Recent research in cooperative Multi-Agent Reinforcement Learning (MARL) has shown significant interest in utilizing Graph Neural Networks (GNNs) for communication learning due to their strong ability to process feature and topological information of agents into message representations for downstream action selection and coordination. However, GNNs generally assume network homogeneity that nodes of the same class tend to be interconnected. In real-world multi-agent systems, such assumptions are often unrealistic, as agents within the same class can be distant from each other. Furthermore, GNN-based MARL methods overlook the crucial role of feature similarity of agents in action coordination, which also restricts their performance. To overcome these limitations, we propose a Multi-Agent communication mechanism with Information preserving graph contrastive Learning (MAIL), which enhances message representation by preserving the comprehensive features of adjacent agents while integrating topological information. Specifically, MAIL considers three distinct graph views: original view, agent feature view, and global topological view. MAIL performs contrastive learning across three views to extract comprehensive information. MAIL effectively learns robust and expressive message representations for downstream tasks. Extensive experiments across various environments demonstrate that MAIL outperforms existing GNN-based MARL methods.

The interpretability of Convolutional Neural Networks (CNNs) has garnered significant attention in computer vision. Integrated Gradients (IG), as a widely used feature attribution method, quantifies the contribution of input features (pixels) to model predictions by accumulating gradients along an interpolation path. However, the linear interpolation path employed by IG can result in inflated attribution scores for irrelevant (unimportant) pixels, introducing noise into saliency maps and reducing their reliability. To address this issue, this paper proposes an improved feature attribution method called Dynamic Weighted Path Integration (DWPI). DWPI incorporates a dynamic interpolation path strategy, prioritizing the movement of the least important pixels to minimize interference from irrelevant pixels in attribution results. Additionally, DWPI leverages the model output rate to weight gradients, amplifying the influence of high-quality gradients and further enhancing the reliability of saliency maps. Experimental results on the ImageNet validation set demonstrate that DWPI outperforms other methods across various models and four standard perturbation test metrics. Ablation study further confirms the effectiveness of the dynamic interpolation path strategy and gradient weighting mechanism.

The proliferation of healthcare data sources, including diverse imaging modalities and biochemical measurements, has created unprecedented opportunities for comprehensive disease prediction. Multi-modal clinical data, encompassing medical imaging reports, biochemical assays, and longitudinal clinical records, provides a rich foundation for developing sophisticated diagnostic models. Graph Neural Networks (GNNs) have emerged as a leading methodological framework, distinguished by their capacity to model complex inter-patient relationships and capture community structures within patient data. Despite their promise, current GNN-based approaches exhibit limitations in handling noisy, low-quality data and often impose overly restrictive graph smoothness constraints. These limitations can obscure patient-specific variations and compromise model robustness. To overcome these challenges, we propose HierSSL (Hierarchical Self-Supervised Learning), a novel multi-modal disease prediction framework that enhances representational learning through dual-scale self-supervision mechanisms operating at both local and global levels. HierSSL's architecture specifically addresses two critical aspects: 1) the capture of local inter-modality dependencies and global community patterns, and 2) the optimization of multi-modal feature integration through an innovative combination of feature consistency constraints and graph contrastive learning. Empirical evaluation across two distinct disease prediction datasets demonstrates that HierSSL achieves statistically significant performance improvements compared to state-of-the-art methods, highlighting its efficacy in robust multi-modal data integration for disease prediction tasks.

Food recommendation systems play a pivotal role in shaping dietary salubrity and fostering sustainable lifestyles by recommending recipes and foodstuffs that align with user preferences. Metadata information of a recipe, encompassing multi-modal descriptions, constituent ingredients, and health-related attributes, can furnish a more holistic perspective on the recipe's profile, thereby augmenting recommendation performance. However, existing state-of-the-art methods often overlook the inherent interdependencies between modalities, ingredients, and health factors, leaving the health information pertaining to recipe characteristics underexploited. Notably, our preliminary investigation on two datasets unveiled that the semantic divergence between health-related knowledge and collaborative filtering signals is more pronounced in comparison to other metadata information, thereby potentially impeding the efficacy of food recommendation systems. To address these limitations, we propose HealthRec, a novel multi-modal food recommendation framework with health-aware knowledge distillation. HealthRec employs a global graph representation learning module to capture high-order dependencies across diverse food-related relations, enriching the representations. Subsequently, a co-attention network is leveraged to capture local, recipe-level knowledge transfer between modality-related and ingredient-related embeddings. Additionally, we exploit external supervision signals derived from WHO recommendations, utilizing knowledge distillation during the training phase to transfer local health-aware knowledge into global collaborative embeddings. Extensive experimentation on real-world datasets demonstrates HealthRec's superiority compared to current state-of-the-art recommendation baselines, highlighting its effectiveness in modeling health-aware food recommendations.

This research proposed a model called rational intelligence and studied its reasoning capability as compared to ChatGPT-4o in counterfactual reasoning tasks. Unlike traditional AI models that rely heavily on data-driven approaches, rational intelligence allows for reasoning over abstract principles and hypothetical scenarios, similar to human cognitive processes. The proposed Rational Intelligence Model (RIM) applies Large Language Models (LLMs) to enable human-like comprehension, knowledge application, and problem-solving capabilities. In complex counterfactual reasoning tasks with scenarios proven to be challenging to human adults, we demonstrate that RIM achieves a clearly higher accuracy rate (76%) compared to ChatGPT-4o (68%), showing its enhanced reasoning capabilities. Additionally, RIM incorporates a self-reflection mechanism to manage knowledge conflicts and gaps, which can further improve its performance and adaptability.

Graph Neural Networks (GNNs) have significantly advanced recommendation systems by capturing complex interplays between user-item relationships and dependencies. However, inherent noise in user behaviors, manifesting as random clicks and diverse browsing patterns, disrupts the structural integrity of graphs, thereby degrading the accuracy and reliability of GNN-based recommendation systems. Traditional graph pruning methods, which remove or reweight connections, often fail to adequately address this noise because they neglect the deeper causal factors influencing user choices, resulting in biased outcomes. To confront these challenges, this paper presents the GNN-based Causal Denoising Framework (GCDF). GCDF employs causal relationships to filter out noisy connections, thus enhancing the performance of GNNs. By utilizing a denoised graph that more accurately reflects the causal interactions among items, GCDF significantly improves the accuracy and reliability of recommendations, as evidenced by comprehensive empirical evaluations.

As AI applications are widely deployed in various fields, the computing cost of AI is skyrocketing. The high cost not only puts pressure on the environment but also poses challenges for researchers entering the field of deep learning. Green AI is gradually gaining attention, aiming to reduce computing costs and make AI application deployment more efficient and environmentally friendly. However, due to the wide variety of deep learning models, there are many challenges in efficiently deploying multiple models on GPUs. An unreasonable model deployment strategy will lead to insufficient utilization of GPU computing resources. We address some of these challenges through AutoMP, a tool that automates performance testing for model placement on GPUs. Through AutoMP, researchers can flexibly initiate a large number of experiments to study which models are suitable for inference tasks on the same GPU, thereby making GPU utilization more efficient. AutoMP provides a user-friendly visual interface and an API to meet users’ needs in different scenarios. AutoMP also provides a complete experimental analysis tool that generates visual charts of experimental data from multiple dimensions and gives experimental conclusions to assist researchers in making decisions. To date, AutoMP has been used by a large number of users for empirical research on deep learning model placement. The cumulative number of experiments has exceeded 10,000. The flexibility and extensibility of AutoMP and our own experience using it show that this tool plays a vital role in promoting Green AI.

Crowdsourcing is a favorable computing paradigm for processing computer-hard tasks by harnessing human intelligence. However, generic crowdsourcing systems may lead to privacy-leakage through the sharing of worker data. To tackle this problem, we propose a novel approach, called iFedCrowd (incentive-boosted Federated Crowdsourcing), to manage the privacy and quality of crowdsourcing projects. iFedCrowd allows participants to locally process sensitive data and only upload encrypted training models, and then aggregates the model parameters to build a shared server model to protect data privacy. To motivate workers to build a high-quality global model in an efficacy way, we introduce an incentive mechanism that encourages workers to constantly collect fresh data to train accurate client models and boosts the global model training. We model the incentive-based interaction between the crowdsourcing platform and participating workers as a Stackelberg game, in which each side maximizes its own profit. We derive the Nash Equilibrium of the game to find the optimal solutions for the two sides. Experimental results confirm that iFedCrowd can complete secure crowdsourcing projects with high quality and efficiency.

Major depressive disorder (MDD) is the most common psychological disorder that affects mental and physical health. To narrow the gap in real world mental healthcare and improve the effectiveness of MDD treatment, an increasing number of artificial intelligence (AI) methods have been proposed to explore electroencephalography (EEG) features, including traditional signal features and measures of brain functional connectivity network (BFCN), for the recognition of depression-related patterns. However, these methods fail to capture long-term dependencies and limit the modeling ability of information transmission dependencies in MDD brain regions. To address these issues, we propose a novel brain functional residual temporal convolution network (BFRTCN) method for MDD recognition. On one hand, this model directly focuses on the connectivity weights of BFCNs to model the information transmission between brain regions, allowing for better differentiation of the differences in information transmission patterns between MDD and normal control (NC). On the other hand, we introduce a residual temporal convolution network (ResiTCN) that utilizes temporal convolution layers to capture short-term changes in brain regions and establish residual connections to help maintain long-term dependencies for improving ability to capture disease variations. Experimental results on benchmark datasets validate the superior performance and time complexity of BFRTCN. Analysis shows that the Beta band MDD transmission mode is relatively stable. There are defects in the brain functional connections between the frontal and right temporal (RT) regions on Alpha and Gamma bands, which can serve as potential biomarkers for MDD recognition.

Counterfactual explanations (CFEs) exemplify how to minimally modify a feature vector to achieve a different prediction for an instance. CFEs can enhance informational fairness and trustworthiness, and provide suggestions for users who receive adverse predictions. However, recent research has shown that multiple CFEs can be offered for the same instance or instances with slight differences. Multiple CFEs provide flexible choices and cover diverse desiderata for user selection. However, individual fairness and model reliability will be damaged if unstable CFEs with different costs are returned. Existing methods fail to exploit flexibility and address the concerns of non-robustness simultaneously. To address these issues, we propose a conceptually simple yet effective solution named Counterfactual Explanations with Minimal Satisfiable Perturbations (CEMSP). Specifically, CEMSP constrains changing values of abnormal features with the help of their semantically meaningful normal ranges. For efficiency, we model the problem as a Boolean satisfiability problem to modify as few features as possible. Additionally, CEMSP is a general framework and can easily accommodate more practical requirements, e.g., casualty and actionability. Compared to existing methods, we conduct comprehensive experiments on both synthetic and real-world datasets to demonstrate that our method provides more robust explanations while preserving flexibility.

Edge computing has been widely used to deploy and service deep learning applications. Equipped with GPUs, edge nodes can process concurrent incoming inference requests of the deep learning model. However, existing methods for inference tasks do not allow efficient parallel handling of user requests. This paper investigates the popular Transformer deep learning model and develops ParaTra, a parallel transformer inference framework for providing parallel inference services to users. In the framework, the Transformer model is partitioned and deployed in users’ devices and the edge node to efficiently utilize their processing power. The concurrent inference tasks with different sizes are dynamically packaged in a scheduling queue and sent in batch to an encoder-decoder pipeline for processing. ParaTra can significantly reduce the overheads of parallel processing and the usage of GPU memory. Experiment results show that ParaTra can save up to 37.1% of GPU memory usage and improve 8.4 times of processing speed.

Learning the directed acyclic graph (DAG) among causal variables is a fundamental pre-task in causal discovery. Available DAG learning solutions canonically focus on homogeneous nodes with multiple variables and assume i.i.d. samples, how to learn DAG on typical attributed heterogeneous network (AHN) composed with different types of inter-dependent nodes and diverse attributes is a practical but more difficult task. In this paper, we propose HetDAG to identify DAG among nodes from heterogeneous network. HetDAG first embeds different types of node attributes and aggregates these embeddings as the node's raw representation. Then it uses contrastive learning with prior network structure to explore latent relationships between nodes and update the representation. Next, HetDAG introduces an attention-based DAG learning module that takes node representations as input to search DAG and orient edges between nodes. To the best of our knowledge, HetDAG is the first study to learn DAG on heterogeneous networks. Extensive experiments on both semi-synthetic and real data show that HetDAG can learn DAG in an efficacy way and outperforms the state-of-the-art approaches. The results on real biological networks confirm that HetDAG can find out the causal relations between lncRNAs and miRNAs.

In the medical scenarios, the demand for secure sharing of medical data and trusted federated computing continues to increase, and blockchain technology can provide secure, credible, and tamper-resistant capabilities, which makes it necessary to combine the two. However, the full application of blockchain to medical scenarios faces two issues. Medical assets such as medical cohort data with strong data correlation and large scale are difficult to be accurately described and safely operated by blockchain. Transactions such as disease diagnosis and hospitalization prediction with many parameters are difficult to execute concurrently in blockchain. Therefore, we propose MAA model, which closely associates patients with assets, separates the logic of asset operations from its storage. At the same time, we propose OPE model with double-layer pipeline concurrency. By constructing the Dependency Graph, and generating multiple blocks with low conflict rates, which are executed simultaneously among multiple Execute Groups, OPE supports the high concurrent execution of medical transactions with high computing power requirements such as trusted federated learning. Experiments show that our model supports multiple types of medical data on-chain compared to existing models, and the concurrency is increased by at least 40% in a high-conflict medical environment.

Phenotype prediction aims to infer the traits of living organisms based on genomics data, which has important applications in biology such as cancer subtype diagnosis and crop breedings. Traditional phenotype prediction approaches only learn the low-dimensional representation of samples, but ignore the interaction between biomolecules and cannot use the topology structure of heterogeneous molecular networks. Furthermore, most of them lack interpretability and do not effectively identify key biomolecules associated with phenotypes. In this paper, we propose a heterogeneous network embedding based solution (PhenoHNE) to predict phenotype by fusing topology information of molecules. PhenoHNE firstly utilizes variational graph autoencode (VGAE) to obtain the embedding representation of molecules in the heterogeneous network. Secondly, PhenoHNE adopts multilayer perceptron (MLP) with attention mechanism to learn sample representation. Finally, it fuses molecular embedding representation and sample representation to predict the phenotype of samples. In this way, the genetics information of heterogeneous molecular network can further guide the feature learning of samples and improve the prediction performance. Experimental results on human and maize datasets confirm that PhenoHNE outperforms competitive methods by a large margin under different evaluation protocols, and it also can effectively identify the key molecules associated with phenotypes of interests.

MOTIVATION: Cancer subtype diagnosis is crucial for its precise treatment and different subtypes need different therapies. Although the diagnosis can be greatly improved by fusing multiomics data, most fusion solutions depend on paired omics data, which are actually weakly paired, with different omics views missing for different samples. Incomplete multiview learning-based solutions can alleviate this issue but are still far from satisfactory because they: (i) mainly focus on shared information while ignore the important individuality of multiomics data and (ii) cannot pick out interpretable features for precise diagnosis. RESULTS: We introduce an interpretable and flexible solution (LungDWM) for Lung cancer subtype Diagnosis using Weakly paired Multiomics data. LungDWM first builds an attention-based encoder for each omics to pick out important diagnostic features and extract shared and complementary information across omics. Next, it proposes an individual loss to jointly extract the specific information of each omics and performs generative adversarial learning to impute missing omics of samples using extracted features. After that, it fuses the extracted and imputed features to diagnose cancer subtypes. Experiments on benchmark datasets show that LungDWM achieves a better performance than recent competitive methods, and has a high authenticity and good interpretability. AVAILABILITY AND IMPLEMENTATION: The code is available at http://www.sdu-idea.cn/codes.php?name=LungDWM. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Clinical pathways are standardized, evidence-based multidisciplinary management plans. Many countries propose their national clinical pathways to improve the quality of care, reduce variation in clinical practice, and increase the efficient use of healthcare resources. Nevertheless, clinical pathways are typically not prescriptive, and the patient’s care journey is an individual one, therefore how to handle the variances of clinical pathways is an important issue. Previous methods construct clinical pathway recommendation models either using national standard clinical pathways to obtain the guidance, or using real-world clinical datasets to obtain clinical experience. However, few research tries to use both of them. This will result in existing algorithms that cannot accurately recommend personalized clinical pathway. To overcome the above problems, we propose P ersonalized C linical P athway Rec ommendation(PCPRec). On the one hand, to obtain general clinical pathway recommendations, we built a novel module to pre-train the self-attention model based on the national standard clinical pathway. So that we can use it as a guide to enhance the accuracy of recommending personalized clinical pathway. On the other hand, we obtain the patient’s treatment history sequence from real-world clinical datasets, and use the self-attention model for training. The purpose is to learn from the experience of the relationship between clinical items to meet patient’s individual needs. Extensive experimental results show that the proposed model achieves the best results compared to state-of-art methods on benchmark datasets.

Genome-phenome association (GPA) play a crucial part in deciphering the complex pathology of phenotypes (i.e., traits and diseases). Heterogeneous network-based GPA solutions can model the complex connections between multi-types of molecules by viewing them as nodes, and give better results than using single/two-omics data alone, but they overwhelmingly need to project other molecules toward homogeneous gene/phenotype nodes for data fusion and prediction, such projections result in information loss. Matrix factorization based data fusion can avoid such projection by integrating multi-type data in a coherent way, but they typically perform linear factorization and cannot mine the nonlinear relationships between molecules, which compromise the GPA analysis. Furthermore, most of them can not synergy network topology and node attribution information in a principle way. In this paper, we propose a deep matrix factorization based solution (DeepGPA) to predict GPAs by fusing heterogeneous molecular network and diverse attributes of nodes. DeepGPA performs deep matrix factorization on the block adjacency matrices of heterogeneous network in a cooperative manner to obtain the nonlinear representations of different moleclues. In addition, it performs low-rank representation learning on the attribute data with the shared nonlinear representations. In this way, both the network topology and node attributes are jointly mined to explore the representations of molecules and complex interplays between molecules and phenotypes. DeepGPA then uses the representational vectors of gene and phenotype nodes to predict GPAs. Experimental results on Maize datasets confirm t hat Deep DPA out performs competitive methods by a large margin under different evaluation protocols.

Crowdsourcing is a popular and relatively economic way to harness human intelligence to process computer-hard tasks. Due to diverse factors (i.e., task difficulty, worker capability, and incentives), the collected answers from various crowd workers are of different quality. Many approaches have been proposed to manage high quality answers and to reduce the budget by modelling tasks, workers, or both. However, most of the existing approaches implicitly assume that the capability of workers is fixed during the crowdsourcing process. But in practice, such capability can be improved by gradually completing easy to hard tasks, alike human beings’ intrinsic self-paced learning ability. In this paper, we investigate crowdsourcing with self-paced workers, whose capability can be gradually boosted as he/she scrutinises and completes easy to hard tasks. Our proposed SPCrowd (Self-Paced Crowd worker) first asks workers to complete a set of golden tasks with known annotations; provides feedback to assist workers with capturing the raw modes of tasks and to spark the self-paced learning, which in turn facilitates the estimation of workers’ quality and tasks’ difficulty. It then introduces a task difficulty model to quantify the difficulty of tasks and rank them from easy to hard, and a benefit maximization criterion for task assignment, which can dynamically monitor the quality of self-paced workers and assign the sorted tasks to capable workers. In this way, a worker can successfully complete hard tasks after he/she completes easier and related tasks. Experimental results on semi-simulated and real crowdsourcing projects show that SPCrowd can better control the quality and save the budget compared to competitive baselines.

Abstract Microtask crowdsourcing is a form of crowdsourcing in which work is decomposed into a set of small, self-contained tasks, which each can typically be completed in a matter of minutes. Due to the various capabilities and knowledge background of the voluntary participants on the Internet, the answers collected from the crowd are ambiguous and the final answer aggregation is challenging. In this process, the choice of quality control strategies is important for ensuring the quality of the crowdsourcing results. Previous work on answer estimation mainly used expectation–maximization (EM) approach. Unfortunately, EM provides local optimal solutions and the estimated results will be affected by the initial value. In this paper, we extend the local optimal result of EM and propose an approximate global optimal algorithm for answer aggregation of crowdsourcing microtasks with binary answers. Our algorithm is expected to improve the accuracy of real answer estimation through further likelihood maximization. First, three worker quality evaluation models are presented based on static and dynamic methods, respectively, and the local optimal results are obtained based on the maximum likelihood estimation method. Then, a dominance ordering model (DOM) is proposed according to the known worker responses and worker categories for the specified crowdsourcing task to reduce the space of potential task-response sequence while retaining the dominant sequence. Subsequently, a Cut-point neighbor detection algorithm is designed to iteratively search for the approximate global optimal estimation in a reduced space, which works on the proposed dominance ordering model (DOM). We conduct extensive experiments on both simulated and real-world datasets, and the experimental results illustrate that the proposed approach can obtain better estimation results and has higher performance than regular EM-based algorithms.

Hospital readmission prediction employs reliable predictive models to evaluate the readmission risk of patients upon discharge. Identifying patients with high readmission risk and paying additional attention to them can ease the burden on both patients and society. Recently, considerable attention has been paid to personalized readmission predictions, i.e., building an independent model for each target patient. However, existing personalized predictive models can be easily affected by data insufficiency and provide poor generalization capabilities. To address these challenges, in this paper, we propose the CLusterbased mUlti-task lEarning model (CLUE) to achieve personalized hospital readmission prediction. CLUE groups patients into different clusters based on a multi-angle similarity metric to preserve the interrelated information of patients with highly similar clinical behaviors. Due to different group characteristics of patients, the clusters of patients are imbalanced. Given that, CLUE treats the hospital readmission prediction for each cluster of patients as one task and learns multiple tasks jointly by parameter sharing mechanisms. In this way, not only can the data insufficiency problem be alleviated by supplementing individual models with the shared information from other clusters, but also the specific information of each cluster can be preserved for personalization. We conduct extensive experiments on a real-world dataset of electronic health records, and show that CLUE significantly outperforms competitive comparative methods.

Commonsense knowledge is critical in human reading comprehension. While machine comprehension has made significant progress in recent years, the ability in handling commonsense knowledge remains limited. Synonyms are one of the most widely used commonsense knowledge. Constructing adversarial dataset is an important approach to find weak points of machine comprehension models and support the design of solutions. To investigate machine comprehension models' ability in handling the commonsense knowledge, we created a Question and Answer Dataset with common knowledge of Synonyms (QADS). QADS are questions generated based on SQuAD 2.0 by applying commonsense knowledge of synonyms. The synonyms are extracted from WordNet. Words often have multiple meanings and synonyms. We used an enhanced lesk algorithm to perform word sense disambiguation to identify synonyms for the context. ELECTRA achieves the state-of-art result on the SQuAD 2.0 dataset in 2019. With about 1/10 scale, ELECTRA can achieve similar performance as BERT does. However, QADS shows that ELECTRA has little ability to handle commonsense knowledge of synonyms. In our experiment, ELECTRA-small can achieve 70% accuracy on SQuAD 2.0, but only 20% on QADS. ELECTRA-large did not perform much better. Its accuracy on SQuAD 2.0 is 88% but dropped significantly to 26% on QADS. In our earlier experiments, BERT, although also failed badly on QADS, was not as bad as ELECTRA. The result shows that even top-performing NLP models have little ability to handle commonsense knowledge which is essential in reading comprehension.

In micro-task crowdsourcing, the answers collected from the crowd are ambiguous and the final answer aggregation is challenging due to the various capabilities and knowledge background of the voluntary participants on the Internet. In this paper, we extend the local optimal result of Expectation-Maximization(EM) approach and propose an approximate global optimal algorithm for answer aggregation of crowdsourcing microtasks with binary answers. Our algorithm is expected to improve the accuracy of real answer estimation through further likelihood maximization. We conduct extensive experiments on both simulated and real-world datasets, and the experimental results illustrate that the proposed approach can obtain better estimation results and has higher performance than regular EM-based algorithms.

Nowadays, with the rapid growth of microblogging networks for news propagation, there are increasingly more people accessing news through such emerging social media. In the meantime, fake news now spreads at a faster pace and affects a larger population than ever before. Compared with traditional text news, the news posted on microblog often has attached images in the context. So how to correctly and autonomously detect fakes news in a multi-modal manner becomes a prominent challenge to be addressed. In this paper, we propose an end-to-end model, named BERT-based domain adaptation neural network for multi-modal fake news detection (BDANN). BDANN comprises three main modules: a multi-modal feature extractor, a domain classifier and a fake news detector. Specifically, the multi-modal feature extractor employs the pretrained BERT model to extract text features and the pretrained VGG-19 model to extract image features. The extracted features are then concatenated and fed to the detector to distinguish fake news. The role of the domain classifier is mainly to map the multi-modal features of different events to the same feature space. To assess the performance of BDANN, we conduct extensive experiments on two multimedia datasets: Twitter and Weibo. The experimental results show that BDANN outperforms the state-of-the-art models. Moreover, we further discuss the existence of noisy images in the Weibo dataset that may affect the results.

As considerable amounts of POI check-in data have been accumulated, successive point-of-interest (POI) recommendation is increasingly popular. Existing successive POI recommendation methods only predict where user will go next, ignoring when this behavior will occur. In this work, we focus on predicting POIs that will be visited by users in the next 24 hours. As check-in data is very sparse, it is challenging to accurately capture user preferences in temporal patterns. To this end, we propose a category-aware deep model CatDM that incorporates POI category and geographical influence to reduce search space to overcome data sparsity. We design two deep encoders based on LSTM to model the time series data. The first encoder captures user preferences in POI categories, whereas the second exploits user preferences in POIs. Considering clock influence in the second encoder, we divide each user’s check-in history into several different time windows and develop a personalized attention mechanism for each window to facilitate CatDM to exploit temporal patterns. Moreover, to sort the candidate set, we consider four specific dependencies: user-POI, user-category, POI-time and POI-user current preferences. Extensive experiments are conducted on two large real datasets. The experimental results demonstrate that our CatDM outperforms the state-of-the-art models for successive POI recommendation on sparse check-in data.

Abnormal co-occurrence medical visit behavior refers to the fraudulent behavior of fraudsters who hold multiple medical insurance cards and frequently purchase medicines within the same time period at the same location to defraud insurance medical insurance fund. Identifying abnormal co-occurrence medical visit behaviors plays a critical role in automated medical insurance fraud detection. However, the conventional methods mainly focus on the mining of frequent patterns, which may lead to misjudgement of normal patients who regularly seek medical treatment for a long time. In order to address this problem, we propose a novel biclustering algorithm, which finds suspicious patient groups who frequent purchased medicines within the same time period at the same location, and filters out normal patients from suspicious patients. Experimental results show that Biclutering-sim outperforms the competitor in detecting medical insurance fraud.