Modern vision-language models (VLMs) deliver impressive predictive accuracy yet offer little insight into 'why' a decision is reached, frequently hallucinating facts, particularly when encountering out-of-distribution data. Neurosymbolic frameworks address this by pairing black-box perception with interpretable symbolic reasoning, but current methods extract their symbols solely from task labels, leaving them weakly grounded in the underlying visual data. In this paper, we introduce a multi-agent system - Concept-RuleNet that reinstates visual grounding while retaining transparent reasoning. Specifically, a multimodal concept generator first mines discriminative visual concepts directly from a representative subset of training images. Next, these visual concepts are utilized to condition symbol discovery, anchoring the generations in real image statistics and mitigating label bias. Subsequently, symbols are composed into executable first-order rules by a large language model reasoner agent - yielding interpretable neurosymbolic rules. Finally, during inference, a vision verifier agent quantifies the degree of presence of each symbol and triggers rule execution in tandem with outputs of black-box neural models, predictions with explicit reasoning pathways. Experiments on five benchmarks, including two challenging medical-imaging tasks and three underrepresented natural-image datasets, show that our system augments state-of-the-art neurosymbolic baselines by an average of 5% while also reducing the occurrence of hallucinated symbols in rules by up to 50%.

Large Language Models (LLMs) have shown immense potential in education, automating tasks like quiz generation and content summarization. However, generating effective presentation slides introduces unique challenges due to the complexity of multimodal content creation and the need for precise, domain-specific information. Existing LLM-based solutions often fail to produce reliable and informative outputs, limiting their educational value. To address these limitations, we introduce SlideBot - a modular, multi-agent slide generation framework that integrates LLMs with retrieval, structured planning, and code generation. SlideBot is organized around three pillars: informativeness, ensuring deep and contextually grounded content; reliability, achieved by incorporating external sources through retrieval; and practicality, which enables customization and iterative feedback through instructor collaboration. It incorporates evidence-based instructional design principles from Cognitive Load Theory (CLT) and the Cognitive Theory of Multimedia Learning (CTML), using structured planning to manage intrinsic load and consistent visual macros to reduce extraneous load and enhance dual-channel learning. Within the system, specialized agents collaboratively retrieve information, summarize content, generate figures, and format slides using LaTeX, aligning outputs with instructor preferences through interactive refinement. Evaluations from domain experts and students in AI and biomedical education show that SlideBot consistently enhances conceptual accuracy, clarity, and instructional value. These findings demonstrate SlideBot’s potential to streamline slide preparation while ensuring accuracy, relevance, and adaptability in higher education.

Large vision-language models such as CLIP have shown strong zero-shot classification performance by aligning images and text in a shared embedding space. However, CLIP models often develop multimodal spurious biases, the undesirable tendency to rely on spurious features. For example, CLIP may infer object types in images based on frequently co-occurring backgrounds rather than the object's core features. This bias significantly impairs the robustness of pre-trained CLIP models on out-of-distribution data, where such cross-modal associations no longer hold. Existing methods for mitigating multimodal spurious bias typically require fine-tuning on downstream data or prior knowledge of the bias, which undermines the out-of-the-box usability of CLIP. In this paper, we first theoretically analyze the impact of multimodal spurious bias in zero-shot classification. Based on this insight, we propose Spuriousness-Aware Guided Exploration (SAGE), a simple and effective method that mitigates spurious bias via guided prompt selection. SAGE requires no training, fine-tuning, or external annotations. It explores on a space of prompt templates and selects the prompts that induces the largest semantic separation between classes, thereby improving worst-group robustness. Extensive experiments on four real-world benchmark datasets and five popular backbone models demonstrate that SAGE consistently improves zero-shot performance and generalization, outperforming previous zero-shot approaches without any external knowledge or model updates.

Spurious bias, a tendency to exploit spurious correlations between superficial input attributes and prediction targets, has revealed a severe robustness pitfall in classical machine learning problems. Multimodal Large Language Models (MLLMs), which leverage pretrained vision and language models, have recently demonstrated strong capability in joint vision-language understanding. However, both the presence and severity of spurious biases in MLLMs remain poorly understood. In this work, we address this gap by analyzing the spurious biases in the multimodal setting and uncovering the specific inference-time data patterns that can manifest this problem. To support this analysis, we introduce MM-SpuBench, a comprehensive, human-verified benchmark dataset consisting of image-class pairs annotated with core and spurious attributes, grounded in our taxonomy of nine distinct types of spurious correlations. The benchmark is constructed using human-interpretable attribute information to capture a wide range of spurious patterns reflective of real-world knowledge. Leveraging this benchmark, we conduct a comprehensive evaluation of the state-of-the-art open-source and proprietary MLLMs with both standard accuracy and the proposed Conditional Generation Likelihood Advantage (CGLA). Our findings highlight the persistence of reliance on spurious correlations and the difficulty of mitigation on our benchmark. We hope this work can inspire new technical strides to mitigate these biases. Our benchmark is publicly available at https://huggingface.co/datasets/mmbench/MM-SpuBench.

Abstract Motivation Large language models (LLMs) have demonstrated strong general reasoning abilities, but applying them to domain-specific tasks such as analysing single-cell RNA sequencing data remains a challenge. A central task in this domain is cell type annotation, which is critical for understanding cellular heterogeneity. Although recent foundation models attempt to automate this process, they typically annotate cells independently, without considering batch-level context or providing explanatory reasoning. To address this limitation, we introduce the CellPuzzles benchmark, which reformulates cell type annotation as a batch-level reasoning task. CellPuzzles spans diverse tissues, diseases, and donor conditions, and requires reasoning across the batch-level cellular context to ensure label uniqueness. Results We find that off-the-shelf LLMs struggle on this task, with the best baseline (OpenAI o1) achieving only 19.0% batch-level accuracy. To fill this gap, we propose Cell-o1, a 7B LLM trained via supervised fine-tuning on distilled reasoning traces, followed by reinforcement learning with batch-level rewards. Cell-o1 achieves state-of-the-art performance, outperforming OpenAI o1 by over 73% and generalizing well across contexts. Further analysis of training dynamics and reasoning behaviors provides insights into batch-level annotation performance and emergent expert-like reasoning. Availability and Implementation Code and data are available at https://github.com/ncbi-nlp/cell-o1.

Abstract Research in scientific domains now generates more than a million articles annually, overwhelming researchers and hindering discovery. This surge has sparked interest in biomedical hypothesis generation (HG), which aims to uncover implicit patterns among biomedical concepts. Most existing methods focus on pairwise link prediction, overlooking the complex, multi-concept relationships underlying many breakthroughs. We introduce HyHG , a temporal Hy pergraph contrastive learning framework for biomedical H ypothesis G eneration, which redefines hypotheses as hyperedges—sets of co-mentioned concepts in an article. By representing articles as hyperedges and organizing them into a temporal hypergraph, HyHG captures the evolution of scientific ideas over time. A transformer-based architecture learns from historical hyperedge sequences to predict future hyperedges—sets of concepts likely to co-occur in the future literature. To distinguish genuine hypotheses from misleading ones, HyHG employs a time-anchored contrastive loss and hard negative sampling based on minimal edits to real hyperedges. We demonstrate that HyHG achieves state-of-the-art performance on three biomedical datasets. Our code and data are available at: https://github.com/amir-hassan25/Temporal-Hypergraph-Contrastive-Learning.

Survival analysis models have evolved significantly with deep learning approaches, yet often lack interpretability and meaningful risk stratification capabilities. We present Interpretable Risk Clustering Intelligence for Survival Analysis (IRIS), a novel framework that addresses the critical task of risk clustering while enhancing both input-level and model-body interpretability. Unlike traditional survival models that perform post-hoc risk clustering, IRIS learns to cluster patients into meaningful risk groups directly from data while providing transparent feature importance estimation through feature contribution functions. We validate IRIS on several benchmark datasets, a real-world Alzheimer's disease dataset, and an electronic health record dataset, showing superior performance in risk clustering and predictive reliability with only a modest decrease in time-to-event prediction accuracy compared to state-of-the-art methods. Our results show that IRIS successfully balances the trade-off between interpretability and prediction performance in risk-based survival analysis, offering clinicians actionable insights for treatment planning and resource allocation.

Compositional reasoning remains a persistent weakness of modern vision language models (VLMs): they often falter when a task hinges on understanding how multiple objects, attributes, and relations interact within an image.Multiple research works have attempted to improve compositionality performance by creative tricks such as improving prompt structure, chain of thought reasoning, etc.A more recent line of work attempts to impart additional reasoning in VLMs using well-trained Large Language Models (LLMs), which are far superior in linguistic understanding than VLMs to compensate for the limited linguistic prowess of VLMs.However, these approaches are either resource-intensive or do not provide an interpretable reasoning process.In this paper, we present "COCO-Tree" -a novel approach that augments VLM outputs with carefully designed neurosymbolic concept trees learned from LLMs to improve VLM's linguistic reasoning.COCO-Tree's beam search-inspired reasoning process boosts compositionality performance and provides a rationale behind VLM predictions.Empirical results on four compositionality benchmarks, Winoground, EqBench, ColorSwap, and SugarCrepe, in seven different open-source VLMs with varying sizes, demonstrate that COCO-Tree significantly improves compositional generalization by 5-10% over baselines.The code is available at: https://github.com/sanchit97/ compositionality-low-res-vlm

Concept Activation Vectors (CAVs) provide a powerful approach for interpreting deep neural networks by quantifying their sensitivity to human-defined concepts. However, when computed independently at different layers, CAVs often exhibit inconsistencies, making cross-layer comparisons unreliable. To address this issue, we propose the Global Concept Activation Vector (GCAV), a novel framework that unifies CAVs into a single, semantically consistent representation. Our method leverages contrastive learning to align concept representations across layers and employs an attention-based fusion mechanism to construct a globally integrated CAV. By doing so, our method significantly reduces the variance in TCAV scores while preserving concept relevance, ensuring more stable and reliable concept attributions. To evaluate the effectiveness of GCAV, we introduce Testing with Global Concept Activation Vectors (TGCAV) as a method to apply TCAV to GCAV-based representations. We conduct extensive experiments on multiple deep neural networks, demonstrating that our method effectively mitigates concept inconsistency across layers, enhances concept localization, and improves robustness against adversarial perturbations. By integrating cross-layer information into a coherent framework, our method offers a more comprehensive and interpretable understanding of how deep learning models encode human-defined concepts. Code and models are available at https://github.com/Zhenghao-He/GCAV.

Human intelligence can continuously and adaptively build multimodal cognition from a series of diverse modalities in the external world. Modality-varying Continual Learning (MVCL) aims to imitate such human intelligence, which trains models on a stream of non-stationary and modality-fluctuating data distributions while sequentially transferring and protecting past knowledge. When an MVCL learner cannot anticipate the complexity of future new modalities and inter-modal interactions, the challenge of dealing with knowledge saturation (KS) with satisfactory parameter efficiency (PE) increases. Existing works focused mainly on overcoming the forgetting of past knowledge but overlooked the critical tradeoff between KS and PEC To address this gap, we propose a novel continual learning frame-work, namely LifelongSkill, that explicitly optimizes this tradeoff. Our key idea is to capture the interpretable inter-task diversity underlying the task stream, and then use this information to guide the parameter-efficient knowledge transfer and necessary network expansion. Specifically, we learn a Latent Knowledge Hypergraph (LKGraph), comprising a variety of semantically-distinct functional capabilities (namely skills) learned from tasks, to represent task diversity through skill co-occurrences. Then, we propose a Skill-wise Node Decoder (SND) to facilitate parameter-efficient network expansion and knowledge transfer guided by LKGraph. Experiment results demonstrate the proposed approach achieves the best tradeoffs between performance and parameter efficiency compared with baselines.

Accurate prediction of patient outcomes from electronic health records (EHRs) is a fundamental task in data mining with practical benefits to clinical decision support and healthcare resource allocation. Over the past few years, with the advent of large language models (LLMs), there has been increasing interest in training LLMs on EHR clinical notes to improve outcome predictions. Despite significant advances, existing approaches have a certain limitation. Specifically, the existing approaches largely model clinical notes as flat token sequences and overlook their intrinsic semi-structured organization into sections (e.g., History of Present Illness and Physical Exam). Moreover, most of the existing approaches ignore the issue of missing data prevalent in real-world EHR clinical notes. To address these challenges, we propose a novel approach that leverages the inherent structure of clinical notes to impute missing sections and learns robust feature representations needed for outcome prediction. In particular, we propose a context-aware section imputation strategy that utilizes multi-head attention to infer missing section representations based on inter-section dependencies within the clinical note. Moreover, to learn disentangled feature representations, we propose orthogonality constraints across the section embeddings. Extensive experiments on multiple benchmark datasets for clinical outcome prediction show that the proposed approach achieves consistent improvements over strong baseline algorithms. The code has been released on github at https://github.com/LeiGong0125Carrot/Strucure-Awared-Clinical-Note-Processing/tree/ICDM-2025

Biomedical research now generates more than a million articles annually, overwhelming researchers and hindering discovery. This surge has sparked interest in biomedical hypothesis generation (HG), which aims to uncover implicit patterns among biomedical concepts. Most existing methods focus on pairwise link prediction, overlooking the complex, multi-concept relationships underlying many breakthroughs. We introduce HyHG, a temporal Hypergraph contrastive learning framework for biomedical Hypothesis Generation, which redefines hypotheses as hyperedges–sets of co-mentioned concepts in an article. By representing articles as hyperedges and organizing them into a temporal hypergraph, HyHG captures the evolution of scientific ideas over time. A transformer-based architecture learns from historical hyperedge sequences to predict future hyperedges–sets of concepts likely to co-occur in future literature. To distinguish genuine hypotheses from misleading ones, HyHG employs a timeanchored contrastive loss and hard negative sampling based on minimal edits to real hyperedges. We demonstrate state-of-the-art performance on three biomedical datasets. Our code and data are available at: https://github.com/amirhassan25/Temporal-Hypergraph-Contrastive-Learning.

Deep neural networks often develop spurious bias, reliance on correlations between non-essential features and classes for predictions. For example, a model may identify objects based on frequently co-occurring backgrounds rather than intrinsic features, resulting in degraded performance on data lacking these correlations. Existing mitigation approaches typically depend on external annotations of spurious correlations, which may be difficult to obtain and are not relevant to the spurious bias in a model. In this paper, we take a step towards self-guided mitigation of spurious bias by proposing NeuronTune, a post hoc method that directly intervenes in a model’s internal decision process. Our method probes in a model’s latent embedding space to identify and regulate neurons that lead to spurious prediction behaviors. We theoretically justify our approach and show that it brings the model closer to an unbiased one. Unlike previous methods, NeuronTune operates without requiring spurious correlation annotations, making it a practical and effective tool for improving model robustness. Experiments across different architectures and data modalities demonstrate that our method significantly mitigates spurious bias in a self-guided way.

As Vision Transformers (ViTs) are increasingly adopted in sensitive vision applications, there is a growing demand for improved interpretability. This has led to efforts to forward-align these models with carefully annotated abstract, human-understandable semantic entities - concepts. Concepts provide global rationales to the model predictions and can be quickly understood/intervened on by domain experts. Most current research focuses on designing model-agnostic, plug-and-play generic concept-based explainability modules that do not incorporate the inner workings of foundation models (e.g., inductive biases, scale invariance, etc.) during training. To alleviate this issue for ViTs, in this paper, we propose ASCENT-ViT, an attention-based, concept learning framework that effectively composes scale and position-aware representations from multiscale feature pyramids and ViT patch representations, respectively. Further, these representations are aligned with concept annotations through attention matrices - which incorporate spatial and global (semantic) concepts. ASCENT-ViT can be utilized as a classification head on top of standard ViT backbones for improved predictive performance and accurate and robust concept explanations as demonstrated on five datasets, including three widely used benchmarks (CUB, Pascal APY, Concept-MNIST) and two real-world datasets (AWA2, KITS). An appendix of the paper with more comprehensive results is available at https://arxiv.org/abs/2501.09221.

Large language models (LLMs) have shown significant potential in scientific disciplines such as biomedicine, particularly in hypothesis generation, where they can analyze vast literature, identify patterns, and suggest research directions. However, a key challenge lies in evaluating the truthfulness of generated hypotheses, as verifying their accuracy often requires substantial time and resources. Additionally, the hallucination problem in LLMs can lead to the generation of hypotheses that appear plausible but are ultimately incorrect, undermining their reliability. To facilitate the systematic study of these challenges, we introduce TruthHypo, a benchmark for assessing the capabilities of LLMs in generating truthful scientific hypotheses, and KnowHD, a knowledge-based hallucination detector to evaluate how well hypotheses are grounded in existing knowledge. Our results show that LLMs struggle to generate truthful hypotheses. By analyzing hallucinations in reasoning steps, we demonstrate that the groundedness scores provided by KnowHD serve as an effective metric for filtering truthful hypotheses from the diverse outputs of LLMs. Human evaluations further validate the utility of KnowHD in identifying truthful hypotheses and accelerating scientific discovery. Our data and source code are available at https://github.com/Teddy-XiongGZ/TruthHypo.

Deep learning models often achieve high performance by inadvertently learning spurious correlations between targets and non-essential features. For example, an image classifier may identify an object via its background that spuriously correlates with it. This prediction behavior, known as spurious bias, severely degrades model performance on data that lacks the learned spurious correlations. Existing methods on spurious bias mitigation typically require a variety of data groups with spurious correlation annotations called group labels. However, group labels require costly human annotations and often fail to capture subtle spurious biases such as relying on specific pixels for predictions. In this paper, we propose a novel post hoc spurious bias mitigation framework without requiring group labels. Our framework, termed ShortcutProbe, identifies prediction shortcuts that reflect potential non-robustness in predictions in a given model's latent space. The model is then retrained to be invariant to the identified prediction shortcuts for improved robustness. We theoretically analyze the effectiveness of the framework and empirically demonstrate that it is an efficient and practical tool for improving a model's robustness to spurious bias on diverse datasets.

Artificial Intelligence (AI) is rapidly reshaping the landscape of scientific discovery by enabling the development of novel models that tackle complex, data- and computation-intensive problems. Scientific challenges, in turn, provide rich, use-inspired settings that push the boundaries of AI research. This virtuous cycle is increasingly driven by cross-disciplinary collaboration, where advances in AI and domain sciences co-evolve to accelerate innovation. In this plenary panel, we will examine the opportunities and challenges in designing cutting-edge AI models for scientific discovery, and high- light the transformative potential of cross-disciplinary partnerships in shaping the future of both AI and science.

Large Language Models (LLMs) have revolutionized interactions between human and artificial intelligence (AI) systems, demonstrating state-of-the-art performance across various domains, including scientific discovery and hypothesis generation. However, the absence of a comprehensive and systematic evaluation framework for LLM-driven research idea generation hinders a rigorous understanding of their strengths and limitations. To address this gap, we propose IdeaBench, a benchmark system that provides a structured dataset and evaluation framework for standardizing the assessment of research idea generation by LLMs. Our dataset comprises titles and abstracts from 2,374 influential papers across eight research domains, along with their 29,408 referenced works, creating a context-rich environment that mirrors human researchers' ideation processes. By profiling LLMs as domain-specific researchers and grounding them in similar contextual constraints, we directly leverage the models' knowledge learned from the pre-training stage to generate new research ideas. To systematically evaluate LLMs' research ideation capability and approximate human assessment, we propose a reference-based metric that aligns with human judgment to quantify idea quality with the assistance of LLMs. Through this evaluation, we find that while LLMs excel at generating novel ideas, they may struggle with generating feasible ideas. IdeaBench serves as a critical resource for benchmarking and comparing LLMs, ultimately advancing research on AI's role in automating scientific discovery.

Deep neural networks often learn and rely on spurious correlations, i.e., superficial associations between non-causal features and the targets. For instance, an image classifier may identify camels based on the desert backgrounds. While it can yield high overall accuracy during training, it degrades generalization on more diverse scenarios where such correlations do not hold. This problem poses significant challenges for out-of-distribution robustness and trustworthiness. Existing methods typically mitigate this issue by using external group annotations or auxiliary deterministic models to learn unbiased representations. However, such information is costly to obtain, and deterministic models may fail to capture the full spectrum of biases learned by the models. To address these limitations, we propose Evidential Alignment, a novel framework that leverages uncertainty quantification to understand the behavior of the biased models without requiring group annotations. By quantifying the evidence of model prediction with second-order risk minimization and calibrating the biased models with the proposed evidential calibration technique, Evidential Alignment identifies and suppresses spurious correlations while preserving core features. We theoretically justify the effectiveness of our method as capable of learning the patterns of biased models and debiasing the model without requiring any spurious correlation annotations. Empirical results demonstrate that our method significantly improves group robustness across diverse architectures and data modalities, providing a scalable and principled solution to spurious correlations.

The past decade has been an inspiring time for artificial intelligence (AI) research. AI systems have transformed norms and practices across industries and have permeated the fabric of human society. Moreover, AI is ushering in a transformative technological age by making remarkable breakthroughs in a number of scientific fields such as protein structure prediction and medical imaging. There is increasing consensus in the wider scientific community that AI is poised to disrupt science by unlocking entirely new approaches, driving new scientific inquiry, and enabling greater scientific leaps with far-reaching social consequences. However, there are substantial barriers preventing science from realizing that potential, and addressing these barriers will require support for advances in AI methods and the adoption of these methods in routine scientific research. In this special day at KDD 2025, we host a series of talks by distinguished researchers on AI for science.

In reinforcement learning from human feedback, preference-based reward models play a central role in aligning large language models to human-aligned behavior. However, recent studies show that these models are prone to reward hacking and often fail to generalize well due to over-optimization. They achieve high reward scores by exploiting shortcuts, that is, exploiting spurious features (e.g., response verbosity, agreeable tone, or sycophancy) that correlate with human preference labels in the training data rather than genuinely reflecting the intended objectives. In this paper, instead of probing these issues one at a time, we take a broader view of the reward hacking problem as shortcut behaviors and introduce a principled yet flexible approach to mitigate shortcut behaviors in preference-based reward learning. Inspired by the invariant theory in the kernel perspective, we propose Preference-based Reward Invariance for Shortcut Mitigation (PRISM), which learns group-invariant kernels with feature maps in a closed-form learning objective. Experimental results in several benchmarks show that our method consistently improves the accuracy of the reward model on diverse out-of-distribution tasks and reduces the dependency on shortcuts in downstream policy models, establishing a robust framework for preference-based alignment.

In the emerging landscape of AI-driven scientific discovery, foundation models hold significant promise for enhancing research ideation and overall scientific advancement. This paper explores a future where foundation models should be able to effectively utilize both external and internal knowledge sources to maximize their role in scientific discovery. The core challenge lies in optimizing two knowledge types: external knowledge, drawn from diverse data sources, and internal knowledge, the parametric understanding acquired during training. We propose a dual-framework solution for this optimization, including X-augmented generation and in-context X learning. X-augmented generation approaches, such as retrieval-augmented generation, knowledge graph-augmented generation, and third-party tool integration, enhance external knowledge processing. In-context X learning methods, including in-context adversarial learning and in-context reinforcement learning, improve models’ internal knowledge adaptation and utility for scientific tasks. We aim to inspire the research community by proposing a bold pathway toward leveraging foundation models as active participants in scientific discovery, tackling the inherent complexity of optimizing vast, multimodal knowledge sources. By addressing this challenge, we envision a future where foundation models catalyze breakthroughs across disciplines, ultimately leading to a more dynamic, collaborative, and insight-driven scientific process.

MOTIVATION: Hypothesis generation is a fundamental problem in biomedical text mining that aims to generate ideas that are new, interesting, and plausible by discovering unexplored links between biomedical concepts. Despite significant advances made by existing approaches, they do not fully leverage the evolutionary properties of biomedical concepts. This is limiting because scientific knowledge continually evolves over time, with new facts being added and old ones becoming obsolete. Thus, it is crucial to capture the evolutionary properties of biomedical concepts from multiple perspectives (e.g. spatial, temporal, and semantic) to generate hypotheses that reflect the up-to-date information landscape of the biomedical domain. RESULTS: We introduce a novel framework, ConceptDrift, that models the hypothesis generation task as a sequence of temporal graphlets and simultaneously encodes spatial, temporal, and semantic change. Unlike existing approaches that treat these dimensions independently, ConceptDrift is the first to provide a holistic understanding of concept evolution by integrating them into a unified framework. Grounded in the theories of the Distributional Hypothesis and Conceptual Change, our method adapts these principles to the unique challenges of large-scale biomedical literature. We conduct extensive experiments across multiple datasets and demonstrate that ConceptDrift consistently outperforms state-of-the-art baselines in generating accurate and meaningful hypotheses. Our framework shows immediate practical benefits for web-based literature mining tools in life sciences and biomedicine, offering more robust and predictive feature representations. AVAILABILITY AND IMPLEMENTATION: https://github.com/amir-hassan25/ConceptDrift (DOI: 10.6084/m9.figshare.29975476).

Generalized Additive Models (GAMs) have long been a powerful white-box tool for the intelligible analysis of tabular data, revealing the influence of each feature on the model predictions. Despite the success of Neural Networks (NNs) in various domains, their application as NN-based GAMs in tabular data analysis remains suboptimal compared to tree-based ones, and the opacity of encoders in NN-GAMs also prevents users from understanding how networks learn the functions. In this work, we propose a new deep tabular learning method, termed Prototypical Neural Additive Model (ProtoNAM), which introduces prototypes into NNs in the framework of GAMs. With the introduced prototype-based feature activation, ProtoNAM can flexibly model the irregular mapping from tabular features to the outputs while maintaining the explainability of the final prediction. We also propose a gradient-boosting inspired hierarchical shape function modeling method, facilitating the discovery of complex feature patterns and bringing transparency into the learning process of each network layer. Our empirical evaluations demonstrate that ProtoNAM outperforms all existing NN-based GAMs, while providing additional insights into the shape function learned for each feature. The source code of ProtoNAM is available at https://github.com/Teddy-XiongGZ/ProtoNAM .

There has been growing concern regarding data privacy during the development and deployment of Multimodal Foundation Models for Artificial General Intelligence (AGI), while Federated Learning (FL) allows multiple clients to collaboratively train models in a privacy-preserving manner. This paper formulates and studies Modality-task Agnostic Federated Learning (AFL) to pave the way toward privacy-preserving AGI. A unique property of AFL is the asymmetrical knowledge relationships among clients due to modality gaps, task gaps, and domain shifts between clients. This raises a challenge in learning an optimal inter-client information-sharing scheme that maximizes positive transfer and minimizes negative transfer for AFL. However, prior FL methods, mostly focusing on symmetrical knowledge transfer, tend to exhibit insufficient positive transfer and fail to fully avoid negative transfer during inter-client collaboration. To address this issue, we propose DisentAFL, which leverages a two-stage Knowledge Disentanglement and Gating mechanism to explicitly decompose the original asymmetrical inter-client information-sharing scheme into several independent symmetrical inter-client information-sharing schemes, each of which corresponds to certain semantic knowledge type learned from the local tasks. Experimental results demonstrate the superiority of our method on AFL than baselines.

Federated Averaging (FedAvg) is known to experience convergence issues when encountering significant clients system heterogeneity and data heterogeneity. Server momentum has been proposed as an effective mitigation. However, existing server momentum works are restrictive in the momentum formulation, do not properly schedule hyperparameters and focus only on system homogeneous settings, which leaves the role of server momentum still an under-explored problem. In this paper, we propose a general framework for server momentum, that (a) covers a large class of momentum schemes that are unexplored in federated learning (FL), (b) enables a popular stagewise hyperparameter scheduler, (c) allows heterogeneous and asynchronous local computing. We provide rigorous convergence analysis for the proposed framework. To our best knowledge, this is the first work that thoroughly analyzes the performances of server momentum with a hyperparameter scheduler and system heterogeneity. Extensive experiments validate the effectiveness of our proposed framework. Due to page limit, we leave all proofs to the full version https://arxiv.org/abs/2312.12670.

Single domain generalization (SDG) aims to train a robust model against unknown target domain shifts using data from a single source domain. Data augmentation has been proven an effective approach to SDG. However, the utility of standard augmentations, such as translate, or invert, has not been fully exploited in SDG; practically, these augmentations are used as a part of a data preprocessing procedure. Although it is intuitive to use many such augmentations to boost the robustness of a model to out-of-distribution domain shifts, we lack a principled approach to harvest the benefit brought from multiple these augmentations. Here, we conceptualize standard data augmentations with learnable parameters as semantics transformations that can manipulate certain semantics of a sample, such as the geometry or color of an image. Then, we propose Adversarial learning with Semantics Transformations (AdvST) that augments the source domain data with semantics transformations and learns a robust model with the augmented data. We theoretically show that AdvST essentially optimizes a distributionally robust optimization objective defined on a set of semantics distributions induced by the parameters of semantics transformations. We demonstrate that AdvST can produce samples that expand the coverage on target domain data. Compared with the state-of-the-art methods, AdvST, despite being a simple method, is surprisingly competitive and achieves the best average SDG performance on the Digits, PACS, and DomainNet datasets. Our code is available at https://github.com/gtzheng/AdvST.

Social determinants of health (SDoH) factors such as poverty, social environment, and unemployment are known to profoundly impact health outcomes. However, extracting SDoH from the electronic health records (EHR) is a challenge due to the unstructured nature of clinical narratives that encode them. To address this, several approaches ranging from rule-based natural language processing to large language models have been proposed in the literature. Despite significant advances, the existing SDoH extraction approaches are not robust to the noise present in clinical notes or discharge summaries and thus yield unsatisfactory performance. In other words, the noisy information in clinical notes leads to the generation of low-quality feature representations of medical concepts that severely impacts the performance of SDoH extraction.In this paper, we propose a novel approach that augments EHR discharge summaries with context-specific semantic knowledge from biomedical literature to generate robust feature representations needed for accurate SDoH extraction. Specifically, our approach identifies key contextual information (e.g., symptoms, diseases, and medications) from EHR discharge summaries and retrieves relevant scientific articles to generate additional semantic context for SDoH classifier. Moreover, to effectively fuse complementary information from both EHR discharge summaries and biomedical literature, we propose a new feature infusion strategy that adaptively fuses feature representations based on their contextual relevance. Experimental results on the benchmark MIMIC-SDoH dataset demonstrate that the proposed approach significantly outperforms baseline algorithms and highlight the role of context-specific feature augmentation in enhancing the accuracy of SDoH extraction.

Machine learning foundation models, particularly large language models (LLMs) such as GPT-4o, have revolutionized traditional applications in computer vision and natural language processing, marking a significant shift in recent years. Building on these advancements, recent efforts have explored the potential of foundation models in hypothesis generation, highlighting their possibility in aiding human researchers in scientific discovery. In this paper, we envision a future where academia increasingly integrates foundation models to accelerate and enhance the process of scientific discovery. Motivated by potential application scenarios of foundation models in scientific research, our vision is anchored in a central question: How can we accelerate scientific discovery with the aid of foundation models? To address this overarching question, we raise two key challenges that need to be addressed: (1) how to effectively harness the parametric knowledge embedded in foundation models to propel scientific discovery? and (2) how to develop rigorous yet scalable methods to evaluate the effectiveness of foundation models in supporting scientific research? To tackle these two challenges, we propose our approaches, termed knowledge-grounded Chain-of-Idea (KG-CoI) hypothesis generation and IdeaBench - Benchmarking LLM hypothesis generators in a customizable manner. Through addressing these challenges, we outline our vision in hope to inspire new ideas and innovations in harnessing foundation models for advancing scientific discovery, paving the way for a new era of research collaboration between humans and artificial intelligence.

Multimodal Federated Learning (MFL) addresses the setup of multiple clients with diversified modality types (e.g. image, text, video, and audio) working together to improve their local personal models in a data-privacy manner. Prior MFL works rely on restrictive compositional neural architecture designs to ensure inter-client information sharing via blockwise model aggregation, limiting their applicability in the real-world **Architecture-personalized MFL (AMFL)** scenarios, where clients may have distinguished multimodal interaction strategies and there is no restriction on local architecture design. The key challenge in AMFL is how to automatically and efficiently tackle the two heterogeneity patterns--statistical and architecture heterogeneity--while maximizing the beneficial information sharing among clients. To solve this challenge, we propose **FedMBridge**, which leverages a topology-aware hypernetwork to act as a bridge that can automatically balance and digest the two heterogeneity patterns in a communication-efficient manner. Our experiments on four AMFL simulations demonstrate the efficiency and effectiveness of our proposed approach.

Interpretability of Deep Neural Networks using concept-based models offers a promising way to explain model behavior through human understandable concepts. A parallel line of research focuses on disentangling the data distribution into its underlying generative factors, in turn explaining the data generation process. While both directions have received extensive attention, little work has been done on explaining concepts in terms of generative factors to unify mathematically disentangled representations and human-understandable concepts as an explanation for downstream tasks. In this paper, we propose a novel method CoLiDR - which utilizes a disentangled representation learning setup for learning mutually independent generative factors and subsequently learns to aggregate the said representations into human-understandable concepts using a novel aggregation/decomposition module. Experiments are conducted on datasets with both known and unknown latent generative factors. Our method successfully aggregates disentangled generative factors into concepts while maintaining parity with state-of-the-art concept-based approaches. Quantitative and visual analysis of the learned aggregation procedure demonstrates the advantages of our work compared to commonly used concept-based models over four challenging datasets. Lastly, our work is generalizable to an arbitrary number of concepts and generative factors - making it flexible enough to be suitable for various types of data.

Adapting large language models (LLMs) to unseen tasks with incontext training samples without fine-tuning remains an important research problem. To learn a robust LLM that adapts well to unseen tasks, multiple meta-training approaches have been proposed such as MetaICL and MetaICT, which involve meta-training pre-trained LLMs on a wide variety of diverse tasks. These meta-training approaches essentially perform in-context multi-task fine-tuning and evaluate on a disjointed test set of tasks. Even though they achieve impressive performance, their goal is never to compute a truly general set of parameters. In this paper, we propose MAML-en-LLM, a novel method for meta-training LLMs, which can learn truly generalizable parameters that not only performs well on disjointed tasks but also adapts to unseen tasks. We see an average increase of 2% on unseen domains in the performance while a massive 4% improvement on adaptation performance. Furthermore, we demonstrate that MAML-en-LLM outperforms baselines in settings with limited amount of training data on both seen and unseen domains by an average of 2%. Finally, we discuss the effects of type of tasks, optimizers and task complexity, an avenue barely explored in metatraining literature. Exhaustive experiments across 7 task settings along with two data settings demonstrate that models trained with MAML-en-LLM outperform SOTA meta-training approaches.

Spurious correlations are brittle associations between certain attributes of inputs and target variables, such as the correlation between an image background and an object class. Deep image classifiers often leverage them for predictions, leading to poor generalization on the data where the correlations do not hold. Mitigating the impact of spurious correlations is crucial towards robust model generalization, but it often requires annotations of the spurious correlations in data -- a strong assumption in practice. In this paper, we propose a novel learning framework based on meta-learning, termed SPUME -- SPUriousness-aware MEta-learning, to train an image classifier to be robust to spurious correlations. We design the framework to iteratively detect and mitigate the spurious correlations that the classifier excessively relies on for predictions. To achieve this, we first propose to utilize a pre-trained vision-language model to extract text-format attributes from images. These attributes enable us to curate data with various class-attribute correlations, and we formulate a novel metric to measure the degree of these correlations' spuriousness. Then, to mitigate the reliance on spurious correlations, we propose a meta-learning strategy in which the support (training) sets and query (test) sets in tasks are curated with different spurious correlations that have high degrees of spuriousness. By meta-training the classifier on these spuriousness-aware meta-learning tasks, our classifier can learn to be invariant to the spurious correlations. We demonstrate that our method is robust to spurious correlations without knowing them a priori and achieves the best on five benchmark datasets with different robustness measures. Our code is available at https://github.com/gtzheng/SPUME.

Current benchmarks for evaluating large language models (LLMs) in medicine are primarily focused on question-answering involving domain knowledge and descriptive reasoning. While such qualitative capabilities are vital to medical diagnosis, in real-world scenarios, doctors frequently use clinical calculators that follow quantitative equations and rule-based reasoning paradigms for evidence-based decision support. To this end, we propose MedCalc-Bench, a first-of-its-kind dataset focused on evaluating the medical calculation capability of LLMs. MedCalc-Bench contains an evaluation set of over 1000 manually reviewed instances from 55 different medical calculation tasks. Each instance in MedCalc-Bench consists of a patient note, a question requesting to compute a specific medical value, a ground truth answer, and a step-by-step explanation showing how the answer is obtained. While our evaluation results show the potential of LLMs in this area, none of them are effective enough for clinical settings. Common issues include extracting the incorrect entities, not using the correct equation or rules for a calculation task, or incorrectly performing the arithmetic for the computation. We hope our study highlights the quantitative knowledge and reasoning gaps in LLMs within medical settings, encouraging future improvements of LLMs for various clinical calculation tasks. MedCalc-Bench is publicly available at: https://github.com/ncbi-nlp/MedCalc-Bench.

MOTIVATION: Gene set enrichment (GSE) analysis allows for an interpretation of gene expression through pre-defined gene set databases and is a critical step in understanding different phenotypes. With the rapid development of single-cell RNA sequencing (scRNA-seq) technology, GSE analysis can be performed on fine-grained gene expression data to gain a nuanced understanding of phenotypes of interest. However, with the cellular heterogeneity in single-cell gene profiles, current statistical GSE analysis methods sometimes fail to identify enriched gene sets. Meanwhile, deep learning has gained traction in applications like clustering and trajectory inference in single-cell studies due to its prowess in capturing complex data patterns. However, its use in GSE analysis remains limited, due to interpretability challenges. RESULTS: In this paper, we present DeepGSEA, an explainable deep gene set enrichment analysis approach which leverages the expressiveness of interpretable, prototype-based neural networks to provide an in-depth analysis of GSE. DeepGSEA learns the ability to capture GSE information through our designed classification tasks, and significance tests can be performed on each gene set, enabling the identification of enriched sets. The underlying distribution of a gene set learned by DeepGSEA can be explicitly visualized using the encoded cell and cellular prototype embeddings. We demonstrate the performance of DeepGSEA over commonly used GSE analysis methods by examining their sensitivity and specificity with four simulation studies. In addition, we test our model on three real scRNA-seq datasets and illustrate the interpretability of DeepGSEA by showing how its results can be explained. AVAILABILITY AND IMPLEMENTATION: https://github.com/Teddy-XiongGZ/DeepGSEA.

Rising usage of deep neural networks to perform decision making in critical applications like medical diagnosis and fi- nancial analysis have raised concerns regarding their reliability and trustworthiness. As automated systems become more mainstream, it is important their decisions be transparent, reliable and understandable by humans for better trust and confidence. To this effect, concept-based models such as Concept Bottleneck Models (CBMs) and Self-Explaining Neural Networks (SENN) have been proposed which constrain the latent space of a model to represent high level concepts easily understood by domain experts in the field. Although concept-based models promise a good approach to both increasing explainability and reliability, it is yet to be shown if they demonstrate robustness and output consistent concepts under systematic perturbations to their inputs. To better understand performance of concept-based models on curated malicious samples, in this paper, we aim to study their robustness to adversarial perturbations, which are also known as the imperceptible changes to the input data that are crafted by an attacker to fool a well-learned concept-based model. Specifically, we first propose and analyze different malicious attacks to evaluate the security vulnerability of concept based models. Subsequently, we propose a potential general adversarial training-based defense mechanism to increase robustness of these systems to the proposed malicious attacks. Extensive experiments on one synthetic and two real-world datasets demonstrate the effectiveness of the proposed attacks and the defense approach. An appendix of the paper with more comprehensive results can also be viewed at https://arxiv.org/abs/2211.16080.

Fairness-aware machine learning has attracted a surge of attention in many domains, such as online advertising, personalized recommendation, and social media analysis in web applications. Fairness-aware machine learning aims to eliminate biases of learning models against certain subgroups described by certain protected (sensitive) attributes such as race, gender, and age. Among many existing fairness notions, counterfactual fairness is a popular notion defined from a causal perspective. It measures the fairness of a predictor by comparing the prediction of each individual in the original world and that in the counterfactual worlds in which the value of the sensitive attribute is modified. A prerequisite for existing methods to achieve counterfactual fairness is the prior human knowledge of the causal model for the data. However, in real-world scenarios, the underlying causal model is often unknown, and acquiring such human knowledge could be very difficult. In these scenarios, it is risky to directly trust the causal models obtained from information sources with unknown reliability and even causal discovery methods, as incorrect causal models can consequently bring biases to the predictor and lead to unfair predictions. In this work, we address the problem of counterfactually fair prediction from observational data without given causal models by proposing a novel framework CLAIRE. Specifically, under certain general assumptions, CLAIRE effectively mitigates the biases from the sensitive attribute with a representation learning framework based on counterfactual data augmentation and an invariant penalty. Experiments conducted on both synthetic and real-world datasets validate the superiority of CLAIRE in both counterfactual fairness and prediction performance.

Multimodal spatiotemporal data (MST) consists of multiple simultaneous spatiotemporal modalities that interact with each other in a dynamic manner. Due to the complexity of MST and the recent desire for the explainability of artificial intelligent systems, disentangled representation learning for MST (DisentMST) has become a significant task, which aims to learn disentangled representations that can expose the underlying spatial semantics, temporal dynamic patterns, and inter-modality interaction modes of the complex MST. One limitation of existing approaches is that they might fail to tolerate the real-world incomplete MST data, where missing information might break the cross-modal spatiotemporal dynamics and bring noise and ambiguity to the learning process. Another limitation is that no existing work systematically reveals the structure of different types of disentangled information. To tackle the two limitations, we define a novel two-level hierarchically structured disentanglement task for MST, which reveals informative and structured disentangled representations for MST as well as digests the real-world MST with incompleteness. We propose a new framework, BiDisentMST, which leverages Gaussian Processes and Graph Factorization on the latent space to achieve our purposes. The experimental results demonstrate the effectiveness of our proposed framework compared with baselines with respect to disentanglement and imputation results.

Deep neural networks are susceptible to human imperceptible adversarial perturbations. One of the strongest defense mechanisms is Adversarial Training (AT). In this paper, we aim to address two predominant problems in AT. First, there is still little consensus on how to set hyperparameters with a performance guarantee for AT research, and customized settings impede a fair comparison between different model designs in AT research. Second, the robustly trained neural networks struggle to generalize well and suffer from tremendous overfitting. This paper focuses on the primary AT framework - Projected Gradient Descent Adversarial Training (PGD-AT). We approximate the dynamic of PGD-AT by a continuous-time Stochastic Differential Equation (SDE), and show that the diffusion term of this SDE determines the robust generalization. An immediate implication of this theoretical finding is that robust generalization is positively correlated with the ratio between learning rate and batch size. We further propose a novel approach, Diffusion Enhanced Adversarial Training (DEAT), to manipulate the diffusion term to improve robust generalization with virtually no extra computational burden. We theoretically show that DEAT obtains a tighter generalization bound than PGD-AT. Our empirical investigation is extensive and firmly attests that DEAT universally outperforms PGD-AT by a significant margin.

The unprecedented advance of wireless human sensing is enabled by the proliferation of the deep learning techniques, which, however, rely heavily on the completeness and representativeness of the data patterns contained in the training set. Thus, deep learning based wireless human perception models usually fail when the human subject is conducting activities that are unseen during the model training. To address this problem, we propose a novel wireless signal augmentation framework, named mmGPE, for Generalized mmWave-based Pose Estimation. In mmGPE, we adopt a physical simulator to generate mmWave FMCW signals. However, due to the imperfect simulation of the physical world, there is a big gap between the signals generated by the physical simulator and the real-world signals collected by the mmWave radar. To tackle this challenge, we propose to integrate the physical signal simulation with deep learning techniques. Specifically, we develop a deep learning-based signal refiner in mmGPE that is capable of bridging the gap and generating realistic signal data. Through extensive evaluations on a COTS mmWave testbed, our mmGPE system demonstrates high accuracy in generating human meshes for unseen activities.

Conditional stochastic optimization has found applications in a wide range of machine learning tasks, such as invariant learning, AUPRC maximization, and meta-learning. As the demand for training models with large-scale distributed data grows in these applications, there is an increasing need for communication-efficient distributed optimization algorithms, such as federated learning algorithms. This paper considers the nonconvex conditional stochastic optimization in federated learning and proposes the first federated conditional stochastic optimization algorithm (FCSG) with a conditional stochastic gradient estimator and a momentum-based algorithm (\emph{i.e.}, FCSG-M). To match the lower bound complexity in the single-machine setting, we design an accelerated algorithm (Acc-FCSG-M) via the variance reduction to achieve the best sample and communication complexity. Compared with the existing optimization analysis for Meta-Learning in FL, federated conditional stochastic optimization considers the sample of tasks. Extensive experimental results on various tasks validate the efficiency of these algorithms.

The minimax problems arise throughout machine learning applications, ranging from adversarial training and policy evaluation in reinforcement learning to AUROC maximization. To address the large-scale distributed data challenges across multiple clients with communication-efficient distributed training, federated learning (FL) is gaining popularity. Many optimization algorithms for minimax problems have been developed in the centralized setting (\emph{i.e.}, single-machine). Nonetheless, the algorithm for minimax problems under FL is still underexplored. In this paper, we study a class of federated nonconvex minimax optimization problems. We propose FL algorithms (FedSGDA+ and FedSGDA-M) and reduce existing complexity results for the most common minimax problems. For nonconvex-concave problems, we propose FedSGDA+ and reduce the communication complexity to $O(\varepsilon^{-6})$. Under nonconvex-strongly-concave and nonconvex-PL minimax settings, we prove that FedSGDA-M has the best-known sample complexity of $O(\kappa^{3} N^{-1}\varepsilon^{-3})$ and the best-known communication complexity of $O(\kappa^{2}\varepsilon^{-2})$. FedSGDA-M is the first algorithm to match the best sample complexity $O(\varepsilon^{-3})$ achieved by the single-machine method under the nonconvex-strongly-concave setting. Extensive experimental results on fair classification and AUROC maximization show the efficiency of our algorithms.

Meta-learning has shown great promise in few-shot image classification where only a small amount of labeled data is available in each classification task. Many training tasks are provided to train a meta-model that can quickly learn new and similar concepts with few labeled samples. Data augmentation is often used to augment training tasks to avoid overfitting. However, existing data augmentation methods are often manually designed and fixed during training, ignoring training dynamics and the difference between various meta-learning settings specified by meta-model architectures and meta-learning algorithms. To address this problem, we add a task transformation layer between a training task and a meta-model such that the right amount of perturbation is added to training tasks for a certain meta-learning setting at a certain training stage. By jointly optimizing the task transformation layer and the meta-model, we avoid the risk of providing tasks that are either too easy or too difficult during training. We design the task transformation layer as a stochastic transformation function, adding the flexibility in how a training task can be transformed. We leverage differentiable data augmentations as the building blocks of the task transformation function for efficient optimization. Extensive experiments show that our method can consistently improve the few-shot generalization performance of various meta-models trained with different meta-learning algorithms, meta-model architectures, and datasets.

MOTIVATION: Despite the success of recent machine learning algorithms' applications to survival analysis, their black-box nature hinders interpretability, which is arguably the most important aspect. Similarly, multi-omics data integration for survival analysis is often constrained by the underlying relationships and correlations that are rarely well understood. The goal of this work is to alleviate the interpretability problem in machine learning approaches for survival analysis and also demonstrate how multi-omics data integration improves survival analysis and pathway enrichment. We use meta-learning, a machine-learning algorithm that is trained on a variety of related datasets and allows quick adaptations to new tasks, to perform survival analysis and pathway enrichment on pan-cancer datasets. In recent machine learning research, meta-learning has been effectively used for knowledge transfer among multiple related datasets. RESULTS: We use meta-learning with Cox hazard loss to show that the integration of TCGA pan-cancer data increases the performance of survival analysis. We also apply advanced model interpretability method called DeepLIFT (Deep Learning Important FeaTures) to show different sets of enriched pathways for multi-omics and transcriptomics data. Our results show that multi-omics cancer survival analysis enhances performance compared with using transcriptomics or clinical data alone. Additionally, we show a correlation between variable importance assignment from DeepLIFT and gene coenrichment, suggesting that genes with higher and similar contribution scores are more likely to be enriched together in the same enrichment sets. AVAILABILITY AND IMPLEMENTATION: https://github.com/berkuva/TCGA-omics-integration.

MOTIVATION: The rapid advance in single-cell RNA sequencing (scRNA-seq) technology over the past decade has provided a rich resource of gene expression profiles of single cells measured on patients, facilitating the study of many biological questions at the single-cell level. One intriguing research is to study the single cells which play critical roles in the phenotypes of patients, which has the potential to identify those cells and genes driving the disease phenotypes. To this end, deep learning models are expected to well encode the single-cell information and achieve precise prediction of patients' phenotypes using scRNA-seq data. However, we are facing critical challenges in designing deep learning models for classifying patient samples due to (i) the samples collected in the same dataset contain a variable number of cells-some samples might only have hundreds of cells sequenced while others could have thousands of cells, and (ii) the number of samples available is typically small and the expression profile of each cell is noisy and extremely high-dimensional. Moreover, the black-box nature of existing deep learning models makes it difficult for the researchers to interpret the models and extract useful knowledge from them. RESULTS: We propose a prototype-based and cell-informed model for patient phenotype classification, termed ProtoCell4P, that can alleviate problems of the sample scarcity and the diverse number of cells by leveraging the cell knowledge with representatives of cells (called prototypes), and precisely classify the patients by adaptively incorporating information from different cells. Moreover, this classification process can be explicitly interpreted by identifying the key cells for decision making and by further summarizing the knowledge of cell types to unravel the biological nature of the classification. Our approach is explainable at the single-cell resolution which can identify the key cells in each patient's classification. The experimental results demonstrate that our proposed method can effectively deal with patient classifications using single-cell data and outperforms the existing approaches. Furthermore, our approach is able to uncover the association between cell types and biological classes of interest from a data-driven perspective. AVAILABILITY AND IMPLEMENTATION: https://github.com/Teddy-XiongGZ/ProtoCell4P.

This paper 1 studies how to schedule hyperparameters to improve generalization of both centralized single-machine stochastic gradient descent (SGD) and distributed asynchronous SGD (ASGD). SGD augmented with momentum variants (e.g., heavy ball momentum (SHB) and Nesterov’s accelerated gradient (NAG)) has been the default optimizer for many tasks, in both centralized and distributed environments. However, many advanced momentum variants, despite empirical advantage over classical SHB/NAG, introduce extra hyperparameters to tune. The error-prone tuning is the main barrier for AutoML. Centralized SGD : We first focus on centralized single-machine SGD and show how to efficiently schedule the hyperparameters of a large class of momentum variants to improve generalization. We propose a unified framework called multistage quasi-hyperbolic momentum (Multistage QHM), which covers a large family of momentum variants as its special cases (e.g. vanilla SGD/SHB/NAG). Existing works mainly focus on only scheduling learning rate α ’s decay, while multistage QHM allows additional varying hyperparameters (e.g., momentum factor), and demonstrates better generalization than only tuning α . We show the convergence of multistage QHM for general nonconvex objectives. Distributed SGD : We then extend our theory to distributed asynchronous SGD (ASGD), in which a parameter server distributes data batches to several worker machines and updates parameters via aggregating batch gradients from workers. We quantify the asynchrony between different workers (i.e., gradient staleness), model the dynamics of asynchronous iterations via a stochastic differential equation (SDE), and then derive a PAC-Bayesian generalization bound for ASGD. As a byproduct, we show how a moderately large learning rate helps ASGD to generalize better. Our tuning strategies have rigorous justifications rather than a blind trial-and-error as we theoretically prove why our tuning strategies could decrease our derived generalization errors in both cases. Our strategies simplify the tuning process and beat competitive optimizers in test accuracy empirically. Our codes are publicly available https://github.com/jsycsjh/centralized-asynchronous-tuning.

With the increasing growth of data and the ability of learning with them, machine learning models are adopted in various domains. However, few of machine learning models are able to reason their prediction, which limits their further applications in real-world tasks. With the potential to address this dilemma, model interpretation has become an important research topic because of the ability to provide the underlying reasons for model predictions at the feature level or concept level. Model interpretation at the concept level focuses on exploring the roles of concepts in model prediction, which enables more compact and understandable interpretations. Concept-level model interpretation requires the identification of the concepts that contribute to model prediction and the exploration of the rules underneath these concepts. To achieve the two objectives, we propose a Concept-level Model Interpretation framework (CMIC) from the perspective of causality. CMIC can automatically detect concepts in data and discover the causal relation between the detected concepts and the model's predicted labels. Furthermore, CMIC ranks the contributions of concepts by their causal effect on the model prediction, reflecting the detected concepts’ importance. We evaluate the proposed CMIC framework on both synthetic and real-world datasets to demonstrate the quality of the provided interpretation.

Providing model explanations has gained significant popularity recently. In contrast with the traditional feature-level model explanations, concept-based explanations can provide explanations in the form of high-level human concepts. However, existing concept-based explanation methods implicitly follow a two-step procedure that involves human intervention. Specifically, they first need the human to be involved to define (or extract) the high-level concepts, and then manually compute the importance scores of these identified concepts in a post-hoc way. This laborious process requires significant human effort and resource expenditure due to manual work, which hinders their large-scale deployability. In practice, it is challenging to automatically generate the concept-based explanations without human intervention due to the subjectivity of defining the units of concept-based interpretability. In addition, due to its data-driven nature, the interpretability itself is also potentially susceptible to malicious manipulations. Hence, our goal in this paper is to free human from this tedious process, while ensuring that the generated explanations are provably robust to adversarial perturbations. We propose a novel concept-based interpretation method, which can not only automatically provide the prototype-based concept explanations but also provide certified robustness guarantees for the generated prototype-based explanations. We also conduct extensive experiments on real-world datasets to verify the desirable properties of the proposed method.

In few-shot image classification, it is challenging for deep neural networks to infer the true class of an image when it contains multiple class-unrelated objects and its label carries no semantic meanings, such as Class 1 or Class 2. In contrast, knowing what are not important in a typical classification task, humans can quickly identify the right class objects with very few images. In this paper, we propose to extract semantic features from a given dataset to filter out class-unrelated objects in a few-shot task. Each semantic feature is meta-learned and represents a common object or pattern shared by many tasks. The strengths of these features in a given image are adjusted by an importance kernel encoding the meta-learned knowledge such that class-unrelated objects can be suppressed, and the few-shot classification performance can be improved. To facilitate learning and identifying semantic features in an image, we further propose an image representation decomposition module to decouple complex correlations between objects in an image embedding. The experimental analysis demonstrates the effectiveness of our method, especially in the extremely low-shot cases.

With the advancement of data collection techniques, end users are interested in how different types of data can collaborate to improve our life experiences. Multimodal Federated Learning (MFL) is an emerging area allowing many distributed clients, each of which can collect data from multiple types of sensors, to participate in the training of some multimodal data-related models without sharing their data. In this paper, we address a novel challenging issue in MFL, the modality incongruity, where clients may have heterogeneous setups of sensors and their local data consists of different combinations of modalities. With the modality incongruity, clients may solve different tasks on different parameter spaces, which escalates the difficulties in dealing with the statistical heterogeneity problem of federated learning; also, it would be hard to perform accurate model aggregation across different types of clients. To tackle these challenges, in this work, we propose the FedMSplit framework, which allows federated training over multimodal distributed data without assuming similar active sensors in all clients. The key idea is to employ a dynamic and multi-view graph structure to adaptively capture the correlations amongst multimodal client models. More specifically, we split client models into smaller shareable blocks and allow each type of blocks to provide a specific view on client relationships. With the graph representation, the underlying correlations between clients can be captured as the edge features in the multi-view graph, and then be utilized to promote local model relations through the neighborhood message passing in the graph. Our experimental results demonstrate the effectiveness of our method under different sensor setups with statistical heterogeneity.

This paper studies the convergence and generalization of a large class of Stochastic Gradient Descent (SGD) momentum schemes, in both learning from scratch and transferring representations with fine-tuning. Momentum-based acceleration of SGD is the default optimizer for many deep learning models. However, there is a lack of general convergence guarantees for many existing momentum variants in conjunction withstochastic gradient. It is also unclear how the momentum methods may affect thegeneralization error. In this paper, we give a unified analysis of several popular optimizers, e.g., Polyak's heavy ball momentum and Nesterov's accelerated gradient. Our contribution is threefold. First, we give a unified convergence guarantee for a large class of momentum variants in thestochastic setting. Notably, our results cover both convex and nonconvex objectives. Second, we prove a generalization bound for neural networks trained by momentum variants. We analyze how hyperparameters affect the generalization bound and consequently propose guidelines on how to tune these hyperparameters in various momentum schemes to generalize well. We provide extensive empirical evidence to our proposed guidelines. Third, this study fills the vacancy of a formal analysis of fine-tuning in literature. To our best knowledge, our work is the first systematic generalizability analysis on momentum methods that cover both learning from scratch and fine-tuning. Our codes are available https://github.com/jsycsjh/Demystify-Hyperparameters-for-Stochastic-Optimization-with-Transferable-Representations .

Counterfactual explanations promote explainability in machine learning models by answering the question “how should the input instance be altered to obtain a desired predicted label?". The comparison of this instance before and after perturbation can enhance human interpretation. Most existing studies on counterfactual explanations are limited in tabular data or image data. In this paper, we study the problem of counterfactual explanation generation on graphs. A few studies have explored to generate counterfactual explanations on graphs, but many challenges of this problem are still not well-addressed: 1) optimizing in the discrete and disorganized space of graphs; 2) generalizing on unseen graphs; 3) maintaining the causality in the generated counterfactuals without prior knowledge of the causal model. To tackle these challenges, we propose a novel framework CLEAR which aims to generate counterfactual explanations on graphs for graph-level prediction models. Specifically, CLEAR leverages a graph variational autoencoder based mechanism to facilitate its optimization and generalization, and promotes causality by leveraging an auxiliary variable to better identify the causal model. Extensive experiments on both synthetic and real-world graphs validate the superiority of CLEAR over state-of-the-art counterfactual explanation methods on graphs in different aspects.

The recent proliferation of various wireless sensing systems and applications demonstrates the advantages of radio frequency (RF) signals over traditional camera-based solutions that are faced with various challenges, such as occlusions and poor lighting conditions. Towards the ultimate goal of imaging human body using RF signals, researchers have been exploring the possibility of constructing the human mesh, a structure capturing not only the pose but also the shape of the human body, from RF signals. In this paper, we introduce M4esh, a novel system that utilizes commercial millimeter wave (mmWave) radar for multi-subject 3D human mesh construction. Our M4esh system can detect and track the subjects on a 2D energy map by predicting the subject bounding boxes on the map, and tackle the subjects' mutual occlusion through utilizing the location, velocity and size information of the subjects' bounding boxes from the previous frames as a clue to estimate the bounding box in the current frame. Through extensive experiments on a real-world COTS millimeter-wave testbed, we show that our proposed M4esh system can accurately localize the subjects and generate their human meshes, which demonstrate the superior effectiveness of the proposed M4esh system.

Fair machine learning aims to mitigate the biases of model predictions against certain subpopulations regarding sensitive attributes such as race and gender. Among the many existing fairness notions, counterfactual fairness measures the model fairness from a causal perspective by comparing the predictions of each individual from the original data and the counterfactuals. In counterfactuals, the sensitive attribute values of this individual had been modified. Recently, a few works extend counterfactual fairness to graph data, but most of them neglect the following facts that can lead to biases: 1) the sensitive attributes of each node's neighbors may causally affect the prediction w.r.t. this node; 2) the sensitive attributes may causally affect other features and the graph structure. To tackle these issues, in this paper, we propose a novel fairness notion - graph counterfactual fairness, which considers the biases led by the above facts. To learn node representations towards graph counterfactual fairness, we propose a novel framework based on counterfactual data augmentation. In this framework, we generate counterfactuals corresponding to perturbations on each node's and their neighbors' sensitive attributes. Then we enforce fairness by minimizing the discrepancy between the representations learned from the original graph and the counterfactuals for each node. Experiments on both synthetic and real-world graphs show that our framework outperforms the state-of-the-art baselines in graph counterfactual fairness, and also achieves comparable prediction performance.

Digging informative knowledge and analyzing contents from the internet is a challenging task as web data may contain new concepts that are lack of sufficient labeled data as well as could be multimodal. Few-shot learning (FSL) has attracted significant research attention for dealing with scarcely labeled concepts. However, existing FSL algorithms have assumed a uniform task setting such that all samples in a few-shot task share a common feature space. Yet in the real web applications, it is usually the case that a task may involve multiple input feature spaces due to the heterogeneity of source data, that is, the few labeled samples in a task may be further divided and belong to different feature spaces, namely hybrid few-shot learning (hFSL). The hFSL setting results in a hybrid number of shots per class in each space and aggravates the data scarcity challenge as the number of training samples per class in each space is reduced. To alleviate these challenges, we propose the Task-adaptive Topological Transduction Network, namely TopoNet, which trains a heterogeneous graph-based transductive meta-learner that can combine information from both labeled and unlabeled data to enrich the knowledge about the task-specific data distribution and multi-space relationships. Specifically, we model the underlying data relationships of the few-shot task in a node-heterogeneous multi-relation graph, and then the meta-learner adapts to each task’s multi-space relationships as well as its inter- and intra-class data relationships, through an edge-enhanced heterogeneous graph neural network. Our experiments compared with existing approaches demonstrate the effectiveness of our method.

MOTIVATION: Biomedical language models produce meaningful concept representations that are useful for a variety of biomedical natural language processing (bioNLP) applications such as named entity recognition, relationship extraction and question answering. Recent research trends have shown that the contextualized language models (e.g. BioBERT, BioELMo) possess tremendous representational power and are able to achieve impressive accuracy gains. However, these models are still unable to learn high-quality representations for concepts with low context information (i.e. rare words). Infusing the complementary information from knowledge-bases (KBs) is likely to be helpful when the corpus-specific information is insufficient to learn robust representations. Moreover, as the biomedical domain contains numerous KBs, it is imperative to develop approaches that can integrate the KBs in a continual fashion. RESULTS: We propose a new representation learning approach that progressively fuses the semantic information from multiple KBs into the pretrained biomedical language models. Since most of the KBs in the biomedical domain are expressed as parent-child hierarchies, we choose to model the hierarchical KBs and propose a new knowledge modeling strategy that encodes their topological properties at a granular level. Moreover, the proposed continual learning technique efficiently updates the concepts representations to accommodate the new knowledge while preserving the memory efficiency of contextualized language models. Altogether, the proposed approach generates knowledge-powered embeddings with high fidelity and learning efficiency. Extensive experiments conducted on bioNLP tasks validate the efficacy of the proposed approach and demonstrates its capability in generating robust concept representations.

Metric learning aims at automatically learning a distance metric from data so that the precise similarity between data instances can be faithfully reflected, and its importance has long been recognized in many fields. An implicit assumption in existing metric learning works is that the learned models are performed in a reliable and secure environment. However, the increasingly critical role of metric learning makes it susceptible to a risk of being malicious attacked. To well understand the performance of metric learning models in adversarial environments, in this article, we study the robustness of metric learning to adversarial perturbations, which are also known as the imperceptible changes to the input data that are crafted by an attacker to fool a well-learned model. However, different from traditional classification models, metric learning models take instance pairs rather than individual instances as input, and the perturbation on one instance may not necessarily affect the prediction result for an instance pair, which makes it more difficult to study the robustness of metric learning. To address this challenge, in this article, we first provide a definition of pairwise robustness for metric learning, and then propose a novel projected gradient descent-based attack method (called AckMetric) to evaluate the robustness of metric learning models. To further explore the capability of the attacker to change the prediction results, we also propose a theoretical framework to derive the upper bound of the pairwise adversarial loss. Finally, we incorporate the derived bound into the training process of metric learning and design a novel defense method to make the learned models more robust. Extensive experiments on real-world datasets demonstrate the effectiveness of the proposed methods.

Inferring brain-effective connectivity networks from neuroimaging data has become a very hot topic in neuroinformatics and bioinformatics. In recent years, the search methods based on Bayesian network score have been greatly developed and become an emerging method for inferring effective connectivity. However, the previous score functions ignore the temporal information from functional magnetic resonance imaging (fMRI) series data and may not be able to determine all orientations in some cases. In this article, we propose a novel score function for inferring effective connectivity from fMRI data based on the conditional entropy and transfer entropy (TE) between brain regions. The new score employs the TE to capture the temporal information and can effectively infer connection directions between brain regions. Experimental results on both simulated and real-world data demonstrate the efficacy of our proposed score function.

Most of existing gradient-based meta-learning approaches to few-shot learning assume that all tasks have the same input feature space. However, in the real world scenarios, there are many cases that the input structures of tasks can be different, that is, different tasks may vary in the number of input modalities or data types. Existing meta-learners cannot handle the heterogeneous task distribution (HTD) as there is not only global meta-knowledge shared across tasks but also type-specific knowledge that distinguishes each type of tasks. To deal with task heterogeneity and promote fast within-task adaptions for each type of tasks, in this paper, we propose HetMAML, a task-heterogeneous model-agnostic meta-learning framework, which can capture both the type-specific and globally shared knowledge and can achieve the balance between knowledge customization and generalization. Specifically, we design a multi-channel backbone module that encodes the input of each type of tasks into the same length sequence of modality-specific embeddings. Then, we propose a task-aware iterative feature aggregation network which can automatically take into account the context of task-specific input structures and adaptively project the heterogeneous input spaces to the same lower-dimensional embedding space of concepts. Our experiments on six task-heterogeneous datasets demonstrate that HetMAML successfully leverages type-specific and globally shared meta-parameters for heterogeneous tasks and achieves fast within-task adaptions for each type of tasks.

Inferring causal effect from observational data has attracted much attention from various domains. Under the potential outcome framework, the estimation of counterfactuals is crucial for the investigation of causal effect at the individual level. Existing representation learning approaches focus on learning one balanced feature space, which ignores certain information predictive to the outcomes. To fully utilize the predictive information, we propose a Subspace learning based Counterfactual Inference (SCI) method to estimate causal effect at the individual level. Different from existing work, SCI learns both a common subspace, which preserves the information across all the treatment groups, and treatment-specific subspaces, which retain the information associated with each specific treatment. Learning from two kinds of subspaces helps SCI obtain better causal effect estimations than state-of-the-art methods, demonstrated by a series of experiments on synthetic and real-world datasets.

Accurate traffic prediction is indispensable for intelligent traffic management. The availability of large-scale road sensing data collected by connected wireless sensors and mobile devices have provided unrealized potential for traffic prediction. However, sensory data is often incomplete due to various factors in the process of data acquisition and transmission. The missingness of traffic data brings a key challenge to the traffic prediction task since the state-of-the-art ML-based traffic prediction models (e.g., Graph Convolutional Networks (GCN)) often rely on spatial and temporal completion of the data. Moreover, existing GCN-based methods usually build a static graph based on geographical distances and are limited in their ability to capture the time-evolving relationships amongst road segments. In this paper, we develop a heterogeneous spatio-temporal prediction framework for traffic prediction using incomplete historical data. In the framework, we build multiple graphs to explicitly model the dynamic correlations among road segments from both geographical and historical aspects, and employ recurrent neural networks to capture temporal correlations for each road segment. We impute missing values in a recurrent process, which is seamlessly embedded in the prediction framework so they can be jointly trained. The proposed framework is evaluated on a public dataset of static sensors and a private dataset collected by our roving sensor system. Experimental results show the effectiveness of the proposed framework compared to state-of-the-art methods, and indicate the potential to be deployed into real-world traffic prediction systems.

One fundamental problem in causality learning is to estimate the causal effects of one or multiple treatments (e.g., medicines in the prescription) on an important outcome (e.g., cure of a disease). One major challenge of causal effect estimation is the existence of unobserved confounders -- the unobserved variables that affect both the treatments and the outcome. Recent studies have shown that by modeling how instances are assigned with different treatments together, the patterns of unobserved confounders can be captured through their learned latent representations. However, the interpretability of the representations in these works is limited. In this paper, we focus on the multi-cause effect estimation problem from a new perspective by learning disentangled representations of confounders. The disentangled representations not only facilitate the treatment effect estimation but also strengthen the understanding of causality learning process. Experimental results on both synthetic and real-world datasets show the superiority of our proposed framework from different aspects.

Pre-trained concept representations are essential to many biomedical text mining and natural language processing tasks. As such, various representation learning approaches have been proposed in the literature. More recently, contextualized embedding approaches (i.e., BERT based models) that capture the implicit semantics of concepts at a granular level have significantly outperformed the conventional word embedding approaches (i.e., Word2Vec/GLoVE based models). Despite significant accuracy gains achieved, these approaches are often computationally expensive and memory inefficient. To address this issue, we propose a new representation learning approach that efficiently adapts the concept representations to the newly available data. Specifically, the proposed approach develops a knowledge-guided continual learning strategy wherein the accurate/stable context-information present in human-curated knowledge-bases is exploited to continually identify and retrain the representations of those concepts whose corpus-based context evolved coherently over time. Different from previous studies that mainly leverage the curated knowledge to improve the accuracy of embedding models, the proposed research explores the usefulness of semantic knowledge from the perspective of accelerating the training efficiency of embedding models. Comprehensive experiments under various efficiency constraints demonstrate that the proposed approach significantly improves the computational performance of biomedical word embedding models.

Stochastic gradient descent (SGD) augmented with various momentum variants (e.g. heavy ball momentum (SHB) and Nesterov's accelerated gradient (NAG)) has been the default optimizer for many learning tasks. Tuning the optimizer's hyperparameters is arguably the most time-consuming part of model training. Many new momentum variants, despite their empirical advantage over classical SHB/NAG, introduce even more hyperparameters to tune. Automating the tedious and error-prone tuning is essential for AutoML. This paper focuses on how to efficiently tune a large class of multistage momentum variants to improve generalization. We use the general formulation of quasi-hyperbolic momentum (QHM) and extend "constant and drop'', the widespread learning rate α scheduler where α is set large initially and then dropped every few epochs, to other hyperparameters (e.g. batch size b, momentum parameter β, instant discount factor ν). Multistage QHM is a unified framework which covers a large family of momentum variants as its special cases (e.g. vanilla SGD/SHB/NAG). Existing works mainly focus on scheduling α's decay, while multistage QHM allows additional varying hyperparameters such as b, β, and ν, and demonstrates better generalization ability than only tuning α. Our tuning strategies have rigorous justifications rather than a blind trial-and-error. We theoretically prove why our tuning strategies could improve generalization. We also show the convergence of multistage QHM for general nonconvex objective functions. Our strategies simplify the tuning process and beat competitive optimizers in test accuracy empirically.

In this paper, we present mmMesh, the first real-time 3D human mesh estimation system using commercial portable millimeter-wave devices. mmMesh is built upon a novel deep learning framework that can dynamically locate the moving subject and capture his/her body shape and pose by analyzing the 3D point cloud generated from the mmWave signals that bounce off the human body. The proposed deep learning framework addresses a series of challenges. First, it encodes a 3D human body model, which enables mmMesh to estimate complex and realistic-looking 3D human meshes from sparse point clouds. Second, it can accurately align the 3D points with their corresponding body segments despite the influence of ambient points as well as the error-prone nature and the multi-path effect of the RF signals. Third, the proposed model can infer missing body parts from the information of the previous frames. Our evaluation results on a commercial mmWave sensing testbed show that our mmMesh system can accurately localize the vertices on the human mesh with an average error of 2.47 cm. The superior experimental results demonstrate the effectiveness of our proposed human mesh construction system.

One fundamental problem in causal inference is to learn the individual treatment effects (ITE) -- assessing the causal effects of a certain treatment (e.g., prescription of medicine) on an important outcome (e.g., cure of a disease) for each data instance, but the effectiveness of most existing methods is often limited due to the existence of hidden confounders. Recent studies have shown that the auxiliary relational information among data can be utilized to mitigate the confounding bias. However, these works assume that the observational data and the relations among them are static, while in reality, both of them will continuously evolve over time and we refer such data as time-evolving networked observational data.

MOTIVATION: Genomic region sets summarize functional genomics data and define locations of interest in the genome such as regulatory regions or transcription factor binding sites. The number of publicly available region sets has increased dramatically, leading to challenges in data analysis. RESULTS: We propose a new method to represent genomic region sets as vectors, or embeddings, using an adapted word2vec approach. We compared our approach to two simpler methods based on interval unions or term frequency-inverse document frequency and evaluated the methods in three ways: First, by classifying the cell line, antibody or tissue type of the region set; second, by assessing whether similarity among embeddings can reflect simulated random perturbations of genomic regions; and third, by testing robustness of the proposed representations to different signal thresholds for calling peaks. Our word2vec-based region set embeddings reduce dimensionality from more than a hundred thousand to 100 without significant loss in classification performance. The vector representation could identify cell line, antibody and tissue type with over 90% accuracy. We also found that the vectors could quantitatively summarize simulated random perturbations to region sets and are more robust to subsampling the data derived from different peak calling thresholds. Our evaluations demonstrate that the vectors retain useful biological information in relatively lower-dimensional spaces. We propose that vector representation of region sets is a promising approach for efficient analysis of genomic region data. AVAILABILITY AND IMPLEMENTATION: https://github.com/databio/regionset-embedding. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

MOTIVATION: Many real-world biomedical interactions such as 'gene-disease', 'disease-symptom' and 'drug-target' are modeled as a bipartite network structure. Learning meaningful representations for such networks is a fundamental problem in the research area of Network Representation Learning (NRL). NRL approaches aim to translate the network structure into low-dimensional vector representations that are useful to a variety of biomedical applications. Despite significant advances, the existing approaches still have certain limitations. First, a majority of these approaches do not model the unique topological properties of bipartite networks. Consequently, their straightforward application to the bipartite graphs yields unsatisfactory results. Second, the existing approaches typically learn representations from static networks. This is limiting for the biomedical bipartite networks that evolve at a rapid pace, and thus necessitate the development of approaches that can update the representations in an online fashion. RESULTS: In this research, we propose a novel representation learning approach that accurately preserves the intricate bipartite structure, and efficiently updates the node representations. Specifically, we design a customized autoencoder that captures the proximity relationship between nodes participating in the bipartite bicliques (2 × 2 sub-graph), while preserving both the global and local structures. Moreover, the proposed structure-preserving technique is carefully interleaved with the central tenets of continual machine learning to design an incremental learning strategy that updates the node representations in an online manner. Taken together, the proposed approach produces meaningful representations with high fidelity and computational efficiency. Extensive experiments conducted on several biomedical bipartite networks validate the effectiveness and rationality of the proposed approach.

Reports on the leaving of the Editor-in-Chief and his replacement.

Indexing biomedical research articles with Medical Subject Headings (MeSH) can greatly facilitate biomedical research and information retrieval. Currently MeSH indexing is performed by human experts. To alleviate the time consumption and monetary cost caused by manual indexing, many automatic MeSH indexing models have been developed, such as MeSHProbeNet, DeepMeSH, and NLM’s official model Medical Text Indexer. In this article, we propose an end-to-end framework, MeSHProbeNet-P, which extends MeSHProbeNet with personalizable MeSH probes. In MeSHProbeNet-P, each MeSH probe carries certain aspects of biomedical knowledge and extracts related information from input articles. MeSHProbeNet-P is able to automatically personalize its MeSH probes for different input articles to ensure that the current MeSH probes best fit the current input article and the most informative features can be extracted from the article. We demonstrate the effectiveness of MeSHProbeNet-P in a real-world large-scale MeSH indexing challenge. MeSHProbeNet-P won the first place in the first batch of Task A in the 2019 BioASQ challenge. The result on the first test set of the challenge is reported in this article. We also provide ablation studies to show the advantages of personalizable MeSH probes.

Causal inference is a critical research topic across many domains, such as statistics, computer science, education, public policy, and economics, for decades. Nowadays, estimating causal effect from observational data has become an appealing research direction owing to the large amount of available data and low budget requirement, compared with randomized controlled trials. Embraced with the rapidly developed machine learning area, various causal effect estimation methods for observational data have sprung up. In this survey, we provide a comprehensive review of causal inference methods under the potential outcome framework, one of the well-known causal inference frameworks. The methods are divided into two categories depending on whether they require all three assumptions of the potential outcome framework or not. For each category, both the traditional statistical methods and the recent machine learning enhanced methods are discussed and compared. The plausible applications of these methods are also presented, including the applications in advertising, recommendation, medicine, and so on. Moreover, the commonly used benchmark datasets as well as the open-source codes are also summarized, which facilitate researchers and practitioners to explore, evaluate and apply the causal inference methods.

The dynamic network data have become ubiquitous with the rapid development of Internet and smart devices. To effectively manage the involved vertices in networks, it is crucial to track the special community patterns and analyze the relationships among vertices. In this paper, we propose a new method to measure the coherence strength, also referred to as community consistency, of a community over a specific observation period. The measurement of community consistency is especially challenging given the dynamic community structure over time, i.e., vertices can leave their original communities and join new communities. In order to interpret the causes of evolving community structure and model the influence of evolving community structure on community consistency, we introduce an influence propagation process having a causal relation with the community consistency. Specifically, a generative model is proposed to combine the influence propagation and the network topological structure at each time step. The proposed influence-based approach for modeling evolution can be instantiated in a variety of real-world network data. The comprehensive experiments on both synthetic and real-world datasets demonstrate the superiority of the proposed framework in estimating the community consistency. Besides, we conduct a case study to show the effectiveness of the proposed method in real-world applications.

Pairwise learning has received much attention recently as it is more capable of modeling the relative relationship between pairs of samples. Many machine learning tasks can be categorized as pairwise learning, such as AUC maximization and metric learning. Existing techniques for pairwise learning all fail to take into consideration a critical issue in their design, i.e., the protection of sensitive information in the training set. Models learned by such algorithms can implicitly memorize the details of sensitive information, which offers opportunity for malicious parties to infer it from the learned models. To address this challenging issue, in this paper, we propose several differentially private pairwise learning algorithms for both online and offline settings. Specifically, for the online setting, we first introduce a differentially private algorithm (called OnPairStrC) for strongly convex loss functions. Then, we extend this algorithm to general convex loss functions and give another differentially private algorithm (called OnPairC). For the offline setting, we also present two differentially private algorithms (called OffPairStrC and OffPairC) for strongly and general convex loss functions, respectively. These proposed algorithms can not only learn the model effectively from the data but also provide strong privacy protection guarantee for sensitive information in the training set. Extensive experiments on real-world datasets are conducted to evaluate the proposed algorithms and the experimental results support our theoretical analysis.

Recently, there are increasingly more attentions paid to an important family of learning problems called pairwise learning, in which the associated loss functions depend on pairs of instances. Despite the tremendous success of pairwise learning in many real-world applications, the lack of transparency behind the learned pairwise models makes it difficult for users to understand how particular decisions are made by these models, which further impedes users from trusting the predicted results. To tackle this problem, in this paper, we study feature importance scoring as a specific approach to the problem of interpreting the predictions of black-box pairwise models. Specifically, we first propose a novel adaptive Shapley-value-based interpretation method, based on which a vector of importance scores associated with the underlying features of a testing instance pair can be adaptively calculated with the consideration of feature correlations, and these scores can be used to indicate which features make key contributions to the final prediction. Considering that Shapley-value-based methods are usually computationally challenging, we further propose a novel robust approximation interpretation method for pairwise models. This method is not only much more efficient but also robust to data noise. To the best of our knowledge, we are the first to investigate how to enable interpretation in pairwise learning. Theoretical analysis and extensive experiments demonstrate the effectiveness of the proposed methods.

Inferring effective connectivity between different brain regions from functional magnetic resonance imaging (fMRI) data is an important advanced study in neuroinformatics in recent years. However, current methods have limited usage in effective connectivity studies due to the high noise and small sample size of fMRI data. In this paper, we propose a novel framework for inferring effective connectivity based on generative adversarial networks (GAN), named as EC-GAN. The proposed framework EC-GAN infers effective connectivity via an adversarial process, in which we simultaneously train two models: a generator and a discriminator. The generator consists of a set of effective connectivity generators based on structural equation models which can generate the fMRI time series of each brain region via effective connectivity. Meanwhile, the discriminator is employed to distinguish between the joint distributions of the real and generated fMRI time series. Experimental results on simulated data show that EC-GAN can better infer effective connectivity compared to other state-of-the-art methods. The real-world experiments indicate that EC-GAN can provide a new and reliable perspective analyzing the effective connectivity of fMRI data.

Cancer has been characterized as a heterogeneous disease, and the classification of cancer subtypes has become a necessity in cancer research, as it can facilitate the subsequent clinical management of patients and provide clinical decision support for clinicians. With the advance of machine learning in the last decade, many researchers employ machine learning to tackle the cancer classification problem. Importantly, traditional machine learning algorithms require a large amount of annotated data for model training. However, collection of large amounts of annotated data is time-consuming and expensive and may not be realistic in real-world activities. Facing data scarcity, meta-learning is proposed to tackle this problem. Meta-learning utilizes prior knowledge learned from related tasks and generalizes to new tasks of limited supervised experience, and it has been applied in many fields to tackle scarce annotated data problem, such as few-shot image classification, drug discovery, etc. As data scarcity is common in cancer research and diagnosis studies, and there are only few previous studies that classify cancers based on limited annotated data. We explore the meta-learning algorithm (MAML) to tackle the scenario where only limited annotated data are available. In this work, our objective is to comprehensively compare MAML among few-shot learning methods (matching network and prototypical network) and traditional machine learning methods (random forest and K-nearest neighbor). Experimental results on The Cancer Genome Atlas (TCGA) cancer patient data demonstrates the effectiveness and superiority of MAML over other methods, including its ability to outperform the other methods using 4.5-fold fewer features.

As an important family of learning problems in healthcare domain, patient similarity learning has received much attention in recent years. Patient similarity learning aims to measure the similarity between a pair of patients according to their historical clinical information, which helps to improve the clinical predictions of the patient of interest. Although patient similarity learning has achieved tremendous success in many real-world applications, the lack of transparency behind the behavior of the learned patient similarity model impedes users from trusting the predicted results, which hampers its further applications in the real world. To tackle this problem, in this paper, we investigate how to enable interpretation in patient similarity learning and propose a global interpretation method for patient similarity learning. Based on the proposed global interpretation method, we can identify a minimal sufficient subset of data features that are sufficient in themselves to justify the global predictions made by the well-trained patient similarity model. The identified minimal sufficient feature subset can help us to better understand the overall behaviors of the learned model across different subpopulations of patients. We also conduct experiments on real-world datasets to evaluate the performance of the proposed global interpretation method.

Missing data, which commonly appears in multivariate time series, has been widely recognized as a key challenge in time series analysis. Many commonly used imputation methods either ignore the temporal dependencies of time series data, or do not adequately utilize the relationships among variables. State-ofthe-art methods on time series imputation are built on Recurrent Neural Networks (RNNs), which utilize the historical information to estimate current values sequentially. However, RNNs rely heavily on the output of nearby timestamps, which may lead to important information lost for long sequences. Moreover, individual variables typically present different dynamics and missingness patterns, which is neglected by the global RNN hidden states. In this paper, we propose an imputation framework to learn both global and local dependencies of multivariate time series, as well as a multi-dimensional self-attention to learn capture distant correlations across both time and feature. Extensive experiments show that the proposed framework outperforms the state-of-the-art methods in the imputation task, and benefits the downstream task.

With the advances in data collection techniques, large amounts of multimodal data collected from multiple sources are becoming available. Such multimodal data can provide complementary information that can reveal fundamental characteristics of real-world subjects. Thus, multimodal machine learning has become an active research area. Extensive works have been developed to exploit multimodal interactions and integrate multi-source information. However, multimodal data in the real world usually comes with missing modalities due to various reasons, such as sensor damage, data corruption, and human mistakes in recording. Effectively integrating and analyzing multimodal data with incompleteness remains a challenging problem. We propose a Heterogeneous Graph-based Multimodal Fusion (HGMF) approach to enable multimodal fusion of incomplete data within a heterogeneous graph structure. The proposed approach develops a unique strategy for learning on incomplete multimodal data without data deletion or data imputation. More specifically, we construct a heterogeneous hypernode graph to model the multimodal data having different combinations of missing modalities, and then we formulate a graph neural network based transductive learning framework to project the heterogeneous incomplete data onto a unified embedding space, and multi-modalities are fused along the way. The learning framework captures modality interactions from available data, and leverages the relationships between different incompleteness patterns. Our experimental results demonstrate that the proposed method outperforms existing graph-based as well as non-graph based baselines on three different datasets.

The past years have witnessed the rapid development of deep reinforcement learning (DRL), which is a combination of deep learning and reinforcement learning (RL). However, the adoption of deep neural networks makes the decision-making process of DRL opaque and lacking transparency. Motivated by this, various interpretation methods for DRL have been proposed. However, those interpretation methods make an implicit assumption that they are performed in a reliable and secure environment. In practice, sequential agent-environment interactions expose the DRL algorithms and their corresponding downstream interpretations to extra adversarial risk. In spite of the prevalence of malicious attacks, there is no existing work studying the possibility and feasibility of malicious attacks against DRL interpretations. To bridge this gap, in this paper, we investigate the vulnerability of DRL interpretation methods. Specifically, we introduce the first study of the adversarial attacks against DRL interpretations, and propose an optimization framework based on which the optimal adversarial attack strategy can be derived. In addition, we study the vulnerability of DRL interpretation methods to the model poisoning attacks, and present an algorithmic framework to rigorously formulate the proposed model poisoning attack. Finally, we conduct both theoretical analysis and extensive experiments to validate the effectiveness of the proposed malicious attacks against DRL interpretations.

Annotated data samples in real-world applications are often limited. Meta-learning, which utilizes prior knowledge learned from related tasks and generalizes to new tasks of limited supervised experience, is an effective approach for few-shot learning. However, standard meta-learning with globally shared knowledge cannot handle the task heterogeneity problem well, i.e., tasks lie in different distributions. Recent advances have explored several ways to trigger task-dependent initial parameters or metrics, in order to customize task-specific information. These approaches learn task contextual information from data, but ignore external domain knowledge that can help in the learning process. In this paper, we propose a task-adaptive network (TAdaNet) that makes use of a domain-knowledge graph to enrich data representations and provide task-specific customization. Specifically, we learn a task embedding that characterizes task relationships and tailors task-specific parameters, resulting in a task-adaptive metric space for classification. Experimental results on a few-shot image classification problem show the effectiveness of the proposed method. We also apply it on a real-world disease classification problem, and show promising results for clinical decision support.

This paper develops the Correlation Networks (CorNet) architecture for the extreme multi-label text classification (XMTC) task, where the objective is to tag an input text sequence with the most relevant subset of labels from an extremely large label set. XMTC can be found in many real-world applications, such as document tagging and product annotation. Recently, deep learning models have achieved outstanding performances in XMTC tasks. However, these deep XMTC models ignore the useful correlation information among different labels. CorNet addresses this limitation by adding an extra CorNet module at the prediction layer of a deep model, which is able to learn label correlations, enhance raw label predictions with correlation knowledge and output augmented label predictions. We show that CorNet can be easily integrated with deep XMTC models and generalize effectively across different datasets. We further demonstrate that CorNet can bring significant improvements over the existing deep XMTC models in terms of both performance and convergence rate. The models and datasets are available at: https://github.com/XunGuangxu/CorNet.

Debugging a traditional network is notoriously difficult due to network devices' heterogeneity and protocols' decentralized nature, but Software-Defined Networking (SDN) is changing this predicament. Recent works have provided very nice approaches for an administrator to perform several fundamental network tests in a single-domain SDN network. However, how to perform these tests securely in multi-domain networks still remains open. In this paper, we study the highly challenging problem of inter-domain forwarding loop test in a SDN environment. We present two novel testing protocols that can be used for inter-domain loop tests. Both protocols are secure in the sense that they protect each domain's private information about its topology and configuration. The first protocol, based on random sampling, is highly efficient with a small error probability diminishing exponentially in the sample size. The second protocol, based on secure set intersection test, guarantees 100 percent accuracy of the result, although not as efficient as the first one. We provide rigorous proofs for the security and accuracy guarantees, and show our protocols have very good efficiency by testing them with real-world network data.

Learning brain effective connectivity (EC) networks from functional magnetic resonance imaging (fMRI) data has become a new hot topic in the neuroinformatics field. However, how to accurately and efficiently learn brain EC networks is still a challenging problem. In this paper, we propose a new algorithm to learn the brain EC network structure using ant colony optimization (ACO) algorithm combining with voxel activation information, named as VACOEC. First, VACOEC uses the voxel activation information to measure the independence between each pair of brain regions and effectively restricts the space of candidate solutions, which makes many unnecessary searches of ants be avoided. Then, by combining the global score increase of a solution with the voxel activation information, a new heuristic function is designed to guide the process of ACO to search for the optimal solution. The experimental results on simulated datasets show that the proposed method can accurately and efficiently identify the directions of the brain EC networks. Moreover, the experimental results on real-world data show that patients with Alzheimers disease (AD) exhibit decreased effective connectivity not only in the intra-network within the default mode network (DMN) and salience network (SN), but also in the inter-network between DMN and SN, compared with normal control (NC) subjects. The experimental results demonstrate that VACOEC is promising for practical applications in the neuroimaging studies of geriatric subjects and neurological patients.

The goal of metric learning is to learn a good distance metric that can capture the relationships among instances, and its importance has long been recognized in many fields. An implicit assumption in the traditional settings of metric learning is that the associated labels of the instances are deterministic. However, in many real-world applications, the associated labels come naturally with probabilities instead of deterministic values, which makes it difficult for the existing metric-learning methods to work well in these applications. To address this challenge, in this article, we study how to effectively learn the distance metric from datasets that contain probabilistic information, and then propose several novel metric-learning mechanisms for two types of probabilistic labels, i.e., the instance-wise probabilistic label and the group-wise probabilistic label. Compared with the existing metric-learning methods, our proposed mechanisms are capable of learning distance metrics directly from the probabilistic labels with high accuracy. We also theoretically analyze the proposed mechanisms and conduct extensive experiments on real-world datasets to verify the desirable properties of these mechanisms.

While deep learning has achieved great success in computer vision and many other fields, currently it does not work very well on patient genomic data with the “big p, small N” problem (i.e., a relatively small number of samples with highdimensional features). In order to make deep learning work with a small amount of training data, we have to design new models that facilitate few-shot learning. Here we present the Affinity Network Model (AffinityNet), a data efficient deep learning model that can learn from a limited number of training examples and generalize well. The backbone of the AffinityNet model consists of stacked k-Nearest-Neighbor (kNN) attention pooling layers. The kNN attention pooling layer is a generalization of the Graph Attention Model (GAM), and can be applied to not only graphs but also any set of objects regardless of whether a graph is given or not. As a new deep learning module, kNN attention pooling layers can be plugged into any neural network model just like convolutional layers. As a simple special case of kNN attention pooling layer, feature attention layer can directly select important features that are useful for classification tasks. Experiments on both synthetic data and cancer genomic data from TCGA projects show that our AffinityNet model has better generalization power than conventional neural network models with little training data.

Estimating brain effective connectivity (EC) from neuroimaging data has recently received wide interest and become a new topic in the neuroinformatics field. Currently, dynamic Bayesian networks (DBN) have been successfully applied to estimating EC from functional magnetic resonance imaging (fMRI) time-series data as they can capture the temporal characteristics of connectivity among brain regions. However, DBN methods assume that activations of brain regions are stationary and follow a Markovian condition, which are strong assumptions that may not be valid in many cases. In this paper, we introduce a novel method to estimate brain effective connectivity networks from fMRI data using non-stationary dynamic Bayesian networks, named as EC-nsDBN. EC-nsDBN can not only capture the non-stationary temporal information from fMRI time-series data but also estimate how interactions among brain regions change dynamically over the fMRI experiments. Systematic experiments on simulated data show that EC-nsDBN has better direction identification ability compared with other state-of-the-art algorithms, and can accurately capture the temporal characteristics of connectivity. Experiments on real-world data sets are also provided to support our analysis.

Treatment effect estimation refers to the estimation of causal effects, which benefits decision-making process across various domains, but it is a challenging problem in real practice. The estimation of causal effects from observational data at the individual level faces two major challenges, i.e., treatment selection bias and missing counterfactuals. Existing methods tackle the selection bias problem by learning a balanced representation and infer the missing counterfactuals based on the learned representation. However, most existing methods learn the representation in a global manner and ignore the local similarity information, which is essential for an accurate estimation of causal effects. Motivated by the above observations, we propose a novel representation learning method, which adaptively extracts fine-grained similarity information from the original feature space and minimizes the distance between different treatment groups as well as the similarity loss during the representation learning procedure. Experiments on three public datasets demonstrate that the proposed method achieves the best performance in causal effect estimation among all the compared methods and is robust to the treatment selection bias.

Although releasing crowdsourced data brings many benefits to the data analyzers to conduct statistical analysis, it may violate crowd users' data privacy. A potential way to address this problem is to employ traditional differential privacy (DP) mechanisms and perturb the data with some noise before releasing them. However, considering that there usually exist conflicts among the crowdsourced data and these data are usually large in volume, directly using these mechanisms can not guarantee good utility in the setting of releasing crowdsourced data. To address this challenge, in this paper, we propose a novel privacy-aware synthesizing method (i.e., PrisCrowd) for crowdsourced data, based on which the data collector can release users' data with strong privacy protection for their private information, while at the same time, the data analyzer can achieve good utility from the released data. Both theoretical analysis and extensive experiments on real-world datasets demonstrate the desired performance of the proposed method.

As an effective way to learn a distance metric between pairs of samples, deep metric learning (DML) has drawn significant attention in recent years. The key idea of DML is to learn a set of hierarchical nonlinear mappings using deep neural networks, and then project the data samples into a new feature space for comparing or matching. Although DML has achieved practical success in many applications, there is no existing work that theoretically analyzes the generalization error bound for DML, which can measure how good a learned DML model is able to perform on unseen data. In this paper, we try to fill up this research gap and derive the generalization error bound for DML. Additionally, based on the derived generalization bound, we propose a novel DML method (called ADroDML), which can adaptively learn the retention rates for the DML models with dropout in a theoretically justified way. Compared with existing DML works that require predefined retention rates, ADroDML can learn the retention rates in an optimal way and achieve better performance. We also conduct experiments on real-world datasets to verify the findings derived from the generalization error bound and demonstrate the effectiveness of the proposed adaptive DML method.

Utilizing multiple modalities to learn a good distance metric is of vital importance for various clinical applications. However, it is common that modalities are incomplete for some patients due to various technical and practical reasons in healthcare datasets. Existing metric learning methods cannot directly learn the distance metric on such data with missing modalities. Nevertheless, the incomplete data contains valuable information to characterize patient similarity and modality relationships, and they should not be ignored during the learning process. To tackle the aforementioned challenges, we propose a metric learning framework to perform missing modality completion and multi-modal metric learning simultaneously. Employing the generative adversarial networks, we incorporate both complete and incomplete data to learn the mapping relationship between modalities. After completing the missing modalities, we use the nonlinear representations extracted by the discriminator to learn the distance metric among patients. Through jointly training the adversarial generation part and metric learning, the similarity among patients can be learned on data with missing modalities. Experimental results show that the proposed framework learns more accurate distance metric on real-world healthcare datasets with incomplete modalities, comparing with the state-of-the-art approaches. Meanwhile, the quality of the generated modalities can be preserved.

Estimating the treatment effect benefits decision making in various domains as it can provide the potential outcomes of different choices. Existing work mainly focuses on covariates with numerical values, while how to handle covariates with textual information for treatment effect estimation is still an open question. One major challenge is how to filter out the nearly instrumental variables which are the variables more predictive to the treatment than the outcome. Conditioning on those variables to estimate the treatment effect would amplify the estimation bias. To address this challenge, we propose a conditional treatment-adversarial learning based matching method (CTAM). CTAM incorporates the treatment-adversarial learning to filter out the information related to nearly instrumental variables when learning the representations, and then it performs matching among the learned representations to estimate the treatment effects. The conditional treatment-adversarial learning helps reduce the bias of treatment effect estimation, which is demonstrated by our experimental results on both semi-synthetic and real-world datasets.

Hypothesis generation (HG) refers to the task of mining meaningful implicit association between unlinked biomedical concepts. The majority of prior studies have focused on uncovering these implicit linkages from static snapshots of the corpus, thereby largely ignoring the temporal dynamics of medical concepts. More recently, a few initial studies attempted to overcome this issue by modelling the temporal change of concepts from natural language text. However, they still fail to leverage the evolutionary features of concepts from contemporary knowledge-bases (KB's) such as semantic lexicons and ontologies. In practice such KB's contain up-to-date information that is important to incorporate, especially, in highly evolving domains such as biomedicine. Furthermore, considering the complementary strength of these sources of information - corpus and ontology - a few natural questions arise: Can joint modelling of (co)-evolutionary dynamics from these resources aid in encoding the temporal features at a granular level? Can the mutual evolution between these intertwined resources lead to better predictive effects? To answer these questions, in this study, we present a novel HG framework that unearths the latent associations between concepts by modeling their co-evolution across complementary sources of information. More specifically, the proposed approach adopts a shared temporal matrix factorization framework that models the co-evolution of concepts across both corpus and KB. Extensive experiments on the largest available biomedical corpus validates the effectiveness of the proposed approach.

In recent years, significant research efforts have been spent towards building intelligent and user-friendly IoT systems to enable a new generation of applications capable of performing complex sensing and recognition tasks. In many of such applications, there are usually multiple different sensors monitoring the same object. Each of these sensors can be regarded as an information source and provides us a unique "view" of the observed object. Intuitively, if we can combine the complementary information carried by multiple sensors, we will be able to improve the sensing performance. Towards this end, we propose DeepFusion, a unified multi-sensor deep learning framework, to learn informative representations of heterogeneous sensory data. DeepFusion can combine different sensors' information weighted by the quality of their data and incorporate cross-sensor correlations, and thus can benefit a wide spectrum of IoT applications. To evaluate the proposed DeepFusion model, we set up two real-world human activity recognition testbeds using commercialized wearable and wireless sensing devices. Experiment results show that DeepFusion can outperform the state-of-the-art human activity recognition methods.

In many time-sensitive applications, knowing the classification results as early as possible while preserving the accuracy is extremely important for further actions. Shapelet-based early classification methods are popular due to their natural interpretability. However, most of the existing shapelet-based methods ignore the distance information between the shapelets and the time series. The distance information, though may contain some noise, can reflect more information between the shapelets and the time series. Some existing works adopt the distance information, but are not robust to the noise in the distance information. To tackle this challenge, we present a novel distance transformation based early classification (DTEC) framework, which transfers the original time series into the distance space. Upon the distance space, a probabilistic classifier is trained, and a novel classification criterion confidence area is proposed in order to overcome the noise brought by the training phase and the dataset. The effectiveness of the proposed framework is validated on three time series benchmarks as well as the extensive datasets selected from UCR time series archive.

MOTIVATION: MEDLINE is the primary bibliographic database maintained by National Library of Medicine (NLM). MEDLINE citations are indexed with Medical Subject Headings (MeSH), which is a controlled vocabulary curated by the NLM experts. This greatly facilitates the applications of biomedical research and knowledge discovery. Currently, MeSH indexing is manually performed by human experts. To reduce the time and monetary cost associated with manual annotation, many automatic MeSH indexing systems have been proposed to assist manual annotation, including DeepMeSH and NLM's official model Medical Text Indexer (MTI). However, the existing models usually rely on the intermediate results of other models and suffer from efficiency issues. We propose an end-to-end framework, MeSHProbeNet (formerly named as xgx), which utilizes deep learning and self-attentive MeSH probes to index MeSH terms. Each MeSH probe enables the model to extract one specific aspect of biomedical knowledge from an input article, thus comprehensive biomedical information can be extracted with different MeSH probes and interpretability can be achieved at word level. MeSH terms are finally recommended with a unified classifier, making MeSHProbeNet both time efficient and space efficient. RESULTS: MeSHProbeNet won the first place in the latest batch of Task A in the 2018 BioASQ challenge. The result on the last test set of the challenge is reported in this paper. Compared with other state-of-the-art models, such as MTI and DeepMeSH, MeSHProbeNet achieves the highest scores in all the F-measures, including Example Based F-Measure, Macro F-Measure, Micro F-Measure, Hierarchical F-Measure and Lowest Common Ancestor F-measure. We also intuitively show how MeSHProbeNet is able to extract comprehensive biomedical knowledge from an input article.

BACKGROUND: Sleep is a complex and dynamic biological process characterized by different sleep patterns. Comprehensive sleep monitoring and analysis using multivariate polysomnography (PSG) records has achieved significant efforts to prevent sleep-related disorders. To alleviate the time consumption caused by manual visual inspection of PSG, automatic multivariate sleep stage classification has become an important research topic in medical and bioinformatics. RESULTS: We present a unified hybrid self-attention deep learning framework, namely HybridAtt, to automatically classify sleep stages by capturing channel and temporal correlations from multivariate PSG records. We construct a new multi-view convolutional representation module to learn channel-specific and global view features from the heterogeneous PSG inputs. The hybrid attention mechanism is designed to further fuse the multi-view features by inferring their dependencies without any additional supervision. The learned attentional representation is subsequently fed through a softmax layer to train an end-to-end deep learning model. CONCLUSIONS: We empirically evaluate our proposed HybridAtt model on a benchmark PSG dataset in two feature domains, referred to as the time and frequency domains. Experimental results show that HybridAtt consistently outperforms ten baseline methods in both feature spaces, demonstrating the effectiveness of HybridAtt in the task of sleep stage classification.

The recent advances in pervasive sensing technologies have enabled us to monitor and analyze the multi-channel electroencephalogram (EEG) signals of epilepsy patients to prevent serious outcomes caused by epileptic seizures. To avoid manual visual inspection from long-term EEG readings, automatic EEG seizure detection has garnered increasing attention among researchers. In this paper, we present a unified multi-view deep learning framework to capture brain abnormalities associated with seizures based on multi-channel scalp EEG signals. The proposed approach is an end-to-end model that is able to jointly learn multi-view features from both unsupervised multi-channel EEG reconstruction and supervised seizure detection via spectrogram representation. We construct a new autoencoder-based multi-view learning model by incorporating both inter and intra correlations of EEG channels to unleash the power of multi-channel information. By adding a channel-wise competition mechanism in the training phase, we propose a channel-aware seizure detection module to guide our multi-view structure to focus on important and relevant EEG channels. To validate the effectiveness of the proposed framework, extensive experiments against nine baselines, including both traditional handcrafted feature extraction and conventional deep learning methods, are carried out on a benchmark scalp EEG dataset. Experimental results show that the proposed model is able to achieve higher average accuracy and f1-score at 94.37% and 85.34%, respectively, using 5-fold subject-independent cross validation, demonstrating a powerful and effective method in the task of EEG seizure detection.

Recent advances in unsupervised language processing methods have created an opportunity to exploit massive text corpora for developing high-quality vector space representation (also known as word embeddings) of words. Towards this direction, practitioners have developed and applied several data driven embedding models with quite good rate of success. However, a drawback of these models lies in their premise of static context; wherein, the meaning of a word is assumed to remain the same over the period of time. This is limiting because it is known that the semantic meaning of a concept evolves over time. While such semantic drifts are routinely observed in almost all the domains; their effect is acute in domain such as biomedicine, where the semantic meaning of a concept changes relatively fast. To address this, in this study, we aim to learn temporally aware vector representation of medical concepts from the timestamped text data, and in doing so provide a systematic approach to formalize the problem. More specifically, a dynamic word embedding based model that jointly learns the temporal characteristics of medical concepts and performs across time-alignment is proposed. Apart from capturing the evolutionary characteristics in an optimal manner, the model also factors in the implicit medical properties useful for a variety of bio-medical applications. Empirical studies conducted on two important bio-medical use cases validates the effectiveness of the proposed approach and suggests that the model not only learns quality embeddings but also facilitates intuitive trajectory visualizations.

The demand for automatic extraction of true information (i.e., truths) from conflicting multi-source data has soared recently. A variety of truth discovery methods have witnessed great successes via jointly estimating source reliability and truths. All existing truth discovery methods focus on providing a point estimator for each object's truth, but in many real-world applications, confidence interval estimation of truths is more desirable, since confidence interval contains richer information. To address this challenge, in this paper, we propose a novel truth discovery method (ETCIBoot) to construct confidence interval estimates as well as identify truths, where the bootstrapping techniques are nicely integrated into the truth discovery procedure. Due to the properties of bootstrapping, the estimators obtained by ETCIBoot are more accurate and robust compared with the state-of-the-art truth discovery approaches. The proposed framework is further adapted to deal with large-scale truth discovery task in distributed paradigm. Theoretically, we prove the asymptotical consistency of the confidence interval obtained by ETCIBoot. Experimentally, we demonstrate that ETCIBoot is not only effective in constructing confidence intervals but also able to obtain better truth estimates.

Word embeddings are finding their increasing application in a variety of biomedical Natural Language Processing (bioNLP) tasks, ranging from drug discovery to automated disease diagnosis. While these word embeddings in their entirety have shown meaningful syntactic and semantic regularities, however, the meaning of individual dimensions remains elusive. This becomes problematic both in general and particularly in sensitive domains such as bio-medicine, wherein, the interpretability of results is crucial to its widespread adoption. To address this issue, in this study, we aim to improve the interpretability of pre-trained word embeddings generated from a text corpora, and in doing so provide a systematic approach to formalize the problem. More specifically, we exploit the rich categorical knowledge present in the biomedical domain, and propose to learn a transformation matrix that transforms the input embeddings to a new space where they are both interpretable and retain their original expressive features. Experiments conducted on the largest available biomedical corpus suggests that the model is capable of performing interpretability that resembles closely to the human-level intuition.

A clinically meaningful distance metric, which is learned from measuring patient similarity, plays an important role in clinical decision support applications. Several metric learning approaches have been proposed to measure patient similarity, but they are mostly designed for learning the metric at only one time point/interval. It leads to a problem that those approaches cannot reflect the similarity variations among patients with the progression of diseases. In order to capture similarity information from multiple future time points simultaneously, we formulate a multi-task metric learning approach to identify patient similarity. However, it is challenging to directly apply traditional multi-task metric learning methods to learn such similarities due to the high dimensional, complex and noisy nature of healthcare data. Besides, the disease labels often have clinical relationships, which should not be treated as independent. Unfortunately, traditional formulation of the loss function ignores the degree of labels' similarity. To tackle the aforementioned challenges, we propose mtTSML, a multi-task triplet constrained sparse metric learning method, to monitor the similarity progression of patient pairs. In the proposed model, the distance for each task can be regarded as the combination of a common part and a task-specific one in the transformed low-rank space. We then perform sparse feature selection for each individual task to select the most discriminative information. Moreover, we use triplet constraints to guarantee the margin between similar and less similar pairs according to the ordered information of disease severity levels (i.e. labels). The experimental results on two real-world healthcare datasets show that the proposed multi-task metric learning method significantly outperforms the state-of-the-art baselines, including both single-task and multi-task metric learning methods.

Recent advances in attention networks have gained enormous interest in time series data mining. Various attention mechanisms are proposed to soft-select relevant timestamps from temporal data by assigning learnable attention scores. However, many real-world tasks involve complex multivariate time series that continuously measure target from multiple views. Different views may provide information of different levels of quality varied over time, and thus should be assigned with different attention scores as well. Unfortunately, the existing attention-based architectures cannot be directly used to jointly learn the attention scores in both time and view domains, due to the data structure complexity. Towards this end, we propose a novel multi-view attention network, namely MuVAN, to learn fine-grained attentional representations from multivariate temporal data. MuVAN is a unified deep learning model that can jointly calculate the two-dimensional attention scores to estimate the quality of information contributed by each view within different timestamps. By constructing a hybrid focus procedure, we are able to bring more diversity to attention, in order to fully utilize the multi-view information. To evaluate the performance of our model, we carry out experiments on three real-world benchmark datasets. Experimental results show that the proposed MuVAN model outperforms the state-of-the-art deep representation approaches in different real-world tasks. Analytical results through a case study demonstrate that MuVAN can discover discriminative and meaningful attention scores across views over time, which improves the feature representation of multivariate temporal data.

Metric learning aims to learn a good distance metric that can capture the relationships among instances, and its importance has long been recognized in many fields. In the traditional settings of metric learning, an implicit assumption is that the associated labels of the instances are deterministic. However, in many real-world applications, the associated labels come naturally with probabilities instead of deterministic values. Thus, the existing metric learning methods cannot work well in these applications. To tackle this challenge, in this paper, we study how to effectively learn the distance metric from datasets that contain probabilistic information, and then propose two novel metric learning mechanisms for two types of probabilistic labels, i.e., the instance-wise probabilistic label and the group-wise probabilistic label. Compared with the existing metric learning methods, our proposed mechanisms are capable of learning distance metrics directly from the probabilistic labels with high accuracy. We also theoretically analyze the two proposed mechanisms and provide theoretical bounds on the sample complexity for both of them. Additionally, extensive experiments based on real-world datasets are conducted to verify the desirable properties of the proposed mechanisms.

Given two topics of interest (A, C) that are otherwise disconnected - for instance two concepts: a disease ("Migraine") and a therapeutic substance ("Magnesium") - this paper attempts to find the conceptual bridges (e.g., serotonin (B)) that connects them in a novel way. This problem of mining implicit linkage is known as hypotheses generation and its potential to accelerate scientific progress is widely recognized. Almost all of the prior studies to tackle this problem ignore the temporal dynamics of concepts. This is limiting because it is known that the semantic meaning of a concept evolves over time. To overcome this issue, in this study, we define this problem as mining time-aware Top-k conceptual bridges, and in doing so provide a systematic approach to formalize the problem. Specifically, the proposed model first extracts relevant entities from the corpus, represents them in time-specific latent spaces, and then further reasons upon it to generate novel and experimentally testable hypotheses. The key challenge in this approach is to learn a mapping function that encodes the temporal characteristics of concepts and aligns the across-time latent spaces. To solve this, we propose an effective algorithm that learns precise mapping sensitive to both global and local semantics of the input query. Both qualitative and quantitative evaluations performed on the largest available biomedical corpus substantiate the importance of leveraging the evolutionary semantics of medical concepts and suggests that the generated hypotheses are novel and worthy of clinical trials.

Predicting the risk of potential diseases from Electronic Health Records (EHR) has attracted considerable attention in recent years, especially with the development of deep learning techniques. Compared with traditional machine learning models, deep learning based approaches achieve superior performance on risk prediction task. However, none of existing work explicitly takes prior medical knowledge (such as the relationships between diseases and corresponding risk factors) into account. In medical domain, knowledge is usually represented by discrete and arbitrary rules. Thus, how to integrate such medical rules into existing risk prediction models to improve the performance is a challenge. To tackle this challenge, we propose a novel and general framework called PRIME for risk prediction task, which can successfully incorporate discrete prior medical knowledge into all of the state-of-the-art predictive models using posterior regularization technique. Different from traditional posterior regularization, we do not need to manually set a bound for each piece of prior medical knowledge when modeling desired distribution of the target disease on patients. Moreover, the proposed PRIME can automatically learn the importance of different prior knowledge with a log-linear model.Experimental results on three real medical datasets demonstrate the effectiveness of the proposed framework for the task of risk prediction

Estimating individual treatment effect (ITE) is a challenging problem in causal inference, due to the missing counterfactuals and the selection bias. Existing ITE estimation methods mainly focus on balancing the distributions of control and treated groups, but ignore the local similarity information that is helpful. In this paper, we propose a local similarity preserved individual treatment effect (SITE) estimation method based on deep representation learning. SITE preserves local similarity and balances data distributions simultaneously, by focusing on several hard samples in each mini-batch. Experimental results on synthetic and three real-world datasets demonstrate the advantages of the proposed SITE method, compared with the state-of-the-art ITE estimation methods.

Patient similarity learning aims to derive a clinically meaningful similarity metric to measure the similarity between a pair of patients according to their historical clinical information, which could help to predict the clinical outcomes of the patient of interest. However, the patient clinical data are usually complex, and contain much irrelevant and redundant information, which makes it difficult to learn the similarity metric with high accuracy. Although some methods have been proposed to address the complex nature of patient data, they overemphasize sparsity-based relevant feature selection and fail to take into consideration the redundant features that are highly correlated with each other, and this heavily degrades the accuracy of the learned results. To address the above challenges, we propose a novel uncorrelated patient similarity learning approach, which can not only select the most relevant features for the learning task, but also guarantee that the selected features have low correlations with each other. Additionally, to address the scenarios where the patient data are distributed across different sites, we extend the proposed approach and design a distributed mechanism, based on which the similarity metric can be accurately learned without directly accessing the raw patient data at each site. The desirable performance of the proposed methods are verified through extensive experiments conducted on both real-world and synthetic datasets.

Truth discovery, with the goal of inferring true information from massive data through aggregating the information from multiple data sources, has attracted significant attention in recent years. It has demonstrated great advantages in real applications since it can automatically learn the reliability degrees of the data sources without supervision and in turn helps to find more reliable information. In many applications, however, the data may arrive in a stream and present various temporal patterns. Unfortunately, there is no existing truth discovery work that can handle such time series data. To tackle this challenge, we propose a novel online truth discovery framework that incorporates the predictions on the time series data into the truth estimation process. By jointly considering the multi-source information and the temporal patterns of the time series data, the proposed framework can improve the accuracy of the truth discovery results as well as the time series prediction. The effectiveness of the proposed framework is validated on both synthetic and real-world datasets.

Motivation: The overwhelming amount of research articles in the domain of bio-medicine might cause important connections to remain unnoticed. Literature Based Discovery is a sub-field within biomedical text mining that peruses these articles to formulate high confident hypotheses on possible connections between medical concepts. Although many alternate methodologies have been proposed over the last decade, they still suffer from scalability issues. The primary reason, apart from the dense inter-connections between biological concepts, is the absence of information on the factors that lead to the edge-formation. In this work, we formulate this problem as a collaborative filtering task and leverage a relatively new concept of word-vectors to learn and mimic the implicit edge-formation process. Along with single-class classifier, we prune the search-space of redundant and irrelevant hypotheses to increase the efficiency of the system and at the same time maintaining and in some cases even boosting the overall accuracy. Results: We show that our proposed framework is able to prune up to 90% of the hypotheses while still retaining high recall in top-K results. This level of efficiency enables the discovery algorithm to look for higher-order hypotheses, something that was infeasible until now. Furthermore, the generic formulation allows our approach to be agile to perform both open and closed discovery. We also experimentally validate that the core data-structures upon which the system bases its decision has a high concordance with the opinion of the experts.This coupled with the ability to understand the edge formation process provides us with interpretable results without any manual intervention. Availability and implementation: The relevant JAVA codes are available at: https://github.com/vishrawas/Medline-Code_v2. Supplementary information: Supplementary data are available at Bioinformatics online.

Clustering cancer patients into subgroups and identifying cancer subtypes is an important task in cancer genomics. Clustering based on comprehensive multi-omic molecular profiling can often achieve better results than those using a single data type, since each omic data type may contain complementary information. However, it is challenging to integrate heterogeneous omic data directly. Based on one popular method - Similarity Network Fusion (SNF), we presented Affinity Network Fusion (ANF), an “upgrade” of SNF with several advantages. Similar to SNF, ANF treats each omic data type as one view of patients and learns a fused affinity matrix for clustering. We applied ANF to a harmonized TCGA dataset consisting of 2193 patients, and generated promising results on clustering patients into correct disease types. Our experimental results also demonstrated the power of feature selection and transformation combined with using ANF in patient clustering. Moreover, eigengap analysis suggests that the learned affinity matrices of four cancer types using our proposed framework may have successfully captured patient group structure and can be used for discovering unknown cancer subtypes.

Predicting patients' risk of developing certain diseases is an important research topic in healthcare. Personalized predictive modeling, which focuses on building specific models for individual patients, has shown its advantages on utilizing heterogeneous health data compared to global models trained on the entire population. Personalized predictive models use information from similar patient cohorts, in order to capture the specific characteristics. Accurately identifying and ranking the similarity among patients based on their historical records is a key step in personalized modeling. The electric health records (EHRs), which are irregular sampled and have varied patient visit lengths, cannot be directly used to measure patient similarity due to lack of an appropriate vector representation. In this paper, we build a novel time fusion CNN framework to simultaneously learn patient representations and measure pairwise similarity. Compared to a traditional CNN, our time fusion CNN can learn not only the local temporal relationships but also the contributions from each time interval. Along with the similarity learning process, the output information which is the probability distribution is used to rank similar patients. Utilizing the similarity scores, we perform personalized disease predictions, and compare the effect of different vector representations and similarity learning metrics.

Epileptic seizure detection has gained increasing attention in clinical therapy. Scalp electroencephalogram (EEG) analysis is a common way to capture brain abnormality for seizure onset detection. This paper presents a novel context-learning based approach using multi-feature fusion to compensate for incomplete description of single feature in epileptic EEG signals. First, EEG scalogram sequence is generated using wavelet transform to represent the time-frequency information. Second, three sets of EEG context features are unsupervisedly learned in parallel by using global principal component analysis (GPCA), stacked denoising autoencoders (SDAEs) and EEG embeddings, respectively. Finally, the multi-features are concatenated into a fixed-length feature vector for seizure classification. The experimental results conducted on two real EEG datasets demonstrate that the proposed cross-patient learning model is able to extract meaningful context features from different perspectives, and hence can detect the onset of epileptic seizure effectively.

The technological advancements in biomedical domain has led to a tremendous growth of unstructured data; primarily a result of increased publication of findings. At the same time, a corresponding interest in the Natural Language Processing (NLP) community to develop scalable methodologies to exploit such massive unlabeled corpora for unsupervised language processing has resulted in new opportunities towards developing semantic sensitive models. Amongst them, the field of word embeddings has garnered significant attention due to its capability to understand implicit semantics. However such data driven models are largely agnostic of the rich explicit semantic knowledge available in the biomedical domain in the form of vocabularies and ontologies. This is problematic because it leads to a poor representation of words with little local context and its effect is acute in biomedical domain. In this paper, we propose a novel model (MeSH2Vec) that jointly exploits both contextual information and available explicit semantic knowledge to learn externally augmented word embeddings. Unlike existing approaches, the proposed methodology is more dexterous in its ability to handle relationships between indirectly related concepts. The 13% improvement in the correlation to experts, shown on experiments involving biomedical concept similarity and relatedness task validates the effectiveness of the proposed approach and demonstrates the importance of incorporating human curated knowledge in the process of generating word embeddings.

Literature based discovery (LBD) is a task that aims to uncover hidden associations between non-interacting scientific concepts by rationally connecting independent nuggets of information. Broadly, prior approaches to LBD include use of: a) distributional statistics and explicit representation, b) graph-theoretic measures, and c) supervised machine learning methods to find associations. However, purely distributional approaches may not necessarily entail semantically meaningful association and graph-theoretic approaches suffer from scalability issues. While supervised machine learning based approaches have the potential to elucidate associations, the training data required is too expensive to generate. In this paper we propose a novel dynamic Medical Subject Heading (MeSH) embedding model which is able to model the evolutionary behavior of medical concepts to uncover latent associations between them. The proposed model allows us to learn the evolutionary trajectories of MeSH embeddings and detect informative terms. Hence, based on the dynamic MeSH embeddings, meaningful medical hypotheses can be efficiently generated. To evaluate the efficacy of the proposed model, we perform both qualitative and quantitative evaluation. The results demonstrate that leveraging the evolutionary features of MeSH concepts is an effective way for predicting novel associations.

Time series data mining has gained increasing attention in health domain. Recently, researchers attempt to employ Natural Language Processing (NLP) to health data mining, in order to learn proper representations of discrete medical concepts from Electronic Health Records (EHRs). However, existing models do not take continuous physiological records into account, which are naturally existed in EHRs. The major challenges for this task are to model non-obvious representations from observed high dimensional biosignals, and to interpret the learned features. To address these issues, we propose Wave2Vec, an end-to-end deep learning model, to bridge the gap between biosignal processing and language modeling. Wave2Vec jointly learns both inherent and embedding representations of biosignals at the same time. To evaluate the performance of our model in clinical task, we carry out experiments on two real world benchmark biosignal datasets. Experimental results show that the proposed Wave2Vec model outperforms the six feature leaning baselines in biosignal processing.

Conventional correlated topic models are able to capture correlation structure among latent topics by replacing the Dirichlet prior with the logistic normal distribution. Word embeddings have been proven to be able to capture semantic regularities in language. Therefore, the semantic relatedness and correlations between words can be directly calculated in the word embedding space, for example, via cosine values. In this paper, we propose a novel correlated topic model using word embeddings. The proposed model enables us to exploit the additional word-level correlation information in word embeddings and directly model topic correlation in the continuous word embedding space. In the model, words in documents are replaced with meaningful word embeddings, topics are modeled as multivariate Gaussian distributions over the word embeddings and topic correlations are learned among the continuous Gaussian topics. A Gibbs sampling solution with data augmentation is given to perform inference. We evaluate our model on the 20 Newsgroups dataset and the Reuters-21578 dataset qualitatively and quantitatively. The experimental results show the effectiveness of our proposed model.

Drug side-effects become a worldwide public health concern, which are the fourth leading cause of death in the United States. Pharmaceutical industry has paid tremendous effort to identify drug side-effects during the drug development. However, it is impossible and impractical to identify all of them. Fortunately, drug side-effects can also be reported on heterogeneous platforms (i.e., data sources), such as FDA Adverse Event Reporting System and various online communities. However, existing supervised and semi-supervised approaches are not practical as annotating labels are expensive in the medical field. In this paper, we propose a novel and effective unsupervised model Sifter to automatically discover drug side-effects. Sifter enhances the estimation on drug side-effects by learning from various online platforms and measuring platform-level and user-level quality simultaneously. In this way, Sifter demonstrates better performance compared with existing approaches in terms of correctly identifying drug side-effects. Experimental results on five real-world datasets show that Sifter can significantly improve the performance of identifying side-effects compared with the state-of-the-art approaches.

A text corpus typically contains two types of context information -- global context and local context. Global context carries topical information which can be utilized by topic models to discover topic structures from the text corpus, while local context can train word embeddings to capture semantic regularities reflected in the text corpus. This encourages us to exploit the useful information in both the global and the local context information. In this paper, we propose a unified language model based on matrix factorization techniques which 1) takes the complementary global and local context information into consideration simultaneously, and 2) models topics and learns word embeddings collaboratively. We empirically show that by incorporating both global and local context, this collaborative model can not only significantly improve the performance of topic discovery over the baseline topic models, but also learn better word embeddings than the baseline word embedding models. We also provide qualitative analysis that explains how the cooperation of global and local context information can result in better topic structures and word embeddings.

Omic data analyses pose great informatics challenges. As an emerging subfield of bioinformatics, omics informatics focuses on analyzing multi-omic data efficiently and effectively, and is gaining momentum. There are two underlying trends in the expansion of omics informatics landscape: the explosion of scattered individual omics informatics tools with each of which focuses on a specific task in both single- and multi- omic settings, and the fast-evolving integrated software platforms such as workflow management systems that can assemble multiple tools into pipelines and streamline integrative analysis for complicated tasks. In this survey, we give a holistic view of omics informatics, from scattered individual informatics tools to integrated workflow management systems. We not only outline the landscape and challenges of omics informatics, but also sample a number of widely used and cutting-edge algorithms in omics data analysis to give readers a fine-grained view. We survey various workflow management systems (WMSs), classify them into three levels of WMSs from simple software toolkits to integrated multi-omic analytical platforms, and point out the emerging needs for developing intelligent workflow management systems. We also discuss the challenges, strategies and some existing work in systematic evaluation of omics informatics tools. We conclude by providing future perspectives of emerging fields and new frontiers in omics informatics.

Electronic medical records (EMRs) store data related to patients information enrolled during their stay in health structures. Data stored into EMRs span from data crawled from biological laboratories to textual description of diseases and diagnostic device results (e.g., biomedical images). Each EMR is related to a diagnosis related group (DRG) record. A DRG record is a record associated with a citizen that has been cured in a hospital. It contains a code, called major diagnostic category (MDC), which summarizes the treated disease and allows to reimburse costs related to patient treatments during his staying in health structures. DRGs are used for administrative process (e.g., costs and reimbursement management) as well as disease monitoring. Associating diagnostic codes with external information (such as environmental and geographical data) and with information filtered from EMRs (e.g., biological results or analytes values) can be useful to monitor citizens wellness status. We propose a methodology to analyze such data based on a multistep process. First, we cross reference data by using a semantics-based clustering procedure, extract information from EMRs, and then, cluster them by looking for similar patterns of diseases. Then, biological records in each disease cluster are analyzed to evaluate intracluster similarity by selecting analytes typologies and values. Finally, biological data is related to diagnosis codes and geometrically projected in areas of interest in order to map calculated outlier patients. We applied the methodology on two case studies: 1) diagnosis codes and biochemical analytes of 20 000 biological analyses about hospitalized patients during one observation year and 2) the correlation between cardiovascular diseases and water quality in a southern Italian region. Preliminary findings show the effectiveness of our method.

Learning semantics based on context information has been researched in many research areas for decades. Context information can not only be directly used as the input data, but also sometimes used as auxiliary knowledge to improve existing models. This survey aims at providing a structured and comprehensive overview of the research on context learning. We summarize and group the existing literature into four categories, Explicit Analysis, Implicit Analysis, Neural Network Models, and Composite Models, based on the underlying techniques adopted by them. For each category, we talk about the basic idea and techniques, and also introduce how context information is utilized as the model input or incorporated into the model to enhance the performance or extend the domain of application as auxiliary knowledge. In addition, we discuss the advantages and disadvantages of each model from both the technical and practical point of view.

The development of Internet and social networks has provided more emerging network data which facilitates the dynamic network analysis. In this paper, we propose a new method to measure coherence strength, also referred to as community consistency, of a community under dynamic settings. In order to better interpret the influence of evolving community structure on community consistency, we model the problem as one of influence propagation processes having a causal relation with the community consistency. To this effect a generative model is proposed to combine the influence propagation and the network topological structure at each time stamp. Our comprehensive experiments on both synthetic and real-world datasets demonstrate the superiority of the proposed framework in estimating the community consistency.

The development of social networks has not only improved the online experience, but also stimulated the advances in knowledge mining so as to assist people in planning their offline social events. Users can explore their favorite events, such as celebrations and symposiums, through the pictures and the posts from their friends on social networks. An effective event recommendation can offer great convenience for both event organizers and participants, which yet remains extremely challenging due to a wide range of practical concerns. In this paper we propose a novel recommendation framework, which combines the information from multiple sources and establishes a connection between the online knowledge and the event participation.

Identification of protein complexes is very important to investigate the characteristics of biological processes. Most of existing protein complex clustering algorithms were often run only on a static protein-protein interaction (PPI) network. The dynamic characteristics of interactions were ignored. In order to solve the problem, a new clustering algorithm (TP-WDPIN) was proposed which is based on the concept of topological potential to measure the importance of proteins in the process of detecting seed proteins and then to mine protein complexes from weighted dynamic PPI network. The algorithm used features of core-attachment of complexes and split low density cores to improve density of cores for achieving better clustering results. Experiment results showed that the proposed TP-WDPIN algorithm has better performance than other algorithms on two PPI databases.

Identifying brain effective connectivity networks from functional magnetic resonance imaging (fMRI) data is an important advanced subject in neuroinformatics in recent years, where the learning method based on bayesian networks (BN) has become a new hot topic in the field. This paper proposes a new method to learn the brain effective connectivity network structure by combining ant colony optimization (ACO) with BN method, named as ACOEC. In the proposed algorithm, a brain effective connectivity network is first mapped onto an ant, and then the ant colony optimization by simulating real ants looking for food is employed to construct network structures and finally an ant with the highest score is obtained as the optimal solution. The experimental results on simulated and real fMRI data sets show that the new method can not only accurately identify the connections and directions of the brain networks, but also quantitatively describe the connection strength of the brain networks, which has a good clinical application prospects.

Functional gene network analysis, such as gene co-expression network analysis, is useful for detecting disease-associated gene modules. Compared with many gene interaction networks in pathway databases, co-expression networks constructed directly from RNA-seq experiment are context-specific and thus more helpful for detecting differential gene modules under defined conditions. However, existing co-expression network inference approaches for RNA-seq data suffer from high noise and biases due to small sample sizes and many confounding factors. In this paper we proposed a framework for constructing robust, context-specific differential gene co-expression networks consisting of only high confidence edges. To detect disease-associated submodules, we devised a new metric to measure module significance scores. Based on this metric, we developed a gene ontology(GO)-driven module discovery algorithm for identifying disease-associated differential modular structures. Experiments on real RNA-seq data shows this framework works well in detecting biologically meaningful signals.

Identifying functional modules in protein-protein interaction (PPI) networks is fundamental to understand cellular organization, processes, and functions. As an emerging evolutionary computational technology, swarm intelligence approaches are now becoming a new research hotspot in identifying functional modules. This paper proposes a new computational approach based on bacterial biological mechanisms for functional module detection in PPI networks (called as BBM-FMD). In BBM-FMD, each bacterium is first initialized to a candidate module partition by a random walk behavior. Then four biological mechanisms of bacteria including chemotaxis, conjugation, reproduction, and elimination and dispersal are simulated to iteratively search for better protein module partitions. At last, two post-processing steps are carried out to refine the obtained module partition. The experimental results on two PPI datasets demonstrate the superior performance of BBM-FMD in detecting functional modules compared with several other algorithms.

Discovering topics in short texts, such as news titles and tweets, has become an important task for many content analysis applications. However, due to the lack of rich context information in short texts, the performance of conventional topic models on short texts is usually unsatisfying. In this paper, we propose a novel topic model for short text corpus using word embeddings. Continuous space word embeddings, which is proven effective at capturing regularities in language, is incorporated into our model to provide additional semantics. Thus we model each short document as a Gaussian topic over word embeddings in the vector space. In addition, considering that background words in a short text are usually not semantically related, we introduce a discrete background mode over word types to complement the continuous Gaussian topics. We evaluate our model on news titles from data sources like abcnews, showing that our model is able to extract more coherent topics from short texts compared with the baseline methods and learn better topic representation for each short document.

The demand for automatic extraction of true information (i.e., truths) from conflicting multi-source data has soared recently. A variety of truth discovery methods have witnessed great successes via jointly estimating source reliability and truths. All existing truth discovery methods focus on providing a point estimator for each object's truth, but in many real-world applications, confidence interval estimation of truths is more desirable, since confidence interval contains richer information. To address this challenge, in this paper, we propose a novel truth discovery method (ETCIBoot) to construct confidence interval estimates as well as identify truths, where the bootstrapping techniques are nicely integrated into the truth discovery procedure. Due to the properties of bootstrapping, the estimators obtained by ETCIBoot are more accurate and robust compared with the state-of-the-art truth discovery approaches. Theoretically, we prove the asymptotical consistency of the confidence interval obtained by ETCIBoot. Experimentally, we demonstrate that ETCIBoot is not only effective in constructing confidence intervals but also able to obtain better truth estimates.

Detecting functional modules from a Protein-Protein Interaction (PPI) network is a fundamental and hot issue in proteomics research, where many computational approaches have played an important role in recent years. However, how to effectively and efficiently detect functional modules in large-scale PPI networks is still a challenging problem. We present a new framework, based on a multiple-grain model of PPI networks, to detect functional modules in PPI networks. First, we give a multiple-grain representation model of a PPI network, which has a smaller scale with super nodes. Next, we design the protein grain partitioning method, which employs a functional similarity or a structural similarity to merge some proteins layer by layer. Thirdly, a refining mechanism with border node tests is proposed to address the protein overlapping of different modules during the grain eliminating process. Finally, systematic experiments are conducted on five large-scale yeast and human networks. The results show that the framework not only significantly reduces the running time of functional module detection, but also effectively identifies overlapping modules while keeping some competitive performances, thus it is highly competent to detect functional modules in large-scale PPI networks.

Despite the ability to model many real world settings as a network, one major challenge in analyzing network data is that important and reliable links between objects are usually obscured by noisy information and hence not readily discernible. In this paper, we propose to detect these important and reliable links - significant edges, from a target network by using multiple auxiliary networks and a limited amount of labelled information. In this process, we first abstract the community knowledge learnt across target and auxiliary networks to detect significant patterns. The mined community knowledge captures the key profile of network relationships and thus can be used to determine whether an existing edge indicates a true or false relationship. Experiments on real world network data show that our two staged solution -- a joint matrix factorisation procedure followed by edge significance score ranking, accurately predicts significant edges in target network by jointly exploring the underlying knowledge embedded in both target and auxiliary networks.

Epileptic seizure is a serious health problem in the world and there is a huge population suffering from it every year. If an algorithm could automatically detect seizures and deliver the patient therapy or notify the hospital, that would be of great assistance. Analyzing the scalp EEG is the most common way to detect the onset of a seizure. In this paper, we proposed the context-learning based EEG analysis for seizure detection (Context-EEG) algorithm. The proposed method aims at extracting both the hidden inherent features within EEG fragments and the temporal features from EEG contexts. First, we segment the EEG signals into EEG fragments of fixed length. Second, we learn the hidden inherent features from each fragment and reduce the dimensionality of the original data. Third, we translate each EEG fragment to an EEG word so that the EEG context can provide us with temporal information. And finally, we concatenate the hidden feature and the temporal feature together to train a binary classifier. The experiment result shows the proposed model is highly effective in detecting seizure.

There is a growing interest in identifying inorganic material affinity classes for peptide sequences due to the development of bionanotechnology and its wide applications. In particular, a selective model capable of learning cross-material affinity patterns can help us design peptide sequences with desired binding selectivity for one inorganic material over another. However, as a newly emerging topic, there are several distinct challenges of it that limit the performance of many existing peptide sequence classification algorithms. In this paper, we propose a novel framework to identify affinity classes for peptide sequences across inorganic materials. After enlarging our dataset by simulating peptide sequences, we use a context learning based method to obtain the vector representation of each amino acid and each peptide sequence. By analyzing the structure and affinity class of each peptide sequence, we are able to capture the semantics of amino acids and peptide sequences in a vector space. At the last step we train our classifier based on these vector features and the heuristic rules. The construction of our models gives us the potential to overcome the challenges of this task and the empirical results show the effectiveness of our models.

The increasing popularity of mobile devices has brought severe challenges to device usability and big data analysis. In this paper we investigate the intellectual recommender system on cell phones by incorporating mobile data analysis. Nowadays with the development of smart phones, more and more applications have emerged on various areas, such as entertainment, education and health care. While these applications have brought great convenience to people's daily life, they also provide tremendous opportunities for analyzing users' interests. In this work we develop an Android background service to collect the user behaviors and analyze their preferences based on their Android application usage. As one of the most intuitive media for visual representation, videos with various types of contents are recommended to users based on a proposed graphical model. The proposed model jointly utilizes the textual descriptions of Android applications and videos, as well as the extracted video content based features. Besides, by analyzing the user's habit of application usage we seamlessly integrate the user's personal interests during the recommendation. The extensive comparisons to multiple baselines reveal the superiority of the proposed model on the recommendation quality. Furthermore, we conduct experiments on personalized recommendation to demonstrate the capacity of the proposed model in effectively analyzing the user's personal interests.

Morden physiological analysis begins to involve more and more types of information. Electroencephalogram (EEG) signals as a typical example is starting to be analyzed with facial expressions videos to detect emotions. Emotions play an important role in the daily life of human beings, the need and importance of automatic emotion recognition has grown with increasing role of human computer interface applications. In this paper, we concentrate on recognition of the emotions jointly from "inner" and "outer" reactions, which are electroencephalogram (EEG) signals and facial expression video. Due to the streaming nature of this problem, the data volume and velocity is very challenging. We address these challenges from the theoretic perspective and propose a real time algorithm based on EEG signals and synchronized facial video to learn feature vector jointly. Our algorithm consists of an unsupervisedly EEG dictionary component based on deep learning theorem, and a probability pooling component transforms a continuous sequential signal into an EEG "sentence" which consists of a sequence of EEG words. The EEG sentence is then jointly learned with video features into a new fixed length feature representation for emotion classification. We overcome several computational challenges on the data based on the idea of convolution and pooling, and we conduct extensive evaluation for each component of our model. We also demonstrate the state-of-the-art classification result on real-world dataset. The superior performances on the emotion recognition task indicates that 1) the natural language scenario can be applied in EEG sequences and 2) borrowing video modality can increase the overall performance.

Sequential data modeling has received growing interests due to its impact on real world problems. Sequential data is ubiquitous - financial transactions, advertise conversions and disease evolution are examples of sequential data. A long-standing challenge in sequential data modeling is how to capture the strong hidden correlations among complex features in high volumes. The sparsity and skewness in the features extracted from sequential data also add to the complexity of the problem. In this paper, we address these challenges from both discriminative and generative perspectives, and propose novel stochastic learning algorithms to model nonlinear variances from static time frames and their transitions. The proposed model, Deep Recurrent Network (DRN), can be trained in an unsupervised fashion to capture transitions, or in a discriminative fashion to conduct sequential labeling. We analyze the conditional independence of each functional module and tackle the diminishing gradient problem by developing a two-pass training algorithm. Extensive experiments on both simulated and real-world dynamic networks show that the trained DRN outperforms all baselines in the sequential classification task and obtains excellent performance in the regression task.

Networks, which characterize object relationships, are ubiquitous in various domains. One very important problem is to detect the nodes of a specific function in these networks. For example, is a user normal or anomalous in an email network? Does a protein play a key role in a protein-protein interaction network? In many applications, the information we have about the networks usually includes both node characteristics and network structures. Both types of information can contribute to the task of learning functional nodes, and we call the collection of node and link information as linked data. However, existing methods only use a few subjectively selected topological features from network structures to detect functional nodes, thus fail to include highly discriminative and meaningful patterns hidden in linked data. To address this problem, a novel Feature Integration based Functional Node Detection (FIND) algorithm is presented. Specifically, FIND extracts the most discriminative information from both node characteristics and network structures in the form of a unified latent feature representation with the guidance of several labeled nodes. Experiments on two real world data sets validate that the proposed method significantly outperforms the baselines on the detection of three different types of functional nodes.

Rapid advances in bionanotechnology have recently generated growing interest in identifying peptides that bind to inorganic materials and classifying them based on their inorganic material affinities. However, there are some distinct characteristics of inorganic materials binding sequence data that limit the performance of many widely-used classification methods when applied to this problem. In this paper, we propose a novel framework to predict the affinity classes of peptide sequences with respect to an associated inorganic material. We first generate a large set of simulated peptide sequences based on an amino acid transition matrix tailored for the specific inorganic material. Then the probability of test sequences belonging to a specific affinity class is calculated by minimizing an objective function. In addition, the objective function is minimized through iterative propagation of probability estimates among sequences and sequence clusters. Results of computational experiments on two real inorganic material binding sequence data sets show that the proposed framework is highly effective for identifying the affinity classes of inorganic material binding sequences. Moreover, the experiments on the structural classification of proteins (SCOP) data set shows that the proposed framework is general and can be applied to traditional protein sequences.

With the booming of healthcare industry and the overwhelming amount of electronic health records (EHRs) shared by healthcare institutions and practitioners, we take advantage of EHR data to develop an effective disease risk management model that not only models the progression of the disease, but also predicts the risk of the disease for early disease control or prevention. Existing models for answering these questions usually fall into two categories: the expert knowledge based model or the handcrafted feature set based model. To fully utilize the whole EHR data, we will build a framework to construct an integrated representation of features from all available risk factors in the EHR data and use these integrated features to effectively predict osteoporosis and bone fractures. We will also develop a framework for informative risk factor selection of bone diseases. A pair of models for two contrast cohorts (e.g., diseased patients versus non-diseased patients) will be established to discriminate their characteristics and find the most informative risk factors. Several empirical results on a real bone disease data set show that the proposed framework can successfully predict bone diseases and select informative risk factors that are beneficial and useful to guide clinical decisions.

Community formation analysis of dynamic networks has been a hot topic in data mining which has attracted much attention. Recently, there are many studies which focus on discovering communities successively from consecutive snapshots by considering both the current and historical information. However, these methods cannot provide us with much historical or successive information related to the detected communities. Different from previous studies which focus on community detection in dynamic networks, we define a new problem of tracking the progression of the community strength-a novel measure that reflects the community robustness and coherence throughout the entire observation period. To achieve this goal, we propose a novel framework which formulates the problem as an optimization task. The proposed community strength analysis also provides foundation for a wide variety of related applications such as discovering how the strength of each detected community changes over the entire observation period. To demonstrate that the proposed method provides precise and meaningful evolutionary patterns of communities which are not directly obtainable from traditional methods, we perform extensive experimental studies on one synthetic and five real datasets: Social evolution, tweeting interaction, actor relationships, bibliography, and biological datasets. Experimental results show that the proposed approach is highly effective in discovering the progression of community strengths and detecting interesting communities.

Informative gene selection is an important topic in the field of bioinformatics which has attracted intensive interest in recent years. It aims to identify the genes which are differentially expressed in different groups, and thus are informative for the classification between the groups. For this purpose, many micro array experiments have been conducted by various medical institutes on their own sets of patients and test subjects. For those institutes who have conducted experiments regarding the same type of disease, it would be beneficial to all of them if they learn on the union of their data to find the informative genes instead of learn just on their own datasets, since the amount of data each institute holds is very limited. However, in many cases, the institutes are not allowed to share their data with others because micro array datasets contain private information about the patients and test subjects. In this paper, we focus on this problem and propose a privacy preserving algorithm that allows multiple parties to perform the widely used informative gene selection method, the Fisher criterion, on the union of their data, without revealing each party's data to others. Basically, we utilize the homomorphic cryptographic system to protect the data during the calculations. Experimental results on real world datasets show the effectiveness of the proposed method.

Nowadays the rapid development in the area of human-computer interaction has given birth to a growing interest on detecting different affective states through smart devices. By using the modern sensor equipment, we can easily collect electroencephalogram (EEG) signals, which capture the information from central nervous system and are closely related with our brain activities. Through the training on EEG signals, we can make reasonable analysis on people's affection, which is very promising in various areas. Unfortunately, the special properties of EEG dataset have brought difficulties for conventional machine learning methods. The main reasons lie in two aspects: the small set of labeled samples and the noisy channel problem. To overcome these difficulties and successfully identify the affective states, we come up with a novel semi-supervised deep structured framework. Compared with previous deep learning models, our method is more adapted to the EEG classification problem. We first adopt a two-level procedure, which involves both supervised label information and unsupervised structure information to jointly make decision on channel selection. And then, we add a generative Restricted Boltzmann Machine (RBM) model for the classification task, and use the training objectives of generative learning and unsupervised learning to jointly regularize the discriminative training. Finally, we extend it to the active learning scenario, which solves the costly labeling problem. The experiments conducted on real EEG dataset have shown both the convincing result on critical channel selection and the superiority of our method over multiple baselines for the affective state recognition.

Markov Clustering (MCL) is a popular algorithm for clustering networks in bioinformatics such as Protein-Protein Interaction (PPI) networks and especially, shows excellent performance in clustering Dynamic Proteinprotein Interaction Networks (DPIN). However, a limitation of MCL and its variants (e.g. regularized MCL and soft regularized MCL) is that the clustering results are mostly dependent on the parameters that user-specified. However we know that different networks with various scales need different parameters. In this article, we propose a new MCL method based on the Firefly Algorithm (FA) to optimize its parameters. The results on DIP dataset show that the new algorithm outperforms the state-of-the-art approaches in terms of accuracy of identifying functional modules on a real DPIN.

Temporal analysis on dynamic networks has become a popularly discussed topic today, with more and more emerging data over time. In this paper we investigate the problem of detecting and tracking the variational communities within a given time period. We first define a metric to measure the strength of a community, called the normalized temporal community strength. And then, we propose our analysis framework. The community may evolve over time, either split to multiple communities or merge with others. We address the problem of evolutionary clustering with requirement on temporal smoothness and propose a revised soft clustering method based on non-negative matrix factorization. Then we use a clustering matching method to find the soft correspondence between different community distribution structures. This matching establishes the connection between consecutive snapshots. To estimate the variational rate and meanwhile address the smoothness during continuous evolution, we propose an objective function that combines the conformity of current variation and historical variational trend. In addition, we integrate the weights to the objective function to identify the temporal outliers. An iterative coordinate descent method is proposed to solve the optimization framework. We extensively evaluate our method with a synthetic dataset and several real datasets. The experimental results demonstrate the effectiveness of our method, which is greatly superior to the baselines on detection of the communities with significant variation over time.

Linked data consist of both node attributes, e.g., Preferences, posts and degrees, and links which describe the connections between nodes. They have been widely used to represent various network systems, such as social networks, biological networks and etc. Knowledge discovery on linked data is of great importance to many real applications. One of the major challenges of learning linked data is how to effectively and efficiently extract useful information from both node attributes and links in linked data. Current studies on this topic either use selected topological statistics to represent network structures, or linearly map node attributes and network structures to a shared latent feature space. However, while approaches based on statistics may miss critical patterns in network structure, approaches based on linear mappings may not be sufficient to capture the non-linear characteristics of nodes and links. To handle the challenge, we propose, to our knowledge, the first deep learning method to learn from linked data. A restricted Boltzmann machine model named LRBM is developed for representation learning on linked data. In LRBM, we aim to extract the latent feature representation of each node from both node attributes and network structures, non-linearly map each pair of nodes to the links, and use hidden units to control the mapping. The details of how to adapt LRBM for link prediction and node classification on linked data have also been presented. In the experiments, we test the performance of LRBM as well as other baselines on link prediction and node classification. Overall, the extensive experimental evaluations confirm the effectiveness of the proposed LRBM model in mining linked data.

Time varying problems usually have complex underlying structures represented as dynamic networks where entities and relationships appear and disappear over time. The problem of efficiently performing dynamic link inference is extremely challenging due to the dynamic nature in massive evolving networks especially when there exist sparse connectivities and nonlinear transitional patterns. In this paper, we propose a novel deep learning framework, i.e., Conditional Temporal Restricted Boltzmann Machine (ctRBM), which predicts links based on individual transition variance as well as influence introduced by local neighbors. The proposed model is robust to noise and have the exponential capability to capture nonlinear variance. We tackle the computational challenges by developing an efficient algorithm for learning and inference of the proposed model. To improve the efficiency of the approach, we give a faster approximated implementation based on a proposed Neighbor Influence Clustering algorithm. Extensive experiments on simulated as well as real-world dynamic networks show that the proposed method outperforms existing algorithms in link inference on dynamic networks.

BACKGROUND: Studies of functional modules in a Protein-Protein Interaction (PPI) network contribute greatly to the understanding of biological mechanisms. With the development of computing science, computational approaches have played an important role in detecting functional modules. RESULTS: We present a new approach using multi-agent evolution for detection of functional modules in PPI networks. The proposed approach consists of two stages: the solution construction for agents in a population and the evolutionary process of computational agents in a lattice environment, where each agent corresponds to a candidate solution to the detection problem of functional modules in a PPI network. First, the approach utilizes a connection-based encoding scheme to model an agent, and employs a random-walk behavior merged topological characteristics with functional information to construct a solution. Next, it applies several evolutionary operators, i.e., competition, crossover, and mutation, to realize information exchange among agents as well as solution evolution. Systematic experiments have been conducted on three benchmark testing sets of yeast networks. Experimental results show that the approach is more effective compared to several other existing algorithms. CONCLUSIONS: The algorithm has the characteristics of outstanding recall, F-measure, sensitivity and accuracy while keeping other competitive performances, so it can be applied to the biological study which requires high accuracy.

Title matching refers roughly to the following problem. We are given two strings of text obtained from different data sources. The texts refer to some underlying physical entities and the problem is to report whether the two strings refer to the same physical entity or not. There are manifestations of this problem in a variety of domains, such as product or bibliography matching, and location or person disambiguation. We propose a new approach to solving this problem, consisting of two main components. The first component uses Web searches to "enrich" the given pair of titles: making titles that refer to the same physical entity more similar, and those which do not, much less similar. A notion of similarity is then measured using the second component, where the tokens from the two titles are modelled as vertices of a "social" network graph. A "strength of ties" style of clustering algorithm is then applied on this to see whether they form one cohesive "community" (matching titles), or separately clustered communities (mismatching titles). Experimental results confirm the effectiveness of our approach over existing title matching methods across several input domains.

A protein-protein interaction (PPI) network is a biomolecule relationship network that plays an important role in biological activities. Studies of functional modules in a PPI network contribute greatly to the understanding of biological mechanism. With the development of life science and computing science, a great amount of PPI data has been acquired by various experimental and computational approaches, which presents a significant challenge of detecting functional modules in a PPI network. To address this challenge, many functional module detecting methods have been developed. In this survey, we first analyze the existing problems in detecting functional modules and discuss the countermeasures in the data preprocess and postprocess. Second, we introduce some special metrics for distance or graph developed in clustering process of proteins. Third, we give a classification system of functional module detecting methods and describe some existing detection methods in each category. Fourth, we list databases in common use and conduct performance comparisons of several typical algorithms by popular measurements. Finally, we present the prospects and references for researchers engaged in analyzing PPI networks.

Discovering functional gene clusters based on gene expression data has been a widely-used method that offers a tremendous opportunity for understanding the functional genomics of a specific disease. Due to its strong power of comprehending and interpreting mass of genes, plenty of studies have been done on detecting and analyzing the gene clusters for various diseases. However, more and more evidence suggest that human diseases are not isolated from each other. Therefore, it's significant and interesting to detect the common functional gene clusters driving the core mechanisms among multiple related diseases. There are mainly two challenges for this task: first, the gene expression from each disease may contain noise; second, the common factors underlying multiple diseases are hard to detect. To address these challenges, we propose a novel deep architecture to discover the mutual functional gene clusters across multiple types of diseases. To demonstrate that the proposed method can discover precise and meaningful gene clusters which are not directly obtainable from traditional methods, we perform extensive experimental studies on both synthetic and real datasets - public gene-expression data of three types of cancers. Experimental results show that the proposed approach is highly effective in discovering the mutual functional gene clusters.

With the rapid development of healthcare industry, the overwhelming amounts of electronic health records (EHRs) have been well documented and shared by healthcare institutions and practitioners. It is important to take advantage of EHR data to develop an effective disease risk management model that not only predicts the progression of the disease, but also provides a candidate list of informative risk factors (RFs) in order to prevent the disease. Although EHRs are valuable sources due to the comprehensive patient information, it is difficult to pinpoint the underlying causes of the disease in order to assess the risk of a patient in developing a target disease. Because of the entangled EHR data, it is also challenging to discriminate between patients suffering from the disease and without the disease for the purpose of selecting RFs that cause the disease. To tackle these challenges, we propose a disease memory (DM) framework which can extract the integrated features by modeling the relationships among RFs and more importantly between RFs and the target disease by establishing a deep graphical model with two types of labels. The variants of DM can model characteristics for patients with disease and without disease respectively via training deep networks with different samples. Experiments on a real bone disease data set show that the proposed framework can successfully predict the bone disease and select the informative RFs that are beneficial and useful to aid clinical decision support. Most of the selected RFs are validated by medical literature and some new RFs will attract interests in medical research. The stable and promising performance on evaluation metrics confirms the effectiveness of our model.

With the ultimate intent of improving the quality of life, identification of human's affective states on the collected electroencephalogram (EEG) has attracted lots of attention recently. In this domain, the existing methods usually use only a few labeled samples to classify affective states consisting of over thousands of features. Therefore, important information may not be well utilized and performance is lowered due to the randomness caused by the small sample problem. However, this issue has rarely been discussed in the previous studies. Besides, many EEG channels are irrelevant to the specific learning tasks, which introduce lots of noise to the systems and further lower the performance in the recognition of affective states. To address these two challenges, in this paper, we propose a novel Deep Belief Networks (DBN) based model for affective state recognition from EEG signals. Specifically, signals from each EEG channel are firstly processed with a DBN for effectively extracting critical information from the over thousands of features. The extracted low dimensional characteristics are then utilized in the learning to avoid the small sample problem. For the noisy channel problem, a novel stimulus-response model is proposed. The optimal channel set is obtained according to the response rate of each channel. Finally, a supervised Restricted Boltzmann Machine (RBM) is applied on the combined low dimensional characteristics from the optimal EEG channels. To evaluate the performance of the proposed Supervised DBN based Affective State Recognition (SDA) model, we implement it on the Deap Dataset and compare it with five baselines. Extensive experimental results show that the proposed algorithm can successfully handle the aforementioned two challenges and significantly outperform the baselines by 11.5% to 24.4%, which validates the effectiveness of the proposed algorithm in the task of affective state recognition.

Critical node detection in dynamic networks is of great value in many areas, such as the evolving of friendship in social networks, the development of epidemics, molecular pathogenesis of diseases and so on. As for detecting critical nodes in dynamic Protein-Protein Interaction Networks (PPINs), there are mainly two challenges: the first is to construct the dynamic PPINs that are not available directly from biological experiments in laboratories; and the second is how to identify the most critical units that are responsible for the dynamic processes. This paper proposes effective framework to tackle these two problems. First of all, this paper proposes to construct the dynamic PPINs by simultaneously modeling the activity of proteins and assembling the dynamic co-regulation protein network at each time point. As result, more comprehensive dynamic PPINs are built. Besides, a novel critical protein detection method that integrates multiple PPI networks into a Deep Belief Network model (referred to as MIDBN) is developed. The integrated model is trained to get hierarchical common representations of multiple sources which are used to reconstruct the original data. The variabilities of the reconstruction errors across the time courses are ranked to finally get the top proteins that have significantly different evolving structural patterns than the other nodes in the dynamic networks. We evaluated our network construction method by comparing the functional representations of the derived networks with that of two other traditional construction methods, and our method achieved superior function analysis results. The ranking results of critical proteins from MIDBN were compared with results from two baseline methods and the comparison results showed that MIDBN had better reconstruction rate and identified more proteins of critical value to yeast cell cycle process.

Transfer learning has benefitted many real-world applications where labeled data are abundant in source domains but scarce on the target domain. As there are usually multiple relevant domains where knowledge can be transferred, Multiple Source Transfer Learning (MSTL) has recently attracted much attention. Most existing MSTL methods work in an offline fashion in that they have to store all the data on the target domain before learning. However, in some time-critical applications where the data arrive sequentially in large volume, a fast and scalable online method that can transfer knowledge from multiple source domains is much needed. To achieve this end, in this paper, we propose a new framework of Online Multiple Source Transfer Learning (OMS-TL). The framework is based on a convex optimization problem where knowledge transferred from multiple source domains are guided by the information on the target domain. The proposed method is fast, scalable and enjoys the theoretical guarantees of standard online algorithms. Extensive experiments are conducted on three real-life data sets. The results show that the performance of OMS-TL is close to that of its offline counterpart, which bears comparable performance to existing baseline methods. Furthermore, the proposed method has great scalability and fast response time.

Community formation analysis of dynamic networks has been a hot topic in data mining which has attracted much attention. Recently, there are many studies which focus on discovering communities successively from each snapshot by considering both current and historical information. However, the detected communities are isolated at a certain snapshot, because these approaches ignore important historical or successive information. Different from previous studies which focus on community detection in dynamic networks, we define a new problem of tracking the progression of the community strength - a novel measure that reflects the community robustness and coherence throughout the entire observation period. The proposed community strength analysis provides significant insights into entity properties and relationships in a wide variety of applications. To tackle this problem, we propose a novel two-stage framework: we first identify communities via non-negative matrix factorization, and then calculate the strength of each detected community corresponding to each specific snapshot by solving an optimization problem. Experimental results show that the proposed approach is highly effective in discovering the progression of community strengths and detecting interesting communities.

Dynamic networks are structures with objects and links between the objects that vary in time. Temporal information in dynamic networks can be used to reveal many important phenomena such as bursts of activities in social networks and human communication patterns in email networks. In this area, one very important problem is to understand dynamic patterns of object roles. For instance, will a user become a peripheral node in a social network? Could a website become a hub on the Internet? Will a gene be highly expressed in gene-gene interaction networks in the later stage of a cancer? In this paper, we propose a novel approach that identifies the role of each object, tracks the changes of object roles over time, and predicts the evolving patterns of the object roles in dynamic networks. In particular, a probability model is proposed to extract latent features of object roles from dynamic networks. The extracted latent features are discriminative in learning object roles and are capable of characterizing network structures. The probability model is then extended to learn the dynamic patterns and make predictions on object roles. We assess our method on two data sets on the tasks of exploring how users' importance and political interests evolve as time progresses on dynamic networks. Overall, the extensive experimental evaluations confirm the effectiveness of our approach for identifying, analyzing and predicting object roles on dynamic networks.

In recent years, information trustworthiness has become a serious issue when user-generated contents prevail in our information world. In this paper, we investigate the important problem of estimating information trustworthiness from the perspective of correlating and comparing multiple data sources. To a certain extent, the consistency degree is an indicator of information reliability--Information unanimously agreed by all the sources is more likely to be reliable. Based on this principle, we develop an effective computational approach to identify consistent information from multiple data sources. Particularly, we analyze vast amounts of information collected from multiple review platforms (multiple sources) in which people can rate and review the items they have purchased. The major challenge is that different platforms attract diverse sets of users, and thus information cannot be compared directly at the surface. However, latent reasons hidden in user ratings are mostly shared by multiple sources, and thus inconsistency about an item only appears when some source provides ratings deviating from the common latent reasons. Therefore, we propose a novel two-step procedure to calculate information consistency degrees for a set of items which are rated by multiple sets of users on different platforms. We first build a Multi-Source Deep Belief Network (MSDBN) to identify the common reasons hidden in multi-source rating data, and then calculate a consistency score for each item by comparing individual sources with the reconstructed data derived from the latent reasons. We conduct experiments on real user ratings collected from Orbitz, Priceline and TripAdvisor on all the hotels in Las Vegas and New York City. Experimental results demonstrate that the proposed approach successfully finds the hotels that receive inconsistent, and possibly unreliable, ratings.

Transfer learning has benefited many real-world applications where labeled data are abundant in source domains but scarce in the target domain. As there are usually multiple relevant domains where knowledge can be transferred, multiple source transfer learning (MSTL) has recently attracted much attention. However, we are facing two major challenges when applying MSTL. First, without knowledge about the difference between source and target domains, negative transfer occurs when knowledge is transferred from highly irrelevant sources. Second, existence of imbalanced distributions in classes, where examples in one class dominate, can lead to improper judgement on the source domains’ relevance to the target task. Since existing MSTL methods are usually designed to transfer from relevant sources with balanced distributions, they will fail in applications where these two challenges persist. In this paper, we propose a novel two-phase framework to effectively transfer knowledge from multiple sources even when there exist irrelevant sources and imbalanced class distributions. First, an effective Supervised Local Weight (SLW) scheme is proposed to assign a proper weight to each source domain's classifier based on its ability of predicting accurately on each local region of the target domain. The second phase then learns a classifier for the target domain by solving an optimization problem which concerns both training error minimization and consistency with weighted predictions gained from source domains. A theoretical analysis shows that as the number of source domains increases, the probability that the proposed approach has an error greater than a bound is becoming exponentially small. Extensive experiments on disease prediction, spam filtering and intrusion detection data sets demonstrate the significant improvement in classification performance gained by the proposed method over existing MSTL approaches.

Lots of recent bioinformatics works have focused on the inference of various types of biological networks, such as gene coexpression networks, protein-protein interaction networks, signal transduction networks, etc. Unfortunately, these raw biological network data often contain much noise, especially the false positive predictions which in many cases hinder accurate reconstruction of biological networks. In addition, since the labeled data is scarce and expensive, we hope that the knowledge from other domains can help handle this lack of labeled data problem. In order to construct a more robust and reliable biological network, we propose a novel link propagation based algorithm to de-noise false positives from the target biological network through propagating information from few labeled samples and a set of auxiliary domain networks. While comparing with many current state-of-the-art algorithms, our proposed approach has shown good performance in de-noising biological network.

Informative Gene Selection is the process of identifying relevant genes that are significantly and differentially expressed in biological procedures. The microarray experiments conducted for this purpose usually implement only less than a hundred of samples to rank the relevance of over thousands of genes. Many irrelevant genes thus may gain statistical importance due to the randomness caused by the small sample problem, while relevant genes may lose focus in the same way. Overcoming such a problem goes beyond what a single microarray dataset can offer and stresses the use of multiple experiment results, which is defined as rank aggregation. In this paper, we propose a novel link prediction based rank aggregation algorithm for the purpose of informative gene selection. Each rank is transferred into a fully connected and weighted network, in which the nodes represent genes and the weights of links stand for priorities between connected nodes (genes). The integration of multiple gene ranks is then formulated as an optimization problem of link prediction on multiple networks, with criterion function favoring the maximization of weighted consensus among each network. We solve the problem through iterative estimation of weights and maximization of consensus among them. In the experimental evaluation, we demonstrate our method on the Prostate Cancer Dataset and compare it with other baseline methods. The results show that our link prediction based rank aggregation method remarkably outperforms all the compared methods, which proves the effectiveness of our framework in finding informative genes from multiple microarray experimental results.

We investigate how to estimate information trustworthiness by considering multiple information sources jointly in a latent matrix space. We particularly focus on user review and recommendation systems, as there are multiple platforms where people can rate items and services that they have purchased, and many potential customers rely on these opinions to make decisions. Information trustworthiness is a serious problem because ratings are generated freely by end-users so that many stammers take advantage of freedom of speech to promote their business or damage reputation of competitors. We propose to simply use customer ratings to estimate each individual source's reliability by exploring correlations among multiple sources. Ratings of items are provided by users of diverse tastes and styles, and thus may appear noisy and conflicting across sources, however, they share some underlying common behavior. Therefore, we can group users based on their opinions, and a source is reliable on an item if its opinions given by latent groups are consistent across platforms. Inspired by this observation, we solve the problem by a two-step model -- a joint matrix factorization procedure followed by reliability score computation. We propose two effective approaches to decompose rating matrices as the products of group membership and group rating matrices, and then compute consistency degrees from group rating matrices as source reliability scores. We conduct experiments on both synthetic data and real user ratings collected from Orbitz, Priceline and Trip Advisor on all the hotels in Las Vegas and New York City. Results show that the proposed method is able to give accurate estimates of source reliability and thus successfully identify inconsistent, conflicting and unreliable information.

Link prediction is an important task in social networks and data mining for understanding the mechanisms by which the social networks form and evolve. In most link prediction researches, it is assumed either a snapshot of the social network or a social network with some missing links is available. Most existing researches therefore approach this problem by exploring the topological structure of the social network using only one source of information. However, in many application domains, in addition to the social network of interest, there are a number of auxiliary information available. In this work, we introduce the pseudo cold start link prediction with multiple sources as the problem of predicting the structure of a social network when only a small subgraph of the social network is known and multiple heterogeneous sources are available. We propose a two-phase supervised method: the first phase generates an efficient feature selection scheme to find the best feature from multiple sources that is used for predicting the structure in the social network. In the second phase, we propose a regularization method to control the risk of over-fitting induced by the first phase. We assess our method empirically over a large data collection obtained from Youtube. The extensive experimental evaluations confirm the effectiveness of our approach.

With the rapid advancement of biology technology, many microarray experiments are conducted towards the same problem of finding informative genes. Therefore, it is important to find a set of informative genes integrating multiple microarray experiments that achieves maximal consensus. Most previous re- searches formulated this problem as a rank aggregation problem. In this paper, we propose a novel Graph-based Consensus Maximization (GCM) model to estimate the conditional probability of each gene being informative, then the genes are ranked by this probability. The estimation of the probabilities is formulated as an optimization problem on a bipartite graph, where the criterion function favors the smoothness of the prediction over the graph and penalizes deviations from the initial input ranked lists from microarray experiments. We solve this problem through iterative propagation of probability estimates among neighboring nodes. In addition, when certain genes have already been identified to be informative, it has never been explored in the literature how to take advantage of such information to improve the consensus result. Our proposed GCM model can be naturally extended to incorporate such information, thus increasing the quality of the predicted result. In the experimental evaluation, we conducted experiments on the five prostate cancer microarray studies. The results showed that our model outperformed other baseline methods in finding informative genes. Furthermore, by adding only one piece of information that some gene is informative, our model yielded a significantly better result. The experimental evaluation demonstrates that the proposed GCM model is effective and superior in finding informative genes from multiple microarray experiments.

This paper proposed a novel method using Bacteria Foraging Optimization(BFO) algorithm to avoid the influence of cluster number on experimental result of clustering PPI networks. The initial position that the bacterium located in was considered to be the cluster center and the positions that the bacterium moved were regarded as the adjacent nodes of cluster center. The algorithm classified the nodes selected in the chemotactic operation into cluster when executing the reproduction and elimination-dispersal operations. The procedure kept on creating new clusters until all the nodes were grouped into the clusters. The simulation result showed that the algorithm not only effectively improved the accuracy of cluster result, but also automatically determined the cluster number.

Bones are firm and solid organs that form the skeleton of the body. Bone consists of living tissues with which bone remodeling occurs asynchronously at various sites and involves resorption by osteoclast, followed by formation of new bone by osteoblast. Although bone is a simple composite of a mineral structure, its structure highly contributes to the strength of the bone. In this paper, we develop a computational framework for microstructural bone dynamics modeling which is capable of quantitative assessment of bone mineral density and bone micro-architecture. Our paper focuses on bone microstructure and remodeling dynamics based on a bone network model. First, we generate a bone network based on our bone model reflecting bone microstructure. Next, we introduce a mathematical model of autocrine and paracrine interactions among osteoblast and osteoclast. It allows us to calculate cell population dynamics and changes in bone mass at multiple sites of bone remodeling. Last, we analyze two bone networks representing healthy bone and bone with osteoporosis with our evaluation measurements. Our study provides an initial framework of bone remodeling simulation that can explain experimental observations in bone biology to explore bone diseases such as osteoporosis.

Protein-protein interactions (PPIs) play fundamental roles in nearly all biological processes and differ based on the composition, affinity and lifetime of the association. A vast amount of PPI data for various organisms is available from MIPS, DIP and other sources. The identification of functional modules in PPI network is of great interest because they often reveal unknown functional ties between proteins and hence predict functions for unknown proteins. In this paper, we propose using functional flow simulation and the topology of the network for the functional module detection and function prediction problem. Our approach is based on the functional influence model that quantifies the influence of a biological component on another. We introduce a flow simulation algorithm to generate a functional profile for each component. In addition, a new clustering method FMD (Functional Module Detection) is designed to associate with functional profiles to detect functional modules. We evaluate the proposed technique on three different yeast networks with MIPS functional categories and compare it with several other existing techniques in terms of precision and recall. Our experiments show that our approach achieves better accuracy than other existing methods.

Identification of a critical location in complex structures is one of the most important issues that affects the quality of safety in human life. Specifically, finding high fracture spots of a bone microstructure in our body is a very important research topic; however, it is still not well understood. In this paper, we study on identifying critical locations in a bone microstructure with our bone network model. In fact, about 25 million people in the United States suffer from osteoporosis, which is a systemic skeletal disease characterized by low bone mass and micro-architectural deterioration of bone tissue leading to enhanced bone fragility and a consequent increase in fracture risk. However, currently available techniques for the diagnosis of osteoporosis and the identification of critical locations of bone microstructure are limited. We create a bone network model based on properties of a bone microstructure and we develop a method, called information propagation, to identify critical locations in a bone network. Our paper focuses on detecting important edges as critical locations for the strength of bone microstructure when there are external forces applied to the bone network. Then, we evaluate results of the method in comparison with existing methods including the weighted betweenness centrality and the weighted bridge coefficient. We conclude with the discussion on advantages and disadvantages among those methods.

New technological advances in large-scale protein-protein interaction (PPI) detection provide researchers a valuable source for elucidating the bimolecular mechanism in the cell. In this paper, we investigate the problem of protein complex detection from noisy protein interaction data, i.e., finding the subsets of proteins that are closely coupled via protein interactions. Many people try to solve the problem by finding dense subgraphs in protein interaction networks with unsupervised methods. Here, we stand from a different point of view, redefining the properties and features for protein complexes and designing a “semi-supervised” method to analyze the problem. First we select some representative topological features and biological features to represent the protein complexes and then utilize the training data to build a multi-layer neural network model and finally detect hidden protein complexes in the protein-protein interaction networks with the obtained model. Experiments show the desirable properties of our proposed algorithm and the effectiveness of our approach.

Knowledge of the association information between the attributes in a data set provides insight into the underlying structure of the data and explains the relationships (independence, synergy, redundancy) between the attributes. Complex models learnt computationally from the data are more interpretable to a human analyst when such interdependencies are known. In this paper, we focus on mining two types of association information among the attributes – correlation information and interaction information which capture multivariate dependencies between the data attributes. Identifying the statistically significant attribute associations is a computationally challenging task – the number of possible associations increases exponentially and many associations contain redundant information when a number of correlated attributes are present. In this paper, we explore efficient data mining methods to discover non-redundant attribute sets that contain significant association information indicating the presence of informative patterns in the data.

BACKGROUND: Protein-protein interactions play a key role in biological processes of proteins within a cell. Recent high-throughput techniques have generated protein-protein interaction data in a genome-scale. A wide range of computational approaches have been applied to interactome network analysis for uncovering functional organizations and pathways. However, they have been challenged because of complex connectivity. It has been investigated that protein interaction networks are typically characterized by intrinsic topological features: high modularity and hub-oriented structure. Elucidating the structural roles of modules and hubs is a critical step in complex interactome network analysis. RESULTS: We propose a novel approach to convert the complex structure of an interactome network into hierarchical ordering of proteins. This algorithm measures functional similarity between proteins based on the path strength model, and reveals a hub-oriented tree structure hidden in the complex network. We score hub confidence and identify functional modules in the tree structure of proteins, retrieved by our algorithm. Our experimental results in the yeast protein interactome network demonstrate that the selected hubs are essential proteins for performing functions. In network topology, they have a role in bridging different functional modules. Furthermore, our approach has high accuracy in identifying functional modules hierarchically distributed. CONCLUSIONS: Decomposing, converting, and synthesizing complex interaction networks are fundamental tasks for modeling their structural behaviors. In this study, we systematically analyzed complex interactome network structures for retrieving functional information. Unlike previous hierarchical clustering methods, this approach dynamically explores the hierarchical structure of proteins in a global view. It is well-applicable to the interactome networks in high-level organisms because of its efficiency and scalability.

Predicting protein function from protein interaction networks has been challenging because of the complexity of functional relationships among proteins. Most previous function prediction methods depend on the neighborhood of or the connected paths to known proteins. However, their accuracy has been limited due to the functional inconsistency of interacting proteins. In this paper, we propose a novel approach for function prediction by identifying frequent patterns of functional associations in a protein interaction network. A set of functions that a protein performs is assigned into the corresponding node as a label. A functional association pattern is then represented as a labeled subgraph. Our frequent labeled subgraph mining algorithm efficiently searches the functional association patterns that occur frequently in the network. It iteratively increases the size of frequent patterns by one node at a time by selective joining, and simplifies the network by a priori pruning. Using the yeast protein interaction network, our algorithm found more than 1400 frequent functional association patterns. The function prediction is performed by matching the subgraph, including the unknown protein, with the frequent patterns analogous to it. By leave-one-out cross validation, we show that our approach has better performance than previous link-based methods in terms of prediction accuracy. The frequent functional association patterns generated in this study might become the foundations of advanced analysis for functional behaviors of proteins in a system level.

Protein interaction networks are significant resources for functional knowledge discovery. However, efficient analysis of the networks has been challenging because of complex connectivity. Protein interaction networks have been characterized by intrinsic features, such as modularity and existence of hubs. The concepts of modules and hubs, extending from specific (local) to general (global), suggest hierarchical structures hidden in the complex networks. Retrieving a protein interaction network into the hierarchical structure is thus a crucial process for better understanding of functional organizations. We present a novel approach for restructuring a protein interaction network to reveal hierarchically organized functional modules and hubs. Our algorithm measures functional similarity between proteins based on the path strength model, and dynamically convert a protein interaction network into a hub-oriented tree structure using the definition of centrality. We identify structural hubs and potential functional modules from the tree structure generated by our algorithm. The experimental results demonstrate that the proteins selected as structural hubs are essential for performing functions. In network topology, they have a role in bridging different modules. Furthermore, our approach has higher accuracy in identifying functional modules than other hierarchical clustering methods.

Biological networks having complex connectivity have been widely studied recently. By characterizing their inherent and structural behaviors in a topological perspective, these studies have attempted to discover hidden knowledge in the systems. However, even though various algorithms with graph-theoretical modeling have provided fundamentals in the network analysis, the availability of practical approaches to efficiently handle the complexity has been limited. In this paper, we present a novel flow-based approach, called flowNet, to efficiently analyze large-sized, complex networks. Our approach is based on the functional influence model that quantifies the influence of a biological component on another. We introduce a dynamic flow simulation algorithm to generate a flow pattern which is a unique characteristic for each component. The set of patterns can be used in identifying functional modules (i.e., clustering). The proposed flow simulation algorithm runs very efficiently in sparse networks. Since our approach uses a weighted network as an input, we also discuss supervised and unsupervised weighting schemes for unweighted biological networks. As experimental results in real applications to the yeast protein interaction network, we demonstrate that our approach outperforms previous graph clustering methods with respect to accuracy.

Gene-gene and gene-environment interactions play important roles in the etiology of complex multi-factorial diseases.With the advancements in genotyping technology, large genetic association studies based on hundreds of thousands of single-nucleotide polymorphisms are a popular option for the study of complex diseases. In this paper we use information theoretic concepts to develop a novel method for detecting statistical gene-gene and gene-environment interactions in complex disease models. We explore the effectiveness of our method with extensive simulations using different gene-gene interaction models and the rheumatoid arthritis dataset from Genetic Analysis Workshop-15. The performance of the method was compared to the well known multi-factor dimensionality reduction (MDR) and generalized MDR (GMDR) methods. We demonstrate that our method is capable of analyzing a diverse range of epidemiological data sets containing evidences for gene-gene interactions.

Predicting function from protein interaction networks has been challenging because of the intricate functional relationships among proteins. Most of the previous function prediction methods depend on the neighborhood of or the connected paths to known proteins, and remain low in accuracy. In this paper, we propose a novel approach for function prediction by detecting frequent patterns of functional associations in a protein interaction network. A set of functions that a protein performs is assigned into the corresponding node as a label. A functional association pattern is then represented as a labeled subgraph. Our FASPAM (frequent functional association pattern mining) algorithm efficiently finds the patterns that occur frequently in the network. It iteratively increases the size of frequent patterns by one node at a time by selective joining, and simplifies the network by a priori pruning. Using the yeast protein interaction network extracted from DIP, the FASPAM algorithm found more than 1,400 frequent patterns. By leave-one-out cross validation, our algorithm predicted functions from the frequent patterns with the accuracy of 86%, which is higher than the results from most previous methods.

This research presents a multi agent model to estimate and predict the spread of Avian Influenza (AI) in various attributes and environments in a given population. AI can be transmitted by air and is a critical hazard to birds, especially chickens and ducks. The virus occurs naturally in birds and is capable of being transmitted from an infected bird to another. It is conceivable that AI could be a major threat to human health if the virus becomes capable of transmission to human beings. The next outbreak of AI could lead to millions of deaths unless a feasible strategy for AI containment can be developed.This paper focuses the flexibility that a multi agent system offers. Agent-based models can closely mimic the situations that exist in real system where several autonomous components may be interacting with each other. The modeling approach offers the advantage of examining the interactions between the agents. This research studies the interactions of three critical factors that characterize AI outbreaks. These properties are quarantine range, incubation period and infection probability. The multi agent model investigates the nature of spreading of AI by incorporating these three properties. We illustrate the potential benefits of multi agent modeling in containing the spreading of AI by presenting how efficiently the virus can be contained. Our work exploits data on the 2008 outbreak of AI in South Korea.

Despite the pervasiveness of networks as models for real world systems ranging from the Internet, the World Wide Web to gene regulation and scientific collaborations, only a limited number of metrics capable of characterizing these systems are available. The existing metrics for characterizing networks have broad specificity and lack the selectivity for many applications. The purpose of this paper is to identify and critically evaluate a metric, termed bridging centrality, which is highly selective for identifying bridges in networks. The properties of bridges are unique compared to the other network metrics. For a diverse range of data sets, we found that networks are highly susceptible to disruption but robust to loss structural integrity upon targeted deletion of bridging nodes. A novel graph clustering approach, termed `bridge cut', utilizing bridging edges as module boundary is also proposed. The modules identified by the bridge cut algorithm are more effective than the other graph clustering methods. Thus, bridging centrality is a network metric with unique properties that may aid in network analysis from element to group level in various areas including systems biology and national security applications.

BACKGROUND: The functional characterization of newly discovered proteins has been a challenge in the post-genomic era. Protein-protein interactions provide insights into the functional analysis because the function of unknown proteins can be postulated on the basis of their interaction evidence with known proteins. The protein-protein interaction data sets have been enriched by high-throughput experimental methods. However, the functional analysis using the interaction data has a limitation in accuracy because of the presence of the false positive data experimentally generated and the interactions that are a lack of functional linkage. RESULTS: Protein-protein interaction data can be integrated with the functional knowledge existing in the Gene Ontology (GO) database. We apply similarity measures to assess the functional similarity between interacting proteins. We present a probabilistic framework for predicting functions of unknown proteins based on the functional similarity. We use the leave-one-out cross validation to compare the performance. The experimental results demonstrate that our algorithm performs better than other competing methods in terms of prediction accuracy. In particular, it handles the high false positive rates of current interaction data well. CONCLUSION: The experimentally determined protein-protein interactions are erroneous to uncover the functional associations among proteins. The performance of function prediction for uncharacterized proteins can be enhanced by the integration of multiple data sources available.

BACKGROUND: Quantitative characterization of the topological characteristics of protein-protein interaction (PPI) networks can enable the elucidation of biological functional modules. Here, we present a novel clustering methodology for PPI networks wherein the biological and topological influence of each protein on other proteins is modeled using the probability distribution that the series of interactions necessary to link a pair of distant proteins in the network occur within a time constant (the occurrence probability). RESULTS: CASCADE selects representative nodes for each cluster and iteratively refines clusters based on a combination of the occurrence probability and graph topology between every protein pair. The CASCADE approach is compared to nine competing approaches. The clusters obtained by each technique are compared for enrichment of biological function. CASCADE generates larger clusters and the clusters identified have p-values for biological function that are approximately 1000-fold better than the other methods on the yeast PPI network dataset. An important strength of CASCADE is that the percentage of proteins that are discarded to create clusters is much lower than the other approaches which have an average discard rate of 45% on the yeast protein-protein interaction network. CONCLUSION: CASCADE is effective at detecting biologically relevant clusters of interactions.

The systematic analysis of protein-protein interactions is a fundamental step for understanding of cellular organization, processes and functions. Functional modules can be identified from the protein interaction networks. However, current unreliable interaction data and complex connectivity of interaction networks have made it challenging. We propose a novel metric, called semantic interactivity, to measure the reliability of protein-protein interactions using gene ontology (GO) annotation data. The protein interaction networks can be converted into a weighted graph representation by assigning the reliability to each edge as a weight. We present an iterative centroid search (ICES) algorithm for optimizing the flow-based modularization method and identifying functional modules in a weighted interaction network. It iteratively performs two procedures: centroid search and flow simulation. Our experimental results show that the accuracy of modules is enhanced during the iteration.

High-throughput techniques for protein-protein interaction detection in a genomic scale have provided us a genomic wide view of molecular interactions of many living organisms. A few approaches were proposed to scrutinize protein-protein interactions of living organisms. By the way, the binary nature of the current protein interaction data sets imposes challenges for effective analysis. Furthermore, their performance was suffered by the intrinsic defect, i.e., high noise level, of high-throughput data. This unpleasantly high false positive rate could lead many devoted researches to erroneous biological conclusions. We propose a novel reliability measurement for protein interactions integrating the similarity in gene ontology and the topological similarity in protein interaction networks. Our metric has been proven to be an effective reliability metric for identifying biologically more reliable interactions through the analyses performed from various view points, e.g., functional homogeneity, subcellular localizational homogeneity, and gene expression correlation, etc.

Recent computational analyses of protein interaction networks have attempted to understand cellular organizations, processes and functions. However, they have encountered difficulties due to unreliable interaction data and the complexity of the networks. In this paper, we propose the integration of protein interaction networks with gene ontology annotations for assessing the reliability of current protein-protein interaction data. The interaction reliability can be used for building weighted protein interaction networks. We apply an information flow-based modularization algorithm to the weighted protein interaction networks. Our experimental results show that the interaction reliability between two proteins is positively correlated to the likelihood of functional and locational associations. We finally demonstrate that our approach identifies accurate modules in the protein interaction networks with high statistical confidence with respect to biological function and cellular localization. Moreover, this algorithm outperforms our previous method (Cho et al., 2006) integrating with genetic co-expressional profiles

Microarray techniques give biologists first peek into the molecular states of living tissues. Previous studies have proven that it is feasible to build sample classifiers using the gene expressional profiles. To build an effective sample classifier, dimension reduction process is necessary since classic pattern recognition algorithms do not work well in high dimensional space. In this paper, we present a novel feature extraction algorithm based on the concept of virtual genes by integrating microarray expression data sets with domain knowledge embedded in gene ontology (GO) annotations. We define semantic similarity to measure the functional associations between two genes using the annotation on each GO term. We then identify the groups of genes, called virtual genes, that potentially interact with each other for a biological function. The correlation in gene expression levels of virtual genes can be used to build a sample classifier. For a colon cancer data set, the integration of microarray expression data with GO annotations significantly improves the accuracy of sample classification by more than 10%.

BACKGROUND: The systematic analysis of protein-protein interactions can enable a better understanding of cellular organization, processes and functions. Functional modules can be identified from the protein interaction networks derived from experimental data sets. However, these analyses are challenging because of the presence of unreliable interactions and the complex connectivity of the network. The integration of protein-protein interactions with the data from other sources can be leveraged for improving the effectiveness of functional module detection algorithms. RESULTS: We have developed novel metrics, called semantic similarity and semantic interactivity, which use Gene Ontology (GO) annotations to measure the reliability of protein-protein interactions. The protein interaction networks can be converted into a weighted graph representation by assigning the reliability values to each interaction as a weight. We presented a flow-based modularization algorithm to efficiently identify overlapping modules in the weighted interaction networks. The experimental results show that the semantic similarity and semantic interactivity of interacting pairs were positively correlated with functional co-occurrence. The effectiveness of the algorithm for identifying modules was evaluated using functional categories from the MIPS database. We demonstrated that our algorithm had higher accuracy compared to other competing approaches. CONCLUSION: The integration of protein interaction networks with GO annotation data and the capability of detecting overlapping modules substantially improve the accuracy of module identification.

Recent computational analyses of protein interaction networks have attempted to understand cellular organizations, processes and functions. Several topology-based clustering methods have been applied to the protein interaction networks for detecting functional modules. However, most of the previous algorithms do not perform well on small-world, scale-free networks. In this paper, we present an efficient approach to identify hierarchical modules in the protein interaction networks. Our algorithm selects a small number of informative proteins from a large network, and transforms the intricate small-world, scale-free network into a simple graph with high modularity. Our results show that this approach remarkably enhances the efficiency. We also demonstrate that it outperforms other previous methods in terms of accuracy

The recent high-throughput bio-techniques have provided us large-scale protein-protein interaction data through systematic identification of physical and genetic interactions among all proteins in an organism. Several previous studies have shown that using protein-protein interaction networks to predict protein function is a big step toward full understanding of the mechanisms of cells. However, the protein-protein interaction data derived from high-throughput experiments are typically very noisy, which presents great challenges to the existing methods. In this paper, we propose a novel common-neighbor-based model and a Bayesian framework to predict protein function on the basis of the small-world property of the protein-protein interaction network. We tested our approach on five data sets from various sources. The experimental results have shown that our approach has a better performance than several representative methods in terms of both precision and recall. In addition, our method is particularly effective to handle the high false-positive and false-negative rates in protein-protein interaction data

MOTIVATION: The size, dimensionality and the limited range of the data values make visualization of single nucleotide polymorphism (SNP) datasets challenging. The purpose of this study is to evaluate the usefulness of 3D VizStruct, a novel multi-dimensional data visualization technique for analyzing patterns in SNP datasets. RESULTS: VizStruct is an interactive visualization technique that reduces multi-dimensional data to two dimensions using the complex-valued harmonics of the discrete Fourier transform (DFT). In the 3D VizStruct extension, the multi-dimensional SNP data vectors are reduced to three dimensions using a combination of the DFT and the Kullback-Leibler divergence. The performance of 3D VizStruct was challenged with several biologically relevant published datasets that included human Chromosome 21, the human lipoprotein lipase (LPL) gene locus and the multi-locus genotypes of coral populations. In every case, the 3D VizStruct mapping provided an intuitive visual description of the key characteristics of the underlying multi-dimensional genotype.

High-throughput methods for detecting protein-protein interactions (PPI) have given researchers an initial global picture of protein interactions on a genomic scale. The huge data sets generated by such experiments pose new challenges in data analysis. Though clustering methods have been successfully applied in many areas in bioinformatics, many clustering algorithms cannot be readily applied on protein interaction data sets. One main problem is that the similarity between two proteins cannot be easily defined. This paper proposes a probabilistic model to define the similarity based on conditional probabilities. We then propose a two-step method for estimating the similarity between two proteins based on protein interaction profile. In the first step, the model is trained with proteins with known annotation. Based on this model, similarities are calculated in the second step. Experiments show that our method improves performance.

A protein is the product of a gene. From the gene expression data, we can find co-expressed genes, which are groups of genes that demonstrate coherent patterns on samples. On the other hand, from the protein interaction data, we can find groups of proteins that frequently interact with each other. If we can conduct a joint mining of both gene expression data and protein interaction data, then we may find the clusters of genes that are co-expressed and also their proteins interact. Such clusters found from the joint mining are interesting and meaningful for at least two reasons. First, both the gene expression data and the protein data are very noisy. The clusters confirmed by both data sets will strongly indicate the correlation/connection among the genes in a cluster. In other words, the clusters found from the joint mining are more reliable. We may thus have the high confidence that the genes in a cluster found as such are regulated by the same mechanism or belong to the same biological process. Second, although highly related, gene expression data and protein interaction data still carry different biological meaning. The coincidence of co-expressed genes and interacting proteins is biologically significant. As indicated in [5], many pathways exhibit two properties: their genes exhibit a similar gene expression profile, and the protein products of the genes often interact.

Nowadays many data mining algorithms focus on clustering methods. There are also a lot of approaches designed for outlier detection. We observe that, in many situations, clusters and outliers are concepts whose meanings are inseparable to each other, especially for those data sets with noise. Thus, it is necessary to treat clusters and outliers as concepts of the same importance in data analysis. In this paper, we present a cluster-outlier iterative detection algorithm, tending to detect the clusters and outliers in another perspective for noisy data sets. In this algorithm, clusters are detected and adjusted according to the intra-relationship within clusters and the inter-relationship between clusters and outliers, and vice versa. The adjustment and modification of the clusters and outliers are performed iteratively until a certain termination condition is reached. This data processing algorithm can be applied in many fields such as pattern recognition, data clustering and signal processing. Experimental results demonstrate the advantages of our approach.

We propose a novel mechanism for routing and bandwidth allocation that exploits the selfish and rational behavior of flows in a network. Our mechanism leads to allocations that simultaneously optimize throughput and fairness criteria. We analyze the performance of our mechanism in terms of the induced Nash equilibrium. We compare the allocations at the Nash equilibrium with throughput-optimal allocations as well as with fairness-optimal allocations. Our mechanism offers a smooth trade-off between these criteria, and allows us to produce allocations that are approximately optimal with respect to both. Our mechanism is also fairly simple and admits an efficient distributed implementation.

Joint mining of multiple data sets can often discover interesting, novel, and reliable patterns which cannot be obtained solely from any single source. For example, in cross-market customer segmentation, a group of customers who behave similarly in multiple markets should be considered as a more coherent and more reliable cluster than clusters found in a single market. As another example, in bioinformatics, by joint mining of gene expression data and protein interaction data, we can find clusters of genes which show coherent expression patterns and also produce interacting proteins. Such clusters may be potential pathways.In this paper, we investigate a novel data mining problem, mining cross-graph quasi-cliques, which is generalized from several interesting applications such as cross-market customer segmentation and joint mining of gene expression data and protein interaction data. We build a general model for mining cross-graph quasi-cliques, show why the complete set of cross-graph quasi-cliques cannot be found by previous data mining methods, and study the complexity of the problem. While the problem is difficult, we develop an efficient algorithm, Crochet, which exploits several interesting and effective techniques and heuristics to efficaciously mine cross-graph quasi-cliques. A systematic performance study is reported on both synthetic and real data sets. We demonstrate some interesting and meaningful cross-graph quasi-cliques in bioinformatics. The experimental results also show that algorithm Crochet is efficient and scalable.

Flexible synchronization models cannot provide a proper way of managing user interactions that change the course of a presentation. In this paper, we present a flexible synchronization model, termed SynchRuler, which allows such user interactions including backward and skip. The synchronization rules, which are based on event-condition-action (ECA) rules, are maintained to handle relationships among streams in SynchRuler. The synchronization rules are manipulated by the receiver-controller-actor (RCA) scheme, where receivers, controllers, and actors are objects to receive events, to check conditions, and to execute actions, respectively. The verification of a multimedia presentation specification is performed with the synchronization model. The correctness of the model and the presentation is controlled with a technique called model checking. Model checker PROMELA/SPIN tool is used for automatic verification of the correctness of LTL (linear temporal logic) formulas.

Effective identification of coexpressed genes and coherent patterns in gene expression data is an important task in bioinformatics research and biomedical applications. Several clustering methods have recently been proposed to identify coexpressed genes that share similar coherent patterns. However, there is no objective standard for groups of coexpressed genes. The interpretation of co-expression heavily depends on domain knowledge. Furthermore, groups of coexpressed genes in gene expression data are often highly connected through a large number of "intermediate" genes. There may be no clear boundaries to separate clusters. Clustering gene expression data also faces the challenges of satisfying biological domain requirements and addressing the high connectivity of the data sets. In this paper, we propose an interactive framework for exploring coherent patterns in gene expression data. A novel coherent pattern index is proposed to give users highly confident indications of the existence of coherent patterns. To derive a coherent pattern index and facilitate clustering, we devise an attraction tree structure that summarizes the coherence information among genes in the data set. We present efficient and scalable algorithms for constructing attraction trees and coherent pattern indices from gene expression data sets. Our experimental results show that our approach is effective in mining gene expression data and is scalable for mining large data sets.

Existing data analysis techniques have difficulty in handling multidimensional data. Multidimensional data has been a challenge for data analysis because of the inherent sparsity of the points. In this paper, we first present a novel data preprocessing technique called shrinking which optimizes the inherent characteristic of distribution of data. This data reorganization concept can be applied in many fields such as pattern recognition, data clustering, and signal processing. Then, as an important application of the data shrinking preprocessing, we propose a shrinking-based approach for multidimensional data analysis which consists of three steps: data shrinking, cluster detection, and cluster evaluation and selection. The process of data shrinking moves data points along the direction of the density gradient, thus generating condensed, widely-separated clusters. Following data shrinking, clusters are detected by finding the connected components of dense cells (and evaluated by their compactness). The data-shrinking and cluster-detection steps are conducted on a sequence of grids with different cell sizes. The clusters detected at these scales are compared by a cluster-wise evaluation measurement, and the best clusters are selected as the final result. The experimental results show that this approach can effectively and efficiently detect clusters in both low and high-dimensional spaces.

Clustering spatial data has various applications. Several clustering algorithms have been proposed to cluster objects in spatial databases. Spatial object distribution has significant effect on the results of clustering. Few of current algorithms consider the distribution of objects while processing clusters. In this paper, we propose an adaptive density-based clustering algorithm, ADBC, which uses a novel adaptive strategy for neighbor selection based on spatial object distribution to improve clustering accuracy. We perform a series of experiments on simulated data sets and real data sets. A comparison with DBSCAN and OPTICS shows the superiority of our new approach.

The ordinary sprite generation techniques focus on camera movement, accurate motion estimation, alignment, and integration. These techniques ignore the resolution of the original images, and the regenerated images from the sprite are likely to have lower resolutions than the original ones. Especially if the scenes have finite depth and zoom-in and zoom-out operations occur, the segments of the scene are captured at different resolutions. The traditional mosaic generation methods either blur the mosaic by integrating lower resolution segments or use unnecessarily large storage for the mosaic. The sprite pyramid (or layered sprite) allows efficient storage of images or video clips of overlapping scenes at different resolutions. Moreover, the images or video frames can be reconstructed from the sprite pyramid at the necessary resolutions

Virtual camera controls (VCC) for digital video enable the viewers to visualize interesting objects from their perspective. VCC also allow us to play the video from different angles. In this sense, VCC do not only support visualization and browsing capabilities but also support playback from different angles. VCC management requires accurate global motion estimation and accurate sprite generation. After motion parameters are detected and the precise sprite is generated, virtual camera controls are used to manipulate the sprite and the original frames to allow interactive spatial browsing and playback that enables panning, tilting, and zooming.

We present a PatternQuest framework to learn the patterns of interest (i.e., the distribution patterns of positive objects) using classification methods and relevance feedback. To improve the performance of multimedia retrieval, our PatternQuest first employs an efficient feature selection method to extract a low-dimensional feature subspace. With the feature selection, PatternQuest can effectively alleviate the curse of dimensionality for learning-based relevance feedback. To discover patterns of interest in the feature subspace effectively, we propose a multiresolution pattern discovery (MPD) approach, which trains an online pattern classification method known as adaptive random forests to filter negative objects, from the neighborhood of the query to the global scope, in a fine to coarse way. With MPD, our PatternQuest method can iteratively capture the patterns of interest with a little training data from the user's feedback. We have carried out extensive experiments on an image database (with 31,438 Corel images) to demonstrate the effectiveness and robustness of our method.

Extensive studies have shown that mining microarray data sets is important in bioinformatics research and biomedical applications. In this paper, we explore a novel type of gene-sample-time microarray data sets, which records the expression levels of various genes under a set of samples during a series of time points. In particular, we propose the mining of coherent gene clusters from such data sets. Each cluster contains a subset of genes and a subset of samples such that the genes are coherent on the samples along the time series. The coherent gene clusters may identify the samples corresponding to some phenotypes (e.g., diseases), and suggest the candidate genes correlated to the phenotypes. We present two efficient algorithms, namely the Sample-Gene Search and the Gene-Sample Search, to mine the complete set of coherent gene clusters. We empirically evaluate the performance of our approaches on both a real microarray data set and synthetic data sets. The test results have shown that our approaches are both efficient and effective to find meaningful coherent gene clusters.

MOTIVATION: DNA arrays permit rapid, large-scale screening for patterns of gene expression and simultaneously yield the expression levels of thousands of genes for samples. The number of samples is usually limited, and such datasets are very sparse in high-dimensional gene space. Furthermore, most of the genes collected may not necessarily be of interest and uncertainty about which genes are relevant makes it difficult to construct an informative gene space. Unsupervised empirical sample pattern discovery and informative genes identification of such sparse high-dimensional datasets present interesting but challenging problems. RESULTS: A new model called empirical sample pattern detection (ESPD) is proposed to delineate pattern quality with informative genes. By integrating statistical metrics, data mining and machine learning techniques, this model dynamically measures and manipulates the relationship between samples and genes while conducting an iterative detection of informative space and the empirical pattern. The performance of the proposed method with various array datasets is illustrated.

MOTIVATION: DNA arrays provide a broad snapshot of the state of the cell by measuring the expression levels of thousands of genes simultaneously. Visualization techniques can enable the exploration and detection of patterns and relationships in a complex data set by presenting the data in a graphical format in which the key characteristics become more apparent. The dimensionality and size of array data sets however present significant challenges to visualization. The purpose of this study is to present an interactive approach for visualizing variations in gene expression profiles and to assess its usefulness for classifying samples. RESULTS: The first Fourier harmonic projection was used to map multi-dimensional gene expression data to two dimensions in an implementation called VizStruct. The visualization method was tested using the differentially expressed genes identified in eight separate gene expression data sets. The samples were classified using the oblique decision tree (OC1) algorithm to provide a procedure for visualization-driven classification. The classifiers were evaluated by the holdout and the cross-validation techniques. The proposed method was found to achieve high accuracy. AVAILABILITY: Detailed mathematical derivation of all mapping properties as well as figures in color can be found as supplementary on the web page http://www.cse.buffalo.edu/DBGROUP/bioinformatics/supplementary/vizstruct. All programs were written in Java and Matlab and software code is available by request from the first author.

DNA microarray technology has now made it possible to simultaneously monitor the expression levels of thousands of genes during important biological processes and across collections of related samples. Elucidating the patterns hidden in gene expression data offers a tremendous opportunity for an enhanced understanding of functional genomics. However, the large number of genes and the complexity of biological networks greatly increases the challenges of comprehending and interpreting the resulting mass of data, which often consists of millions of measurements. A first step toward addressing this challenge is the use of clustering techniques, which is essential in the data mining process to reveal natural structures and identify interesting patterns in the underlying data. Cluster analysis seeks to partition a given data set into groups based on specified features so that the data points within a group are more similar to each other than the points in different groups. A very rich literature on cluster analysis has developed over the past three decades. Many conventional clustering algorithms have been adapted or directly applied to gene expression data, and also new algorithms have recently been proposed specifically aiming at gene expression data. These clustering algorithms have been proven useful for identifying biologically relevant groups of genes and samples. In this paper, we first briefly introduce the concepts of microarray technology and discuss the basic elements of clustering on gene expression data. In particular, we divide cluster analysis for gene expression data into three categories. Then, we present specific challenges pertinent to each clustering category and introduce several representative approaches. We also discuss the problem of cluster validation in three aspects and review various methods to assess the quality and reliability of clustering results. Finally, we conclude this paper and suggest the promising trends in this field.

Clustering the time series gene expression data is an important task in bioinformatics research and biomedical applications. Recently, some clustering methods have been adapted or proposed. However, some concerns still remain, such as the robustness of the mining methods, as well as the quality and the interpretability of the mining results. In this paper, we tackle the problem of effectively clustering time series gene expression data by proposing algorithm DHC, a density-based, hierarchical clustering method. We use a density-based approach to identify the clusters such that the clustering results are of high quality and robustness. Moreover, the mining result is in the form of a density tree, which uncovers the embedded clusters in a data set. The inner-structures, the borders and the outliers of the clusters can be further investigated using the attraction tree, which is an intermediate result of the mining. By these two trees, the internal structure of the data set can be visualized effectively. Our empirical evaluation using some real-world data sets show that the method is effective, robust and scalable. It matches the ground truth provided by bioinformatics experts very well in the sample data sets.

DNA microarray technology is now widely used in basic biomedical research for mRNA expression profiling and are increasingly being used to explore patterns of gene expression in clinical research. Automatically detecting phenotype structures from gene expression profiles can provide deep insight into the nature of many diseases as well as lead in the development of new drugs. While most of the previous studies focus on only mining empirical phenotype structure which the experiment controls, it is also interesting to detect possible hidden phenotype structures underlying gene expression profiles.Since the number of samples is usually limited, such data sets are very sparse in high-dimensional gene space. Furthermore, most of the genes of interest are buried in large amount of noise. Unsupervised phenotype structure discovery of such sparse high-dimensional data sets present interesting but challenging problems. In this paper, we propose the model of simultaneously mining both empirical and hidden phenotype structures from gene expression data. We demonstrate the effectiveness and efficiency of the proposed method on various real-world data sets.

Relevance feedback has been an indispensable component for multimedia retrieval systems. In this paper, we present an adaptive pattern discovery method, which addresses relevance feedback by interactively discovering meaningful patterns of relevant objects. To facilitate pattern discovery, we first present a dynamic feature extraction method, which aims to alleviate the curse of dimensionality by extracting a feature subspace using balanced information gain. In the feature subspace, we train an online pattern classification method called adaptive random forests to classify multimedia objects as relevant or irrelevant. Our adaptive random forests adapts the traditional classification method known as random forests for relevance feedback. It improves the efficiency of pattern discovery by choosing the most-informative samples for online learning. Extensive experiments are carried out on a Corel image set (with 31,438 images) to evaluate the performance of our method as compared against the state-of-the-art approaches.

Over-segmentation could be relieved by adopting a divisive image segmentation model. This also requires the binary classification of whether a segmented region corresponds to a single semantic object. In this paper, we propose a model to address this classification problem, by detecting if a region contains both "background" and "foreground" regions. When "background" and "foreground" both present, the region is considered to have multiple objects, otherwise it corresponds to a single object. We implement the model based on certain image features of the region that effectively tell the difference between "background" and "foreground". Experiments show that our model can effectively perform the classification tasks.

Relevance feedback can effectively improve the performance of content-based multimedia retrieval systems. To be effective, a relevance feedback approach must be able to efficiently capture the user's query concept from a very limited number of training samples. To address this issue, we propose a novel adaptive classification method using random forests, which is a machine learning algorithm with proven good performance on many traditional classification problems. With random forests, our method reduces the relevance feedback to a two-class classification problem and classifies database objects as relevant or irrelevant. From the relevant object set, our approach returns the top k nearest neighbors of the query to the user. Briefly speaking, our relevance feedback method has the following dominant features. First, our method is able to address the multimodal distribution of relevant points, because it trains a nonparametric and nonlinear classifier, i.e., random forests, for relevance feedback. Second, it does not overfit training data because it uses an ensemble of tree classifiers to classify multimedia objects. Experiments on a Corel image set (with 31,438 images) show that our method significantly outperforms the state-of-the-art relevance feedback approaches.

Discovering coherent gene expression patterns in time-series gene expression data is an important task in bioinformatics research and biomedical applications. In this paper, we propose an interactive exploration framework for mining coherent expression patterns in time-series gene expression data. We develop a novel tool, coherent pattern index graph, to give users highly confident indications of the existences of coherent patterns. To derive a coherent pattern index graph, we devise an attraction tree structure to record the genes in the data set and summarize the information needed for the interactive exploration. We present fast and scalable algorithms to construct attraction trees and coherent pattern index graphs from gene expression data sets. We conduct an extensive performance study on some real data sets to verify our design. The experimental results strongly show that our approach is more effective than the state-of-the-art methods in mining real gene expression data, and is scalable in mining large data sets.

Mining microarray gene expression data is an important research topic in bioinformatics with broad applications. While most of the previous studies focus on clustering either genes or samples, it is interesting to ask whether we can partition the complete set of samples into exclusive groups (called phenotypes) and find a set of informative genes that can manifest the phenotype structure. In this paper, we propose a new problem of simultaneously mining phenotypes and informative genes from gene expression data. Some statistics-based metrics are proposed to measure the quality of the mining results. Two interesting algorithms are developed: the heuristic search and the mutual reinforcing adjustment method. We present an extensive performance study on both real-world data sets and synthetic data sets. The mining results from the two proposed methods are clearly better than those from the previous methods. They are ready for the real-world applications. Between the two methods, the mutual reinforcing adjustment method is in general more scalable, more effective and with better quality of the mining results.

We introduce the ClusterTree, a new indexing approach for representing clusters generated by any existing clustering approach. A cluster is decomposed into several subclusters and represented as the union of the subclusters. The subclusters can be further decomposed, which isolates the most related groups within the clusters. A ClusterTree is a hierarchy of clusters and subclusters which incorporates the cluster representation into the index structure to achieve effective and efficient retrieval. Our cluster representation is highly adaptive to any kind of cluster. It is well accepted that most existing indexing techniques degrade rapidly as the dimensions increase. The ClusterTree provides a practical solution to index clustered data sets and supports the retrieval of the nearest-neighbors effectively without having to linearly scan the high-dimensional data set. We also discuss an approach to dynamically reconstruct the ClusterTree when new data is added. We present the detailed analysis of this approach and justify it extensively with experiments.

High-dimensional data representation in which each data item (termed target object) is described by many features, is a necessary component of many applications. For example, in DNA microarrays, each sample (target object) is represented by thousands of genes as features. Pattern discovery of target objects presents interesting but also very challenging problems. The data sets are typically not task-specific, many features are irrelevant or redundant and should be pruned out or filtered for the purpose of classifying target objects to find empirical pattern. Uncertainty about which features are relevant makes it difficult to construct an informative feature space. This paper proposes an iterative strategy for pattern discovery in high-dimensional data sets. In this approach, the iterative process consists of two interactive components: discovering patterns within target objects and pruning irrelevant features. The performance of the proposed method with various real data sets is also illustrated.

A large number of algorithms have been proposed for texture image retrieval and analysis. While each algorithm has its own characteristics and is suitable for indexing some particular categories of images, there is a lacking of an effective strategy to integrate these algorithms to maximize the retrieval performance based on the algorithms available. Our first step in this direction is to build up a texture image retrieval system which adaptively chooses the "right" transform to query texture database. Experiments on the Brodatz texture set show that the adaptive image retrieval system significantly outperforms the commonly used single-algorithm based ones.

Global motion estimation (GME) techniques have been developed and usually applied on video where motion takes place often. Although these methods produce accurate results where frequent motion occurs, they turn out to be inefficient if motion is not so often in the video, as in semi-dynamic video. In this paper, we propose motion sensors that indicate the existence of motion and yield a quick approximation to the motion when motion exists thus removing the computations of the hierarchical evaluation of low-pass filtered images as in iterative descent methods.

Users of image databases often prefer to retrieve relevant images by categories. Unfortunately, images are usually indexed by low-level features like color, texture and shape, which often fail to capture high-level concepts well. To address this issue, relevance feedback has been extensively used to associate low-level image features with highlevel concepts. Among all existing relevance feedback approaches, query movement and feature re-weighting have been proven to be suitable for large-scaled image databases with high dimensional image features. We present a feature re-weighting approach using relevant images as well as irrelevant ones in the relevance feedback. As far as feature re-weighting approaches are concerned, one of their common drawbacks is that the feature re-weighting process is prone to be trapped by suboptimal states. To overcome this problem, we introduce a disturbing factor, which is based on the Fisher criterion, to push the feature weights out of sub-optimum. Experimental results on a large-scaled image database with 31,438 COREL images demonstrate the effectiveness of the presented method.

We present a method to extract coarse boundary features from discrete cosine transform (DCT) compressed blocks. Video sequences usually contain enormous data for video processing. Decompressing data and then processing uncompressed data is computational intensive. Moreover, this may yield erroneous results due to processing unnecessary data. Our goal is to extract coarse boundary features without decompressing original data, and then to eliminate insignificant blocks for decompressing. The extracted features provide information about the smoothness, the boundary visibility, and the boundary structure of a block. We give an application of using coarse features for video object segmentation.

The mosaic generation methods benefit from global motion estimation (GME) methods, which yield almost accurate estimation of motion parameters. However, the generated mosaics are usually more blurred than the original frames due to the image warping stage and errors in motion estimation. The transformed coordinates resulting from GME are generally real numbers whereas images are sampled into integer values. Although GME methods generate proper motion parameters, a slight error in motion estimation may propagate to subsequent mosaic generation steps. We propose a method to generate clearer mosaics from video. The temporal integration of images is performed using the histemporal filter based on the histogram of values within an interval. The initial frame in the video sequence is registered at a higher resolution to generate a high resolution mosaic. Instead of warping of each frame, the frames are warped into the mosaic at intervals. This reduces the blurring in the mosaic.

We present a self-adjustable metadatabase aimed at improving the performance of the relevance feedback module extensively used in content-based image retrieval systems. Our metadatabase provides a mechanism for accumulating the optimized relevance feedback records (which are called metadata records) obtained from previous queries. Each metadata record in the metadatabase includes optimal query, feature weights, and identifiers of relevant and/or irrelevant images, and can be effectively used to guide future queries. With the metadatabase, the relevance feedback module admits a noticeable improvement on its performance for category-based search, especially when the relevant images form multiple classes in the feature space. Experiments on a Corel image set (with 31,438 images) show that our method has at least a 15% improvement on average precision and recall over relevance-feedback-only approaches.

A large number of algorithms have been proposed to retrieve and analyze texture images. While much effort has been made to find algorithms applicable to all textures for superior retrieval performance, less work has been done to adaptively integrate various texture retrieval and analysis algorithms. As no individual texture retrieval algorithm is suited for every texture category, a hybrid scheme would outperform any individual method. In this paper, an adaptive retrieval scheme (ARS) for texture image indexing is proposed to dynamically adapt different transforms to different texture patterns for better retrieval performance. The experiments on the Brodatz texture database show that ARS significantly outperforms any individual transform.

Visualization enables us to find structures, features, patterns and relationship in a dataset by presenting the data in various graphical forms with possible interactions. Recent development of DNA microarray technology can be used to measure the expression levels of thousands of genes simultaneously. It has already had a significant impact on the field of bioinformatics, requiring innovative techniques to efficiently and effectively extract, analysis and visualize these fast growing data. In this paper, we present VizCluster, an interactive visualization approach to cluster analysis, and its application on microarray data. VizCluster combines the merits of both high dimensional scatter-plot and parallel coordinates. Integrated with useful features, it can give a simple, fast, intuitive and yet powerful view of the data set. VizCluster supports three major analyzing modes: cluster/class discovery, class prediction, and class assessment. Its primary applications are the classification of samples on microarray datasets. The experiments are based on gene expression data from a study of multiple sclerosis and leukemia patients.

The effectiveness of content-based image retrieval can be enhanced using heterogeneous features embedded in the images. However, since the features in texture, color, and shape are generated using different computation methods and thus may require different similarity measurements, the integration of the retrievals on heterogeneous features is a nontrivial task. We present a semantics-based clustering and indexing approach, termed SemQuery, to support visual queries on heterogeneous features of images. Using this approach, the database images are classified based on their heterogeneous features. Each semantic image cluster contains a set of subclusters that are represented by the heterogeneous features that the images contain. An image is included in a semantic cluster if it falls within the scope of all the heterogeneous clusters of the semantic cluster. We also design a neural network model to merge the results of basic queries on individual features. A query processing strategy is then presented to support visual queries on heterogeneous features. An experimental analysis is conducted and presented to demonstrate the effectiveness and efficiency of the proposed approach.

We present a new flow and congestion control scheme, PLUS (Probe-Loss Utilization Streaming protocol), for distributed multimedia presentation systems. This scheme utilizes probing of the network situation and an effective adjustment mechanism to data loss to support multimedia presentations. The proposed scheme is also designed to scale with increasing number of PLUS-based streaming traffic and to live in harmony with TCP-based traffic. The novelty of the PLUS protocol is that it utilizes the knowledge of its future bottleneck bandwidth in probing the current network situation. This can be achieved by a priori knowledge of the multimedia data before a presentation is requested by a client. Compression schemes like MPEG introduce dependencies on media units. I frames are needed to successfully decode P and B frames, and P frames are needed to decode B frames. A loss of an I or P frame automatically eliminates dependent media units. Our probing scheme increases the successful transmission of critical I and P packets without the overhead of error-correction-schemes. Probing is done using B-frame packets. The advantage is that we use data packets as probe packets. With the PLUS protocol we address the need to avoid congestion rather than react to it. Experiments demonstrate the effectiveness of the approach in utilizing network resources and decreasing loss ratios.

We present a framework for measuring the complexity of image databases, which characterizes the databases for image retrieval. Motivated from the concept of text corpus perplexity, the complexity of image databases is formulated based on image database statistics and information theory. We propose a quantitative metric which can be used to measure the degree of difficulty to retrieve images based on contents of the database. This metric is independent of queries, hence, it is objective. Experiments on both synthetic and real-world images demonstrate that the proposed measurement is highly effective in quantitatively measuring the contents of image databases for content-based retrieval.

The success of text-based retrieval motivates us to investigate analogous techniques which can support the querying and browsing of image data. However, images differ significantly from text both syntactically and semantically in their mode of representing and expressing information. Thus, the generalization of information retrieval from the text domain to the image domain is non-trivial. This paper presents a framework for information retrieval in the image domain which supports content-based querying and browsing of images. A critical first step to establishing such a framework is to construct a codebook of "keywords" for images which is analogous to the dictionary for text documents. We refer to such "keywords" in the image domain as "keyblocks." In this paper, we first present various approaches to generating a codebook containing keyblocks at different resolutions. Then we present a keyblock-based approach to content-based image retrieval. In this approach, each image is encoded as a set of one-dimensional index codes linked to the keyblocks in the codebook, analogous to considering a text document as a linear list of keywords. Generalizing upon text-based information retrieval methods, we then offer various techniques for image-based information retrieval. By comparing the performance of this approach with conventional techniques using color and texture features, we demonstrate the effectiveness of the keyblock-based approach to content-based image retrieval.

DNA arrays can be used to measure the expression levels of thousands of genes simultaneously. Most research is focusing on interpretation of the meaning of the data. However, the majority of methods are supervised, with less attention having been paid to unsupervised approaches which are important when domain knowledge is incomplete or hard to obtain. In this paper we present a new framework for unsupervised analysis of gene expression data which applies an interrelated two-way clustering approach to the gene expression matrices. The goal of clustering is to find important gene patterns and perform cluster discovery on samples. The advantage of this approach is that we can dynamically use the relationships between the groups of genes and samples while iteratively clustering through both gene-dimension and sample-dimension. We illustrate the method on gene expression data from a study of multiple sclerosis patients. The experiments demonstrate the effectiveness of this approach.

The development of the new video coding standard, MPEG- 4, has triggered many video segmentation algorithms that address the generation of the video object planes (VOPs). The background of a video scene is one kind of VOPs where all other video objects are layered on. In this paper, we propose a method for the generation of the stationary background in a MPEG compressed video sequence. If the objects move frequently and all the components of the background are visible in the video sequence, the background macroblocks can be constructed by using Discrete Cosine Transform (DCT) DC coefficients of the blocks. After the generation of the stationary background, the moving objects can be extracted by taking the difference between the frames and the background.

Keyblock, which is a new framework we proposed for content-based image retrieval, is a generalization of the textbased information retrieval technology in the image domain. In this framework, keyblocks, which are analogous to keywords in text document retrieval, can be constructed by exploiting the method of Vector Quantization (VQ). Then an image can be represented as a list of keyblocks similar to a text document which can be considered as a list of keywords. Based on this image representation, various feature models can be constructed for supporting image retrieval. In this paper, we present a new feature representation model which use the keyblock-keyblock correlation matrix, termed keyblock-thesaurus, to facilitate the image retrieval. The feature vectors of this new model incorporate the effect of correlation between keyblocks, thus being more effective in representing image content.

The gap between the synchronization specification and the synchronization model limits user interactions for a multimedia presentation. The middle-tier for multimedia synchronization handles the synchronization rules that are directly extracted from the specification. In addition to these rules, the middle-tier also manages implicit synchronization rules which are not specified but can be extracted from other rules. The synchronization rules generated by the middle-tier assists the synchronization model to provide user interactions while keeping the synchronization specification minimal. We give examples of how these rules are generated from SMIL expressions.

A heterogeneous distributed database environment integrates a set of autonomous database systems to provide global database functions. A flexible transaction approach has been proposed for the heterogeneous distributed database environments. In such an environment, flexible transactions can increase the failure resilience of global transactions by allowing alternate (but in some sense equivalent) executions to be attempted when a local database system fails or some subtransactions of the global transaction abort. We study the impact of compensation, retry, and switching to alternative executions on global concurrency control for the execution of flexible transactions. We propose a new concurrency control criterion for the execution of flexible and local transactions, termed F-serializability, in the error-prone heterogeneous distributed database environments. We then present a scheduling protocol that ensures F-serializability on global schedules. We also demonstrate that this scheduler avoids unnecessary aborts and compensation.

We propose a new framework termed Keyblock for content-based image retrieval, which is a generalization of the text-based information retrieval technology in the image domain. In this framework, methods for extracting comprehensive image features are provided, which are based on the frequency of representative blocks, termed keyblocks, of the image database. Keyblocks, which are analogous to index terms in text document retrieval, can be constructed by exploiting the vector quantization (VQ) method which has been used for image compression. By comparing the performance of our approach with the existing techniques using color feature and wavelet texture feature, the experimental results demonstrate the effectiveness of the framework in image retrieval.

Image databases contain data with high dimensions. Finding interesting patterns in these databases poses a very challenging problem because of the scalability, lack of domain knowledge and complex structures of the embedded clusters. High dimensionality adds severely to the scalability problem. In this paper, we introduce WaveCluster/sup +/, a novel approach to apply wavelet-based techniques for clustering high-dimensional data. Using a hash-based data structure to represent the data set, we offer a detailed technique to apply a wavelet transform on the hashed feature space. We demonstrate that the cost of clustering can be reduced dramatically yet maintaining all the advantages of wavelet-based clustering. This hash-based data representation can be applied for any grid-based clustering approaches. The experimental results show the effectiveness and efficiency of our method on high-dimensional data sets.

Article Free Access Share on ACQ: an automatic clustering and querying approach for large image databases Authors: Dantong Yu Department of Computer Science and Engineering, State University of New York at Buffalo, Buffalo, NY Department of Computer Science and Engineering, State University of New York at Buffalo, Buffalo, NYView Profile , Aidong Zhang Department of Computer Science and Engineering, State University of New York at Buffalo, Buffalo, NY Department of Computer Science and Engineering, State University of New York at Buffalo, Buffalo, NYView Profile Authors Info & Claims MULTIMEDIA '99: Proceedings of the seventh ACM international conference on Multimedia (Part 2)October 1999 Pages 95–98https://doi.org/10.1145/319878.319904Published:01 October 1999Publication History 2citation298DownloadsMetricsTotal Citations2Total Downloads298Last 12 Months5Last 6 weeks0 Get Citation AlertsNew Citation Alert added!This alert has been successfully added and will be sent to:You will be notified whenever a record that you have chosen has been cited.To manage your alert preferences, click on the button below.Manage my AlertsNew Citation Alert!Please log in to your account Save to BinderSave to BinderCreate a New BinderNameCancelCreateExport CitationPublisher SiteeReaderPDF

Advances in multimedia computing offer new approaches to support online access to information from a variety of sources such as video, audio, images, and transparencies. The key to success in providing distributed real time retrieval, is the control of network delay and congestion to keep up the desired quality of service (QoS). We present a new distributed multimedia database environment, called NetMedia. This environment can support controlled delivery of video, audio, and text data, across Internet. NetMedia can be the base of various distributed multimedia applications. As an example we present NetMedia-Virtual-Classroom, an asynchronous distance learning tool, which uses the advantages of the NetMedia framework.

The State University of New York at Buffalo is developing a geographic image retrieval system termed GiView. From a large-scale geographic image database, the system retrieves relevant images that contain parts similar to a given query image. The system extracts texture and color features of geographic images. One key design of the system is a practical image decomposition scheme called nona-tree that uses a hierarchical over-lapping window structure. This treatment offers a compromise to alleviate the difficulty of object recognition which is still an open problem in computer vision. The system also takes the advantage of multi-resolution properties of wavelet transforms so that the system can extract texture features in different scales. This design takes into account the multi-scale nature of geographic images and improves retrieval efficiency. A third key design of the system is to cluster database images based on their similarity to a set of geographic image templates. The templates are selected according to needs of particular projects, and multiple cluster approaches are implemented to accommodate different needs.

Distributed multimedia presentation systems require flexibility and good quality of service for multimedia data presentations. To ensure flexible and satisfiable presentations of multimedia data, collaborations between servers, network, and clients must be carefully designed to retrieve the data from the disk and transfer the data to the client. In this paper we investigate a multi-level buffering and feedback scheme to guarantee the on-time retrieval and transmission of multimedia data in the distributed multimedia presentation environments. In the multi-level buffering scheme, both client and server buffers are designed and collaborated to support the end-to-end multimedia data transmission. These buffers can absorb the rate variations of media streams in different locations. Local and remote feedback messages are used to dynamically prevent buffer overflow/underflow problems. Experiments demonstrate that highly smooth and synchronized presentations of multimedia streams can be supported.

Different visual databases have been designed in various locations. The global integration of such databases can enable users to access data across the world in a transparent manner. In this paper, we investigate an approach to the design and creation of an integrated information system which supports global visual query access to various visual databases over the Internet. Specifically, a metaserver, including a hierarchical metadatabase, a metasearch agent and a query manager, is designed to support such an integration. The metadatabase houses abstracted data about individual remote visual databases. To support visual content-based queries, the abstracted data in the metadatabase reflect the semantics of each visual database. The query manager extracts the feature contents from the queries. The metasearch agent processes the queries by matching their feature contents with the metadata. A list of relevant database sites is derived for efficient retrieval of the query in the selected databases. The performance of the system is refined based on the user's feedback. The proposed system is implemented using Java in a Web-based environment.

The effectiveness of the content-based image retrieval can be enhanced using the heterogeneous features embedded in the images. However, since the features in texture, color, and shape are generated using different computation methods and thus may require different similarity measurements, the integration of the...

With the advances in multimedia databases and the popularization of the Internet, it is now possible to access large image and video repositories distributed throughout the world. One of the challenging problems in such access is how the information in the respective databases can be summarized to enable an intelligent selection of relevant database sites based on visual queries. This paper presents an approach to solve this problem based on image content-based indexing of a metadatabase at a query distribution server. The metadatabase records a summary of the visual content of the images in each database through image templates and statistical features characterizing the similarity distributions of the images. The selection of the databases is done by searching the metadatabase using a ranking algorithm that uses the query's similarity to a template and the features of the databases associated with the template. Two selection approaches, termed mean-based and histogram-based approaches, are presented. The database selection mechanisms have been implemented in a metaserver, and extensive experiments have been performed to demonstrate the effectiveness of the database selection approaches.

Distributed multimedia systems require good quality of service for real-time data transmission. Ethernets, which use a non-deterministic protocol, cannot guarantee the requested bandwidth and there is a large variance in the data rate. A novel approach to supporting multimedia applications over Ethernets, where constant data rates and bandwidth requirements are guaranteed, is proposed. Unlike known solutions, the authors' approach does not require any changes either to the existing software or the Ethernet hardware.

made available at other locations for purposes ranting from diagnostics, distance learning, to electronic c;rnI merce.Accessing these repositories in a distributed setting over either a proprietary or a public network poses a number of challenges.With the advances in multimedia databases and the popularization of the Internet, it is now possible to access large image and video repositories distributed throughout the world.One of the challenging problems in such an access is how the information in the respective databases can be summarized to enable an intelligent selection of relevant database sites based on visual queries.This paper presents an approach to this problem that is based on image content-based indexing of a metadatabase at a query distribution server.The metadatabase records a summary of the visual content of the images per database through image templates and statistical features characterizing the similarity value distributions of images.The selection of databases is done by searching the metadatabase using a histogram-based ranking algorithm that uses query similarity to a template and the features of databases associated with the template.The database selection mechanism has been implemented as a metaserver and extensive experiments have been done to demonstrate the effectiveness of the database ranking algorithm.

Efficient means of content based image indexing is an important problem in large visual information systems. Research in this area is still in its preliminary stages. We present a new method for content based image query from image databases. Our retrieval approach is based on fractal dimension estimation of image features which are generated using various image data transformation techniques. Results are presented for image query using the wavelet transform as the image transform and alternating sequential filters (ASF) for the fractal dimension estimation.

Develops a formal basis for research into the reliability aspects of transaction processing in multidatabase systems (MDBSs). We define a new correctness notion called 'global committability' for the correct unilateral commit and the retry recovery of global transactions in an autonomous MDBS environment. This notion makes it easier to ensure the isolation property of global transactions when the retry approach is applied. The formalization work illustrates that the conventional serializability and recoverability notions are not sufficient to specify the correct execution (i.e. isolated execution and recovery) of global transactions when the unilateral commit and the retry recovery are used to ensure the atomicity of global transactions. This work is significant because the unilateral commit and the retry recovery are an attractive complementary means to the undo recovery (whose correct schedule is specified by the conventional recoverability notion) for advanced transaction applications with the characteristics of site autonomy and long-lived execution.

No abstract available.

Global transaction management requires cooperation from local sites to ensure the consistent and reliable execution of global transactions in a distributed database system. In a heterogeneous distributed database (or multidatabase) environment, various local sites make conflicting assertions of autonomy over the execution of global transactions. A flexible transaction model for the specification of global transactions makes it possible to deal robustly with these conflicting requirements. This paper presents an approach that preserves the semi-atomicity (a weaker form of atomicity) of flexible transactions, allowing local sites to autonomously maintain serializability and recoverability. We offer a fundamental characterization of the flexible transaction model and precisely define the semi-atomicity. We investigate the commit dependencies among the subtransactions of a flexible transaction. These dependencies are used to control the commitment order of the subtransactions. We next identify those restrictions that must be placed upon a flexible transaction to ensure the maintenance of its semi-atomicity. As atomicity is a restrictive criterion, semi-atomicity enhances the class of executable global transactions.