Multimodal sentiment analysis aims to accurately identify sentiment orientations by integrating information from multiple modalities such as text, audio, and video. However, a key challenge in multimodal fusion is effectively balancing the sufficiency and necessity of information across modalities. Traditional models often fail to qualify and capture this balance due to the presence of noise and redundant information in multimodal data, leading to suboptimal performance in sentiment analysis. To address this issue, we propose a novel multimodal sentiment analysis method calledUCMIB-PNS, which is guided by information bottleneck and probabilistic causality. The method employs anUncertainCross-ModalInformationBottleneck(UCMIB)module to reduce redundant information within modalities and maximize discriminative information. The UCMIB utilizes codebooks to dynamically record the distributions of samples and employs random sampling to conduct uncertain modeling across different modalities. It integrates uncertainty-aware contrastive learning and KL divergence for dynamic comparison and compression of information from different modalities. Moreover, UCMIB-PNS uses differentiableProbability ofNecessity andSufficiency(PNS)estimators to estimate and re-weight the sufficiency and necessity of modalities by constructing several counterfactual scenarios through end-to-end learning. Experiments conducted on four publicly available multimodal sentiment analysis datasets demonstrate that UCMIB-PNS achieves optimal performance on both clean and noisy data. Extended experiments further validate the method's robustness under different types of noise.

Heterophilic graph neural networks (GNNs) have gained prominence for their ability to learn effective representations in graphs with diverse, attribute-aware relationships. While existing methods leverage attribute inference during message passing to improve performance, they often struggle with challenging heterophilic graphs. This is due to edge distribution shifts introduced by diverse connection patterns, which blur attribute distinctions and undermine message-passing stability. This paper introduces H₂OGNN, a novel framework that reframes edge attribute inference as an out-of-distribution (OOD) detection problem. H₂OGNN introduces a simple yet effective symbolic energy regularization approach for OOD learning, ensuring robust classification boundaries between homophilic and heterophilic edge attributes. This design significantly improves the stability and reliability of GNNs across diverse connectivity patterns. Through theoretical analysis, we show that H₂OGNN addresses the graph denoising problem by going beyond feature smoothing, offering deeper insights into how precise edge attribute identification boosts model performance. Extensive experiments on nine benchmark datasets demonstrate that H₂OGNN not only achieves state-of-the-art performance but also consistently outperforms other heterophilic GNN frameworks, particularly on datasets with high heterophily.

ABSTRACT Lung cancer remains the leading cause of cancer‐related mortality worldwide. Early detection is critical to improving treatment outcomes and survival rates. Positron emission tomography/computed tomography (PET/CT) is a widely used imaging modality for identifying lung tumours. However, limitations in imaging resolution and the complexity of cancer characteristics make detecting small lesions particularly challenging. To address this issue, we propose a novel coarse‐to‐fine detection framework to reduce missed diagnoses of small lung lesions in PET/CT images. Our method integrates a stacked detection structure with a multi‐attention guidance mechanism, effectively leveraging spatial and contextual information from small lesions to enhance lesion localisation. Experimental evaluations on a PET/CT dataset of 225 patients demonstrate the effectiveness of our method, achieving remarkable results with a precision of 81.74%, a recall of 76.64%, and an mAP of 84.72%. The proposed framework not only improves the detection accuracy of small target lesions in the lung but also provides a more reliable solution for early diagnosis.

Accurate segmentation of metastatic lesions is crucial for improving the quality of patient care, particularly in the context of bone scans. However, existing automated methods, which are predominantly data‐driven, exhibit limited performance and lack interpretability. To address these challenges, we propose a novel two‐stage framework that integrates human domain knowledge with data patterns to enhance CNN‐based metastasis lesion segmentation in bone scans. The proposed method comprises two phases: Stage I detects hotspots in bone scans using a CNN‐based model, while Stage II identifies actual metastases by leveraging clinical knowledge of uptake intensity asymmetry. Our approach incorporates a dual‐sampling scheme inspired by diagnostic patterns and an enhanced feature extractor within the hotspot segmentation network, thus augmenting the detection capabilities of traditional data‐driven CNN models. The assessment of symmetrical uptake intensity starts with the symmetry axis of the trunk in the image, followed by a composite similarity measure that considers both geometric symmetry and intensity consistency. Experimental evaluations on 302 clinical cases reveal that our proposed segmentation network improves the Dice similarity coefficient score by 4.34% compared to the baseline method. Furthermore, integrating clinical knowledge significantly reduces false positives, improving the class pixel accuracy score by 2.39% and demonstrating notable adaptability to other segmentation models. Comparative analysis with existing models for metastasis lesion segmentation demonstrates the superior performance of our approach. By incorporating domain knowledge into data patterns, our method enhances automated segmentation performance and bridges the gap between domain expertise and data‐driven methodologies in the automated analysis of low‐resolution bone scans.

Facial expressions in videos inherently mirror the dynamic nature of real-world facial events. Consequently, facial expression recognition (FER) should employ a dynamic graph-based representation to effectively capture the relational structure of facial expressions rather than relying on conventional grid or sequence methods. However, existing graph-based approaches have their limitations. Frame-level graph methods provide a coarse representation of the facial graph across time and space, while landmark-based graph methods need to introduce additional facial landmarks, resulting in a static graph structure. To address these challenges, we propose spatial-temporal relation-aware dynamic graph convolutional networks (ST-RDGCN). This fine-grained relation modeling approach enables the dynamic modeling of evolving facial expressions in videos through dynamic space-time graphs, eliminating the need for facial landmarks. ST-RDGCN encompasses three graph construction paradigms: dynamic independent space graph, dynamic joint space-time graph, and dynamic cross space-time graph. Furthermore, we propose a relation-aware space-time graph convolution (RSTG-Conv) operator to learn informative spatiotemporal correlations in dynamic space-time graphs. In extensive experimental evaluations, our ST-RDGCN demonstrates state-of-the-art performance on the five popular video-based FER datasets, achieving overall accuracy scores of 99.69%, 91.67%, 56.51%, 69.37%, and 49.03% on the CK+, Oulu-CASIA, AFEW, DFEW, and FERV39k datasets, respectively. In particular, our ST-RDGCN outperforms the current best method by 3.6% in UAR on the most challenging FERV39k dataset. Furthermore, our analysis reveals that the dynamic cross space-time graph scheme is the most effective among the three dynamic graph construction schemes.

Accurately identifying associations between human genes (proteins) and clinical phenotypes is critical for advancing drug development and precision medicine. While the human phenotype ontology (HPO) standardizes clinical phenotypes, current computational approaches for predicting human protein-phenotype associations suffer from two limitations: (1) underutilization of multimodal protein-related information and (2) lack of state-of-the-art deep learning representations tailored to diverse data modalities, such as text and sequence. To overcome these limitations, we introduce MultiFusion2HPO, a novel multimodal model that integrates diverse features and advanced learning methods from multiple data sources to enhance the prediction of human protein-HPO associations. MultiFusion2HPO leverages five critical modalities: textual information (TFIDF-D2V and BioLinkBERT embeddings), protein sequence data (InterPro and ESM2), protein-protein interaction (PPI) networks, gene ontology (GO) annotation, and gene expression. Comprehensive experiments on benchmark datasets demonstrate the superiority of MultiFusion2HPO over the state-of-the-art methods, DeepPheno and HPOLabeler. These results underscore the effectiveness of integrating multimodal protein data to improve the accuracy of human protein-HPO association predictions.

Graph anomaly detection (GAD) has attracted increasing interest due to its critical role in diverse real-world applications. Graph neural networks (GNNs) offer a promising avenue for GAD, leveraging their exceptional capacity to model complex graph structures and relationships. However, existing GNN-based models encounter challenges in addressing the GAD's fundamental issue-anomaly camouflage, where anomalies mimic normal instances, leading to indistinguishable features. In this article, we propose a novel approach, termed correlation information enhanced GAD (CIE-GAD). Specifically, drawing on the observation that the distribution of homophilic and heterophilic edges differs between abnormal and normal samples, we construct a hypergraph to learn the co-occurrence relationships among adjacent edges. By enhancing the extraction of sample correlation information, we effectively tackle feature similarity caused by anomaly camouflage, thereby enhancing the performance of GAD. Furthermore, we develop a spectral convolution mechanism based on node-level attention fusion, enabling the capture of multifrequency signals. This module performs adaptive fusion tailored to the unique frequency information requirements of each node, mitigating the local heterophily problem. Extensive experiments on various real-world GAD datasets demonstrate that the proposed CIE-GAD outperforms state-of-the-art methods. Notably, our approach achieves AUC-PR improvements of up to 3.47%, with an average gain of 1.5%, demonstrating its effectiveness in detecting anomalies in graph data.

In UniProtKB, each protein is linked to numerous publications covering topics such as sequence, function, and structure, which are annotated manually or through automated methods. Given the vast number of proteins and literature, manual annotation is time-consuming and labour-intensive. Although UniProtKB offers automated annotations, their quality often falls short. Therefore, developing an accurate automated classifier to identify the topics of publications associated with each protein is imperative for advancing biomedical knowledge discovery. Classifying publications in UniProtKB involves protein-publication pairs characterized by multi-label, label co-occurrence, and class imbalance, which increases complexity. This paper proposes a novel method called PubLabeler, which simultaneously considers protein description and scientific literature texts as input. PubLabeler employs the PubMedBERT model to encode input texts and integrates label co-occurrence information into the model parameters. Additionally, it uses focal loss to update parameters, allowing the model to focus more on classes with a few instances. Using newly annotated literature from Swiss-Prot in 2023 as a test set, PubLabeler achieved superior results in both micro and macro metrics, showing a 28.5% improvement in macro-F1 compared to UniProtKB's automated annotation method, UPCLASS. Furthermore, we validated PubLabeler's effectiveness in TrEMBL annotation, showcasing its comprehensive prediction results compared to TrEMBL's automated annotations. These findings highlight PubLabeler's reliability and potential to advance protein-related information extraction and knowledge discovery.

Multimodal sentiment analysis leverages information from multiple sensors to achieve a comprehensive interpretation of emotions. However, different modalities do not always boost each other as expected. They compete with each other, leading to some modalities being under-optimized during the training process. To address this issue, we propose Adaptive Gradient Scaling with Sparse Mixture-of-Experts (AGS-SMoE) . We first discuss the issue of modal preemption in unified multimodal learning from the perspective of causal preemption. Driven by actual cause, we use the gradient norms from different encoders at two fusion stages as evidence, estimating the current modal preemption state using a parameter-free method. Then, based on the dynamic preemption factor, we design a gradient scaling method to balance optimization for different encoders. Furthermore, we use Mixture-of-Experts to sparsify and perceive multimodal tokens in different preemption states. As a result, our experiments on four multimodal sentiment analysis datasets have achieved state-of-the-art results. Moreover, our method improves modal representation learning at different stages. Extensive experiments confirm that our method can alleviate the modal preemption problem in a plug-and-play manner. Our code is available at https://github.com/TheShy-Dream/AGS-SMoE .

The capability of cyber-physical power system (CPPS) to recover from cascading failures caused by extreme events and restore prefailure functionality is a critical focus in resilience research. In contrast to the strongly coupled systems studied by most researchers, this article examines weakly coupled CPPS, exploring result-oriented recovery approaches to enhance system resilience. Various repair methods are compared in terms of the resilience of weakly connected CPPS across different coupling modes and probabilities of failover. Utilizing the Q-learning algorithm, an optimized sequence for network restoration is obtained to minimize the negative influence of failures on network functionality while reducing power loss. The proposed method's effectiveness and generalizability have been comprehensively verified through simulation experiments by establishing weakly coupled CPPS for the IEEE 39, IEEE 118, and IEEE 300 networks and their corresponding scale-free networks. Its rationality was verified through two recovery mechanisms: single-node recovery and multinode recovery. By comparing the proposed method with heuristic recovery methods and optimization-based recovery methods, we found that it can significantly accelerate network recovery, and improve network resilience, achieving better resilience centrality. These findings provide valuable insights for decision making in CPPS recovery work.

Biomedical Named Entity Recognition (NER) is a crucial task in extracting information from biomedical texts. However, the diversity of professional terminology, semantic complexity, and the widespread presence of synonyms pose significant challenges. Traditional methods that rely on sequence labeling training datasets often struggle to handle these complexities. To address this, we introduce a novel framework for BioNER, termed DMNER, which leverages external knowledge and operates in two steps: entity boundary detection and entity category identification through matching. The core of DMNER is its second step, which determines the entity category by retrieving similar entities and their categories from a knowledge dictionary using semantic similarity matching. Our experiments on 10 biomedical datasets demonstrate that DMNER outperforms baselines across these tasks, proving its effectiveness and adaptability. DMNER is versatile and can be applied to various NER tasks, including supervised NER, distantly supervised NER, and NER on multiple datasets with disjoint label sets. The DMNER code is publicly available1.

The prevalent solution for BioNER involves using representation learning techniques combined with sequence labeling. However, such methods are inherently task-specific, demonstrate poor generalizability, and often require a dedicated model for each dataset. To leverage the versatile capabilities of recent large language models (LLMs), several approaches have explored generative techniques for entity extraction. Yet, these approaches often fall short compared to previous sequence labeling approaches. In this paper, we utilize the open-sourced LLM LLaMA2 as the backbone model, and design specific instructions to distinguish between different types of entities and datasets. By combining the LLM’s understanding of instructions with sequence labeling techniques, we train a model using a mix of datasets capable of extracting various types of entities. Given that the backbone LLMs lacks specialized medical knowledge, we also integrate external entity knowledge bases and employ instruction tuning to enable the model to densely recognize curated entities. Our parameter-efficient training model, VANER, significantly outperforms previous LLMs-based models. For the first time, as an LLM-based model, VANER surpasses the majority of conventional state-of-the-art BioNER systems, achieving the highest F1 scores across three datasets.

Real-world data often contain incomplete views with varying degrees of missing information. While there are existing methods for learning representations from such data, effectively utilizing all incomplete view data and ensuring robustness to different levels of completeness remains a challenging task. To address this problem, we propose a novel framework named IMRL-AGI. IMRL-AGI combines the anchor graph-based Graph Convolutional Network (GCN) and information bottleneck. Specifically, the framework starts by constructing an anchor graph to effectively captures the nonlinear information between instances. Next, an anchor graph-based GCN is designed to extract feature information from various views. IMRL-AGI maximizes the mutual information between the views obtained by the common representation and the anchor-graph-based GCN, ensuring the accurate extraction of view information. Furthermore, the minimization of mutual information is applied to promote diversity and reduce redundancy in the multi-view representation. Extensive experiments are conducted on several real-world datasets, and the results demonstrate the superiority of IMRL-AGI.

Message passing (MP) is crucial for effective graph neural networks (GNNs). Most local message-passing schemes have been shown to underperform on heterophily graphs due to the perturbation of updated representations caused by local redundant heterophily information. However, our experiment findings indicate that the distribution of heterophily information during MP can be disrupted by disentangling local neighborhoods. This finding can be applied to other GNNs, improving their performance on heterophily graphs in a more flexible manner compared to most heterophily GNNs with complex designs. This article proposes a new type of simple message-passing neural network called Flow2GNN. It uses a two-way flow message-passing scheme to enhance the ability of GNNs by disentangling and redistributing heterophily information in the topology space and the attribute space. Our proposed message-passing scheme consists of two steps in topology space and attribute space. First, we introduce a new disentangled operator with binary elements that disentangle topology information in-flow and out-flow between connected nodes. Second, we use an adaptive aggregation model that adjusts the flow amount between homophily and heterophily attribute information. Furthermore, we rigorously prove that disentangling in message-passing can reduce the generalization gap, offering a deeper understanding of how our model enhances other GNNs. The extensive experiment results show that the proposed model, Flow2GNN, not only outperforms state-of-the-art GNNs, but also helps improve the performance of other commonly used GNNs on heterophily graphs, including GCN, GAT, GCNII, and H2GCN, specifically for GCN, with up to a 25.88% improvement on the Wisconsin dataset.

Deep learning (DL) based knowledge tracing (KT) models have challenges for uninterpretable prediction and parameter representation in educational applications, though they achieved remarkable outcomes in predicting the exercise performance of students. This paper proposes a novel knowledge tracing model of high precision and interpretability (namedXKT) for questions with multiple knowledge concepts based on cognitive learning theories and multidimensional item response theory (MIRT). TheXKTconsists of three differentiable network components: multi-feature embedding, cognition processing network, andMIRT-based neural predictor, which aim to provide an explainable prediction of student exercise performance. Specifically, inXKT, multi-feature embedding learns the rich semantic representation (e.g., knowledge distribution information) to enhance knowledge tracing using a cognition processing network. The cognition processing network performs selective perception, ability memory processing, and long-term knowledge memory processing to ensure the explainable factor representation for theMIRT-based neural predictor. Lastly, theMIRT-based neural predictor employs psychometric parameters to interpret student exercise predictions better. Extensive experiments on four real-world datasets show thatXKToutperforms existingKTmethods in predicting future learner responses. Moreover, ablation studies further show thatXKToffers good interpretability of student performance predictions with multiple knowledge concepts, indicating excellent potential in real-world educational applications.

Human Phenotype Ontology (HPO)-based approaches have gained popularity in recent times as a tool for genomic diagnostics of rare diseases. However, these approaches do not make full use of the available information on disease and patient phenotypes. We present a new method called Phen2Disease, which utilizes the bidirectional maximum matching semantic similarity between two phenotype sets of patients and diseases to prioritize diseases and genes. Our comprehensive experiments have been conducted on six real data cohorts with 2051 cases (Cohort 1, n = 384; Cohort 2, n = 281; Cohort 3, n = 185; Cohort 4, n = 784; Cohort 5, n = 208; and Cohort 6, n = 209) and two simulated data cohorts with 1000 cases. The results of the experiments showed that Phen2Disease outperforms the three state-of-the-art methods when only phenotype information and HPO knowledge base are used, particularly in cohorts with fewer average numbers of HPO terms. We also observed that patients with higher information content scores have more specific information, leading to more accurate predictions. Moreover, Phen2Disease provides high interpretability with ranked diseases and patient HPO terms presented. Our method provides a novel approach to utilizing phenotype data for genomic diagnostics of rare diseases, with potential for clinical impact. Phen2Disease is freely available on GitHub at https://github.com/ZhuLab-Fudan/Phen2Disease.

A number of machine learning (ML) approaches for drug discovery have been available that rely only on sequential (1D) and planar (2D) information without effectively using the 3D information for generating features of drugs. However, 3D information of small molecules can reflect relative position of atoms more directly, which affects molecular properties. In this work, we present a new deep learning model called Drug3D-DTI for drug-target interaction prediction. Drug3D-DTI takes advantage of molecular spatial information, i.e., atom proximity in three-dimensional (3D) structures. We comprehensively evaluated the performance of Drug3D-DTI on two datasets with two tasks of regression and classification. In particular, we compared Drug3D-DTI with several existing methods including the two cutting-edge methods for compound-protein interaction prediction. From the experimental results, Drug3D-DTI clearly outperformed other methods under all settings. Further, this performance improvement was validated by ablation experiments and a case study. The implementation of Drug3D-DTI is available at (https://github.com/zhiruiliao/Drug3D-DTI).

Grant support (GS) in the MEDLINE database refers to funding agencies and contract numbers. It is important for funding organizations to track their funding outcomes from the GS information. As such, how to accurately and automatically extract funding information from biomedical literature is challenging. In this paper, we present a pipeline system called GrantExtractor that is able to accurately extract GS information from fulltext biomedical literature. GrantExtractor effectively integrates several advanced machine learning techniques. In particular, we use a sentence classifier to identify funding sentences from articles first. A bi-directional LSTM and the CRF layer (BiLSTM-CRF), and pattern matching are then used to extract entities of grant numbers and agencies from these identified funding sentences. After removing noisy numbers by a multi-class model, we finally match each grant number with its corresponding agency. Experimental results on benchmark datasets have demonstrated that GrantExtractor clearly outperforms all baseline methods. It is further evident that GrantExtractor won the first place in Task 5C of 2017 BioASQ challenge, with achieving the Micro-recall of 0.9526 for 22,610 articles. Moreover, GrantExtractor has achieved the Micro F-measure score as high as 0.90 in extracting grant pairs.

Dynamic networks are networks in which nodes and edges come and go at different times. The dynamic nature of this type of networks has imposed a challenge in network visualisation. Many techniques have been proposed to visualise dynamic networks in an effort of not only capturing the dynamic features of networks over the time but also laying out the network in a way that is fast and makes the visualisation aesthetically pleasing. However, empirical research has shown that the current techniques have limitations in one way or another. In the paper, we present an interactive visualisation called DGViewer which gives users control and flexibility allowing them to decide what, when and how to view visualisations to improve the analysis and comprehension of dynamic networks. A preliminary usability study indicates that users are positive with this user-centred visualisation for better user experience and task performance.

MOTIVATION: With the rapidly growing biomedical literature, automatically indexing biomedical articles by Medical Subject Heading (MeSH), namely MeSH indexing, has become increasingly important for facilitating hypothesis generation and knowledge discovery. Over the past years, many large-scale MeSH indexing approaches have been proposed, such as Medical Text Indexer, MeSHLabeler, DeepMeSH and MeSHProbeNet. However, the performance of these methods is hampered by using limited information, i.e. only the title and abstract of biomedical articles. RESULTS: We propose FullMeSH, a large-scale MeSH indexing method taking advantage of the recent increase in the availability of full text articles. Compared to DeepMeSH and other state-of-the-art methods, FullMeSH has three novelties: (i) Instead of using a full text as a whole, FullMeSH segments it into several sections with their normalized titles in order to distinguish their contributions to the overall performance. (ii) FullMeSH integrates the evidence from different sections in a 'learning to rank' framework by combining the sparse and deep semantic representations. (iii) FullMeSH trains an Attention-based Convolutional Neural Network for each section, which achieves better performance on infrequent MeSH headings. FullMeSH has been developed and empirically trained on the entire set of 1.4 million full-text articles in the PubMed Central Open Access subset. It achieved a Micro F-measure of 66.76% on a test set of 10 000 articles, which was 3.3% and 6.4% higher than DeepMeSH and MeSHLabeler, respectively. Furthermore, FullMeSH demonstrated an average improvement of 4.7% over DeepMeSH for indexing Check Tags, a set of most frequently indexed MeSH headings. AVAILABILITY AND IMPLEMENTATION: The software is available upon request. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

MOTIVATION: Annotating human proteins by abnormal phenotypes has become an important topic. Human Phenotype Ontology (HPO) is a standardized vocabulary of phenotypic abnormalities encountered in human diseases. As of November 2019, only <4000 proteins have been annotated with HPO. Thus, a computational approach for accurately predicting protein-HPO associations would be important, whereas no methods have outperformed a simple Naive approach in the second Critical Assessment of Functional Annotation, 2013-2014 (CAFA2). RESULTS: We present HPOLabeler, which is able to use a wide variety of evidence, such as protein-protein interaction (PPI) networks, Gene Ontology, InterPro, trigram frequency and HPO term frequency, in the framework of learning to rank (LTR). LTR has been proved to be powerful for solving large-scale, multi-label ranking problems in bioinformatics. Given an input protein, LTR outputs the ranked list of HPO terms from a series of input scores given to the candidate HPO terms by component learning models (logistic regression, nearest neighbor and a Naive method), which are trained from given multiple evidence. We empirically evaluate HPOLabeler extensively through mainly two experiments of cross validation and temporal validation, for which HPOLabeler significantly outperformed all component models and competing methods including the current state-of-the-art method. We further found that (i) PPI is most informative for prediction among diverse data sources and (ii) low prediction performance of temporal validation might be caused by incomplete annotation of new proteins. AVAILABILITY AND IMPLEMENTATION: http://issubmission.sjtu.edu.cn/hpolabeler/. CONTACT: [email protected]. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Functional imaging has successfully been applied to capture functional changes in the pathological tissues of a body in recent years. Nuclear medicine functional imaging has been used to acquire information about areas of concerns (e.g. lesions and organs) in a non‐invasive manner, enabling semi‐automated or automated decision‐making for disease diagnosis, treatment, evaluation, and prediction. Focusing on functional nuclear medicine images, in this study, the authors review existing work on the classification of single‐photon emission computed tomography, positron emission tomography, and their hybrid modalities with computed tomography and magnetic resonance imaging images by using convolutional neural network (CNN) techniques. Specifically, they first present an overview of nuclear imaging and the CNN technique, such as nuclear imaging modalities, nuclear image data format, CNN architecture, and the main CNN classification models. According to the diseases of concern, they then classify the existing CNN‐based work on the classification of functional nuclear images into three different categories. For the typical work in each of these categories, they present details about their research objectives, adopted CNN models, and achieved main results. Finally, they discuss research challenges and directions for developing technological solutions to classify nuclear medicine images based on the CNN technique.

The prediction of precise protein-ligand binding activities can accelerate drug discovery by virtual screening-a computational technique that predicts whether a small molecule ligand is able to bind to a specific target. Thus, it is crucial to improve the performance of virtual screening. However, previous models for solving this problem are either ligand-based or structure-based. In this paper, we propose a universal deep neural network model called DeepDock that predicts protein-ligand interaction by using both ligand and structure information. Using the combination of two types of information, our model consists of embedding, convolution, max pooling, and fully-connected layers. In particular, different types of inputs are concatenated before being fed into the fully-connected layers. In the experiments, we compare our approach to the competing methods against two benchmark datasets under different settings. The experiment results have demonstrated that DeepDock can improve predictive performance by more than 4% on both DUD-E and MUV datasets in terms of AUPR.

Graph visualization finds wide applications in different areas. As the popularity of social network sites is increasing, it becomes particularly useful in visual analysis of these sites. A number of algorithms for graph visualization have been developed over the past decades. The issue on how to design and develop algorithms by taking into account the characteristics of real graphs such as scale-free and hierarchical structures, however, has not been well addressed. In this paper, we extend the concept of a node degree to a node global degree for a node in a graph, and present an algorithm that computes their scores of all nodes. By taking advantage of the common structure features of real networks, two scalable extensions of this algorithm are further provided that are able to approximate computation results. Based on node global degrees, a filtering approach is presented to reduce the visual complexity of a layout. Extensive experiments have demonstrated the performance of the proposed algorithms in terms of two common evaluation metrics, as well as visualization results. In addition, we have implemented the algorithms in a prototype system, which enable users to explore a graph at continuous levels of details in real time, as evidenced by several real examples.

Given a query photo characterizing a location-aware landmark shot by a user, landmark retrieval is about returning a set of photos ranked in their similarities to the query. Existing studies on landmark retrieval focus on conducting a matching process between candidate photos and a query photo by exploiting location-aware visual features. Notwithstanding the good results achieved, these approaches are based on an assumption that a landmark of interest is well-captured and distinctive enough to be distinguished from others. In fact, distinctive landmarks may be badly selected, e.g. changes on viewpoints or angles. This will discourage the recognition results if a biased query photo is issued. In this paper, we present a novel technique that exploits user communities in social media networks. Given a biased query photo containing some landmarks taken by a user, we select multiple users to complement this user for retrieval. Multiple photos are then used to enrich the query photo, constituting a more representative yet robust multi-query set. A pattern mining method is developed to obtain a compact feature representation of photos from the multi-query set. Such a representation is utilized to efficiently query the database so as to improve retrieval results. Extensive experiments on real-world datasets demonstrate the effectiveness and efficiency of our approach.

More often than not, a multimedia data described by multiple features, such as color and shape features, can be naturally decomposed of multi-views. Since multi-views provide complementary information to each other, great endeavors have been dedicated by leveraging multiple views instead of a single view to achieve the better clustering performance. To effectively exploit data correlation consensus among multi-views, in this paper, we study subspace clustering for multi-view data while keeping individual views well encapsulated. For characterizing data correlations, we generate a similarity matrix in a way that high affinity values are assigned to data objects within the same subspace across views, while the correlations among data objects from distinct subspaces are minimized. Before generating this matrix, however, we should consider that multi-view data in practice might be corrupted by noise. The corrupted data will significantly downgrade clustering results. We first present a novel objective function coupled with an angular based regularizer. By minimizing this function, multiple sparse vectors are obtained for each data object as its multiple representations. In fact, these sparse vectors result from reaching data correlation consensus on all views. For tackling noise corruption, we present a sparsity-based approach that refines the angular-based data correlation. Using this approach, a more ideal data similarity matrix is generated for multi-view data. Spectral clustering is then applied to the similarity matrix to obtain the final subspace clustering. Extensive experiments have been conducted to validate the effectiveness of our proposed approach.

Content discovery is a critical issue in unstructured Peer-to-Peer (P2P) networks as nodes maintain only local network information. However, similarly without global information about human networks, one still can find specific persons via his/her friends by using social information. Therefore, in this paper, we investigate the problem of how social information (i.e., friends and background information) could benefit content discovery in P2P networks. We collect social information of 384,494 user profiles from Facebook, and build a social P2P network model based on the empirical analysis. In this model, we enrich nodes in P2P networks with social information and link nodes via their friendships. Each node extracts two types of social features–Knowledge and Similarity–and assigns more weight to the friends that have higher similarity and more knowledge. Furthermore, we present a novel content discovery algorithm which can explore the latent relationships among a node’s friends. A node computes stable scores for all its friends regarding their weight and the latent relationships. It then selects the top friends with higher scores to query content. Extensive experiments validate performance of the proposed mechanism. In particular, for personal interests searching, the proposed mechanism can achieve 100% of Search Success Rate by selecting the top 20 friends within two-hop. It also achieves 6.5 Hits on average, which improves 8x the performance of the compared methods.

As a new paradigm, Software-as-a-service (SaaS) is becoming increasingly important in information technology industry because it provides a cost-effective alternative over traditional packaged applications. Apart from an application delivery model, SaaS is also a business model that encompasses a broad spectrum of business, marketing, and technical opportunities, as well as issues and challenges. Introducing the concept of ESaaS that is defined as Education Software-as-a-Service, this paper compares it with traditional software, and discusses how such software can support education systems as a tool for enhancing information, teaching and learning in cloud environment. This paper then focuses on the development model of ESaaS, providing systematic steps on how to develop ESaaS.

Services are an indispensable component in cloud computing. Web services are particularly important. As an increasing number of Web services provides equivalent functions, one common issue faced by users is the selection of the most appropriate one based on quality. This article presents a conceptual framework that characterizes the quality of Web services, an algorithm that quantifies them, and a system architecture that ranks Web services by using the proposed algorithm. In particular, the algorithm, called UsageQoS that computes the scores of quality of service (QoS) of Web services within a community, makes use of the usage frequencies of Web services. The frequencies are defined as the numbers of times invoked by other services in a given time period. The UsageQoS algorithm is able to optionally take user ratings as its initial input. The proposed approach has been validated by extensively experimenting on several datasets, including two real datasets. The results of the experiments have demonstrated that our approach is capable of estimating QoS parameters of Web services, regardless of whether user ratings are available or not.

Advances in technology offers new opportunities in enhancing teaching and learning. The new technologies enable individuals to personalize the environment in which they work or learn, a range of tools to meet their interests and needs. In this paper, we try to explore the salient features of the nature and educational potential of `cloud computing' (CC) in order to exploit the affordance of CC in teaching and learning in a higher education context. It is evident that cloud computing has a significant place in the higher education landscape both as a ubiquitous computing tool and a powerful platform. Although, the adoption of cloud computing promises various benefits to an organization, a successful adoption of cloud computing in an organization, particularly in educational institutes requires an understanding of different dynamics and expertise in diverse domains. This paper aims at a roadmap of Cloud Computing for Higher Education (CCHE) which provides with a number of steps for adopting cloud computing.

Wandering is among the most frequent, problematic, and dangerous behaviors for elders with dementia. Frequent wanderers likely suffer falls and fractures, which affect the safety and quality of their lives. In order to monitor outdoor wandering of elderly people with dementia, this paper proposes a real-time method for wandering detection based on individuals' GPS traces. By representing wandering traces as loops, the problem of wandering detection is transformed into detecting loops in elders' mobility trajectories. Specifically, the raw GPS data is first preprocessed to remove noisy and crowded points by performing an online mean shift clustering. A novel method called θ_WD is then presented that is able to detect loop-like traces on the fly. The experimental results on the GPS datasets of several elders have show that the θ_WD method is effective and efficient in detecting wandering behaviors, in terms of detection performance (AUC > 0.99, and 90% detection rate with less than 5 % of the false alarm rate), as well as time complexity.

Two-view methods have been well developed to identify human actions. However, in a case where the corresponding imaged points cannot induce distinguished measures, the performance of the methods deteriorates. For this reason, we propose a new view-invariant measure for human action recognition by enforcing tri-view constraints in this paper. This new measurement method can be tolerant to different rates of human actions and the anthropometric proportions. We apply our approach to video synchronization by imposing both the similarity ratio and the consistency in the trifocal tensor over entire video sequences. By testing on both synthetic and real data, our method has achieved higher tolerance to noise levels, as well as higher identification accuracy than the traditional two-view method. Experimental results demonstrate that our approach can identify human pose transitions, in spite of dynamic time-lines, different viewpoints and unknown camera parameters.

Two-view methods have been well developed to identify human actions. However, in a case where the corresponding imaged points cannot induce distinguished measures, the performance of the methods deteriorates. For this reason, we propose a new view-invariant measure for human action recognition by enforcing tri-view constraints in this paper. We apply our approach to video synchronization by imposing both the similarity ratio and the consistency in the trifocal tensor over entire video sequences. By testing on both synthetic and real data, our method has achieved higher tolerance to noise levels, as well as higher identification accuracy than the traditional two-view method. Experimental results demonstrate that our approach can identify human pose transitions, despite of dynamic time-lines, different viewpoints, and unknown camera parameters.

Identity fraud has become one of major concerns for broad communities. The new information era needs a new digital identity infrastructure to support next generation Internet. This article suggests a hierarchical structure for digital identities. We define and classify all digital identities into three broad categories: Object, People and Organization. This paper is the first to systematically address the classification of digital identities. More and more individuals and communities heavily rely on the network. Many countries are trying their own digital identity initiatives, like E-passport, national smart card, etc. This article intends to initiate a discussion on a universal digital identity infrastructure for our future. We believe that in the near future all paper-based identities will be replaced by digital identities. A robust digital identity infrastructure will take a vital role in the future information age.

Using local data information, the recently proposed local Fisher Discriminant Analysis (LFDA) algorithm18 provides a new way of handling the multimodal issues within classes where the conventional Fisher Discriminant Analysis (FDA) algorithm fails. Like the FDA algorithm (global counterpart), the LFDA suffers when it is applied to the higher dimensional data sets. In this paper, we propose a new formulation by which a robust algorithm can be formed. The new algorithm offers more robust results for higher dimensional data sets when compared with the LFDA in most cases. By extensive simulation studies, we have demonstrated the practical usefulness and robustness of our new algorithm in data visualization.

A parallel design and implementation of FM-index is presented in this paper. In applications, the performance of the FM-index is crucial, which is a self-contained, highly compressed indexing algorithm. With the popularity of multi-core processors, parallel computing allows the FM-index to run faster by performing multiple computations simultaneously when possible. Our approach works by splitting input data into overlapping blocks with equal size, and running them through the FM-index algorithm simultaneously on multiple processors. After analyzing and refactoring the sequential version, we organize the data flows of all operations according to a unified parallel framework. The experimental results show that, in general our approach has achieved a significant and sub-linear speedup on widespread symmetrical multi-processing architectures. This will greatly reduce the running time of executing operations on large data sets.

In this paper, least square approach is applied in subspace intersection (SI) method for the problem of bearing estimation in shallow water. Based on that, a method called constrained least square subspace intersection method (CLS-SI) is proposed. The mathematic expressions of CLS-SI are given. In addition, the relationship between CLS-SI and MUSIC is discussed. Simulations show that the performance of the new method proposed is better than that of the original SI method.

As more and more organisations and institutions are moving towards the e-learning strategy, more and more disparate data are distributed by different e-learning systems. How to effectively use this vast amount of distributed data becomes a big challenge. This paper addresses this challenge and works out a new mechanism to implement data integration for e-learning. A workflow management system based (WFMS-based) data integration model is contributed to the e-learning

The Web graph has recently been used to model the link structure of the Web. The studies of such graphs can yield valuable insights into Web algorithms for crawling, searching and discovery of Web communities. This paper proposes a new approach to clustering the Web graph. The proposed algorithm identifies a small subset of the graph as "core" members of clusters, and then incrementally constructs the clusters by a selection criterion. Two qualitative criteria are proposed to measure the quality of graph clustering. We have implemented our algorithm and tested a set of arbitrary graphs with good results. Applications of our approach include graph drawing and Web visualization.