IEEE Transactions on Big Data最新文献

{"title":"Incremental Causal Graph Learning for Online Cyberattack Detection in Cyber-Physical Infrastructures","authors":"Arun Vignesh Malarkkan;Dongjie Wang;Haoyue Bai;Yanjie Fu","doi":"10.1109/TBDATA.2025.3648314","DOIUrl":"https://doi.org/10.1109/TBDATA.2025.3648314","url":null,"abstract":"The escalating threat of cyberattacks on real-time critical infrastructures poses significant risks to public safety, necessitating detection methods that can effectively capture complex system interdependencies and adapt to evolving attack patterns. Traditional real-time anomaly detection techniques often produce excessive false positives due to their statistical sensitivity to high data variability and class imbalance. To address these limitations, recent research has explored modeling causal relationships among system components. However, prior work predominantly focuses on offline causal graph-based approaches that require static historical data and fail to generalize to real-time settings. These methods are fundamentally constrained by: (1) their inability to adapt to dynamic shifts in data distribution without retraining, and (2) the risk of catastrophic forgetting when lacking timely supervision in live systems. To overcome these challenges, we propose <bold>INCADET</b>, a novel framework for <italic>incremental causal graph learning</i> tailored to real-time cyberattack detection. INCADET dynamically captures evolving system behavior by incrementally updating causal graphs across streaming time windows. The framework comprises three modules: <bold>1) Early Symptom Detection:</b> Detects transitions in system status using divergence in edge-weight distributions across sequential causal graphs. <bold>2) Incremental Causal Graph Learning:</b> Leverages experience replay and edge reinforcement to continually refine causal structures while preserving prior knowledge. <bold>3) Causal Graph Classification:</b> Employs Graph Convolutional Networks (GCNs) to classify system status using the learned causal graphs. Extensive experiments on real-world critical infrastructure datasets demonstrate that INCADET achieves superior accuracy, robustness, and adaptability compared to both offline, online, and deep temporal baselines in evolving attack scenarios.","PeriodicalId":13106,"journal":{"name":"IEEE Transactions on Big Data","volume":"12 3","pages":"881-892"},"PeriodicalIF":8.1,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148504290","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Preference-Aware Multimodal Contrastive Learning Recommendation","authors":"Hairong Wang;Zhihang Yi;Zhaojing Xu;Jing Wang;Hongying Li","doi":"10.1109/TBDATA.2026.3668549","DOIUrl":"https://doi.org/10.1109/TBDATA.2026.3668549","url":null,"abstract":"Multimodal recommendation systems struggle to effectively integrate heterogeneous modalities and select informative negative samples, which limits their ability to capture users’ true preferences. To address these challenges, we propose Preference-Aware Multimodal Contrastive Learning (PAMCL), a novel framework that leverages a lightweight large language model (LLM) to generate explicit user preference descriptions for guiding the recommendation process. Our approach introduces a simplified multimodal graph aggregation mechanism that unifies feature propagation and contrastive learning, enhancing semantic consistency while reducing computational complexity. Additionally, we develop a preference-guided multiple negative sampling strategy that utilizes LLM-generated preference embeddings to identify challenging negative samples while avoiding false negatives. Experimental results on three public multimodal recommendation benchmarks demonstrate that PAMCL significantly outperforms state-of-the-art methods by effectively addressing false negative issues in sampling and noise contamination in multimodal fusion.","PeriodicalId":13106,"journal":{"name":"IEEE Transactions on Big Data","volume":"12 3","pages":"1033-1044"},"PeriodicalIF":8.1,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148507425","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"PsycHin: Finding the Trade-Off With Heterogeneous Information Network Toward Psychometric Preference Assessment","authors":"Qingren Wang;Xu Wen;Juan Hou;Hong Zhong;Yiwen Zhang","doi":"10.1109/TBDATA.2026.3653641","DOIUrl":"https://doi.org/10.1109/TBDATA.2026.3653641","url":null,"abstract":"In psychometry, psychometric profiles of an individual subject as well as correlations among the psychometric profiles are crucial supports for assessing the psychometric preference of the individual subject. Intuitively, while sufficient psychometric profiles in terms of amount and types yield high assessment accuracy, too many psychometric profiles increase workloads of subjects, because measuring a psychometric profile of an individual subject requires filling out one psychological questionnaire at least. The trade-off between assessment accuracy and participant burden constitutes a core challenge in psychometric research, because it directly affects both the statistical validity of findings and the cognitive feasibility of large-scale evaluation. Existing assessment frameworks often struggle to reconcile efficiency with precision. This challenge underscores the need for a new modeling paradigm capable of reducing measurement redundancy while preserving interpretability and reliability. Therefore, inspired by the semantic field, this paper defines a new concept, namely <i>profile field</i>, and proposes a novel framework, i.e., <i>PsycHin</i>, to achieve the trade-off. Specifically, <i>PsycHin</i> can be decomposed into three phases: <inline-formula><tex-math>$i$</tex-math></inline-formula>) modeling the psychologically cognitive processes, where <i>PsycHin</i> learns multiple explicit and implicit correlations among psychometric profiles based on a heterogeneous information network and defines a profile meta structure to guide instance extraction; <inline-formula><tex-math>$ii$</tex-math></inline-formula>) fitting data and calculating similarities among subjects, where <i>PsycHin</i> utilizes <i>tanh</i> to make up for the rating biases caused by the discrete scores from options of psychological questionnaires and employs entropy to measure the cognitive levels of subjects on same objects; and <inline-formula><tex-math>$iii$</tex-math></inline-formula>) constructing profile fields, where <i>PsycHin</i> calculates the importance of psychometric profiles and filters the noisy ones by leveraging the contribution rates of psychometric profiles. Our experimental results not only show that <i>PsycHin</i> performs well, but also demonstrate that incorporating the heterogeneous information network into psychometric preference assessment does improve the performance of <i>PsycHin</i>.","PeriodicalId":13106,"journal":{"name":"IEEE Transactions on Big Data","volume":"12 3","pages":"962-976"},"PeriodicalIF":8.1,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148506503","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"GBDE: Granular-Ball Textual Adversarial Attacks With Differential Evolution","authors":"Zeli Wang;Tianhao Pu;Shuyin Xia;Longlong Lin;Linpin Liu;Guoyin Wang","doi":"10.1109/TBDATA.2025.3644670","DOIUrl":"https://doi.org/10.1109/TBDATA.2025.3644670","url":null,"abstract":"Neural network models in natural language processing have demonstrated remarkable proficiency, but theyare still vulnerable to adversarial attacks, where minute and indistinguishable alterations can profoundly disturb their performance. Existing methods struggle to define synonyms accurately and rely on greedy traversal or gradient computation, making them difficult to deploy. Inspired by these, this paper proposes a novel text adversarial attack method, which constructs rich candidate adversarial examples by non-trivially combining the <underline>G</u>ranular-<underline>B</u>all computing and <underline>D</u>ifferential <underline>E</u>volution, termed GBDE. Through iterative optimization guided by a meticulously crafted loss function, GBDE iteratively optimizes to generate effective and high-quality adversarial examples. Specifically, GBDE realizes a multi-granularity granular-ball clustering method following the data features, to establish robust synonym sets. It then performs synonym substitutions on target samples to create candidate adversarial samples and further diversifies these samples through a customized differential evolution process. Additionally, the loss function incorporates a reward and punishment mechanism to steer the optimization of attacks. Empirical results showcase GBDE’s superiority when contrasted with cutting-edge models. Notably, GBDE produces high-quality adversarial texts with a larger success rate. Besides, ablation studies confirm the contributions of granular-ball clustering and differential evolution in text attacks, and the rationality of the loss function design.","PeriodicalId":13106,"journal":{"name":"IEEE Transactions on Big Data","volume":"12 3","pages":"843-854"},"PeriodicalIF":8.1,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148508344","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"TrustChain: A Blockchain Framework for Auditing and Verifying Aggregators in Decentralized Federated Learning","authors":"Ehsan Hallaji;Roozbeh Razavi-Far;Mehrdad Saif","doi":"10.1109/TBDATA.2025.3639977","DOIUrl":"https://doi.org/10.1109/TBDATA.2025.3639977","url":null,"abstract":"The serverless nature of Decentralized Federated Learning (DFL) requires allocating the aggregation role to specific participants in each federated round. Current DFL architectures ensure the trustworthiness of the aggregator node upon selection. However, most of these studies overlook the possibility that the aggregating node may turn rogue and act maliciously after being nominated. To address this problem, this paper proposes a DFL structure, called <monospace>TrustChain</monospace>, that scores the aggregators before selection based on their past behavior and additionally audits them after the aggregation. To do this, the statistical independence between the client updates and the aggregated model is continuously monitored using the Hilbert-Schmidt Independence Criterion (HSIC). The proposed method relies on several principles, including blockchain, anomaly detection, and concept drift analysis. The designed structure is evaluated on several federated datasets and attack scenarios with different numbers of Byzantine nodes.","PeriodicalId":13106,"journal":{"name":"IEEE Transactions on Big Data","volume":"12 3","pages":"741-751"},"PeriodicalIF":8.1,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148508624","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Learn to Cluster Human Mobility Pattern for Post-Disaster Analysis","authors":"Wenjing Li;Jinyu Chen;Yuhao Yao;Peiran Li;Hill Hiroki Kobayashi;Haoran Zhang;Xuan Song;Ryosuke Shibasaki;Xiaodan Shi","doi":"10.1109/TBDATA.2026.3653656","DOIUrl":"https://doi.org/10.1109/TBDATA.2026.3653656","url":null,"abstract":"Disaster has a great impact on human mobility patterns. Clustering mobility patterns by generalizing group-level characteristics from diverse individual trajectories plays a vital role in informing post-disaster recovery strategies. However, predefined clustering criteria or supervised labeling alone are insufficient to adequately capture the dynamics of mobility patterns during disaster events. Besides, the existing methods are not enough to capture the sensitive changes in mobility patterns in disaster events and handle the data of high-dimensional. This research investigates an embedded deep learning-based method which can automatically extract the groups’ short-term feature of mobility patterns and achieves short-term mobility pattern clustering during disaster. The proposed method employs a Transformer-based temporal encoder to capture intra-day sequence patterns and integrates a VAE component with an embedded latent variable that directly encodes group-level mobility modes. We also design a compactness–separation loss that explicitly encourages within-mode feature compactness and between-mode feature separation. Based on massive mobile data, we conduct mobility pattern clustering on the case of 2011 Fukushima Earthquake. Compared to conventional clustering approaches, the proposed model structure is more discriminative and can capture the more sensitive changes between pre- and post- events. Compared to baselines with different loss functions, proposal methods can make more accurate fitting result and obtain more discrete clusters modes. Additionally, sensitivity analysis is conducted to examine the influence of key hyperparameters within the model. Based on the clustering outcomes, five representative mobility patterns are identified. We further analyze the spatial-temporal characteristics of mobility pattern changes during the disaster events and the recovery period of mobility pattern.","PeriodicalId":13106,"journal":{"name":"IEEE Transactions on Big Data","volume":"12 3","pages":"936-947"},"PeriodicalIF":8.1,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148507799","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Multi-Scale Temporal Interpolation Patch Graph Neural Network for Multivariate Time Series Forecasting With Missing Values","authors":"Wenchang Zhang;Xiao Wu;Linjiang Zheng;Xiao Cai;Jiang Zhong","doi":"10.1109/TBDATA.2025.3639915","DOIUrl":"https://doi.org/10.1109/TBDATA.2025.3639915","url":null,"abstract":"Recent studies have revealed the significant potential of spatiotemporal graph neural networks for multivariate time series forecasting with missing values. These methods typically represent interactions between time series as graph structures, where each time step is encoded as a graph node. Existing methods largely rely on self-learned or predefined graph structures to capture spatial dependencies within temporal data. However, in real-world applications, interactions between nodes are often directional and may span long temporal distances. Moreover, at high missing rates, interpolation across time scales often produces varied results in reconstructing missing time series. To address these challenges, this paper investigates the directed interactions of missing values in time series an d their multi-scale interpolation process. We propose the Multi-scale Temporal Interpolation Patch Graph Neural Network (MTIPG), which enables precise spatiotemporal modeling with limited data. Specifically, we first design a hierarchical structure combined with dilated convolutions to capture feature interpolation of missing sequences at specific time scales. Then, we quantify the positional relationships between patches using the dot product similarity between the head and tail embeddings of missing sequences, facilitating implicit modeling of missing variables and directional information propagation. Finally, we integrate these modules to generate accurate prediction results. Extensive experiments on four real-world datasets demonstrate that MTIPG consistently outperforms state-of-the-art baselines across various missing rates, achieving accurate predictions for all variables’ future values, even with up to 60% missing data.","PeriodicalId":13106,"journal":{"name":"IEEE Transactions on Big Data","volume":"12 3","pages":"752-767"},"PeriodicalIF":8.1,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148508908","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Accelerating Heterogeneous Tensor Parallelism via Flexible Workload Control","authors":"Zhigang Wang;Xu Zhang;Ning Wang;Chuanfei Xu;Yu Gu;Hui Lu;Dawei Zhao;Zhihong Tian","doi":"10.1109/TBDATA.2026.3657302","DOIUrl":"https://doi.org/10.1109/TBDATA.2026.3657302","url":null,"abstract":"Transformer-based foundation models are becoming deeper and larger. For fast training, their billions of parameters (tensors) are split onto parallel tasks running on many modern yet expensive accelerators. To amortize the huge hardware investments, it is cost-effective to share the aggregated resources among multi-tenants. However, resource contention yields the heavy straggling problem. Existing works feature contributions for the traditional data parallelism. They cannot work well for the new tensor parallelism, due to the dependency among split tensors. This paper is the first attempt on accelerating heterogeneous tensor parallelism. We summarize specific challenges, including the very frequent synchronizations and the heavy tensor computation workloads. Our solution is to resize dimensions of parameters on demand, to quickly and dynamically balance workloads. The accuracy loss is reduced through priority resizing. We also migrate workloads between tasks, without any loss of accuracy. The most efficient communication primitives are selected and then scheduled in a non-redundant manner, to reduce the runtime latency. Our final hybrid solution is built on top of resizing and migration. By studying the tradeoff between accuracy and efficiency, it can smartly hit the “sweet spot”. Extensive experiments validate the effectiveness of our proposals.","PeriodicalId":13106,"journal":{"name":"IEEE Transactions on Big Data","volume":"12 3","pages":"994-1005"},"PeriodicalIF":8.1,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148506441","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"TERM Model: Tensor Ring Mixture Model for Density Estimation","authors":"Ruituo Wu;Jiani Liu;Bing Li;Anh-Huy Phan;Ivan V. Oseledets;Ce Zhu;Yipeng Liu","doi":"10.1109/TBDATA.2025.3648288","DOIUrl":"https://doi.org/10.1109/TBDATA.2025.3648288","url":null,"abstract":"Probabilistic modeling is a core challenge in statistical machine learning. Tensor-based probabilistic graph methods address interpretability and stability concerns encountered in neural network approaches and allow tractable inference (e.g., marginal inference and conditional inference). In this paper, we introduce tensor ring decomposition for density estimation, which reduces the number of permutation candidates compared to existing methods, while simultaneously enhancing expressive power and maintaining tractable inference. Different non-negative strategies for density function results in two variants: Born TRDE offers simpler inference and sampling but with slightly lower accuracy, while Energy TRDE, though more complex, achieves superior performance. Furthermore, a mixture model that incorporates multiple permutation candidates with adaptive weights is designed, resulting in increased expressive flexibility and comprehensiveness. Unlike existing methods that focus on finding a single optimal permutation, our approach, inspired by ensemble learning, demonstrates that combining multiple suboptimal permutations can yield superior results. Experiments demonstrate that the proposed approach excels in estimating probability density functions and sampling, capturing intricate details with competitive or superior performance compared to existing state-of-the-art (SOTA) tractable density methods.","PeriodicalId":13106,"journal":{"name":"IEEE Transactions on Big Data","volume":"12 3","pages":"893-908"},"PeriodicalIF":8.1,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148508349","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Survey on Federated Unlearning: Challenges and Opportunities","authors":"Hyejun Jeong;Shiqing Ma;Amir Houmansadr","doi":"10.1109/TBDATA.2026.3668538","DOIUrl":"https://doi.org/10.1109/TBDATA.2026.3668538","url":null,"abstract":"Federated learning (FL), introduced in 2017, enables collaborative learning across mutually distrusting parties without sharing raw data, enabling privacy-preserving model training. However, emerging regulations and practical demands require models to be able to <italic>forget</i> learned data, leading to growing interest in Machine Unlearning (MU). In the context of FL, many techniques developed for unlearning in centralized settings are not trivially applicable. This is due to interactivity, stochasticity, heterogeneity, and limited data accessibility. This has motivated a distinct research area of <italic>federated unlearning</i> (FU). This survey provides the first <italic>practice-oriented synthesis</i> of FU, foregrounding aspects often overlooked in prior surveys: data-distribution modeling (and non-IID simulation), dataset selection, and FL system configurations. We introduce a taxonomy that separates influence removal from performance recovery, and compare FU approaches across unlearning targets, aggregation assumptions, and reproducibility signals. By analyzing datasets, evaluation metrics, and code availability, we surface key trends that differentiate FU from centralized MU. We highlight challenges unique to FU, such as interactive training, stochastic client participation, and data isolation, and synthesize open problems around scalability, fairness, and unlearning in foundation models. Our survey aims to guide FU research by clarifying methodological gaps, system assumptions, and promising directions.","PeriodicalId":13106,"journal":{"name":"IEEE Transactions on Big Data","volume":"12 3","pages":"702-720"},"PeriodicalIF":8.1,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148509697","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}