IEEE Transactions on Computational Social Systems最新文献

{"title":"A Digital-Twin-Enabled Cyber-Physical Framework for Predictive Infection-Risk Monitoring and Secure Health-Data Surveillance","authors":"Dheeraj Kumar;Sandeep Kumar Sood;Keshav Singh Rawat;Sunil Prajapat","doi":"10.1109/TCSS.2026.3703478","DOIUrl":"https://doi.org/10.1109/TCSS.2026.3703478","url":null,"abstract":"Timely monitoring of airborne infectious diseases remains a major challenge for healthcare systems. Existing solutions rarely combine real-time sensing, secure data handling, predictive risk assessment, and network-level surveillance in a single framework. This work presents a cyber-physical system architecture for infection-risk monitoring organised into three layers: an IoT-assisted physical space, a digital-twin-enabled cyber-physical space, and a user-service layer. Data integrity across the layers is enforced by a reputation-based Byzantine fault tolerance mechanism over Ethereum nodes, and an XGBoost ensemble classifier performs symptom-driven severity stratification. The framework is evaluated as a controlled feasibility study on a real public COVID-19 dataset and, separately, on a synthetic H1N1 dataset generated only for feasibility testing, using 10-fold cross-validation and baseline comparison. It attains an average accuracy of 96.69%, recall of 93.37%, precision of 95.53%, an F-score of 93.20%, and a false-positive rate of 18.47%. The elevated false-positive rate is reported as a deployment constraint requiring decision-threshold tuning and clinical oversight. A NetLogo-based social-network-analysis module provides neighbourhood-level exposure awareness through a risk-probability-of-infection metric. The contribution is an integration-level framework; the reported numbers are simulation-based feasibility indicators and are not a claim of clinical diagnostic performance.","PeriodicalId":13044,"journal":{"name":"IEEE Transactions on Computational Social Systems","volume":"13 5","pages":"5733-5744"},"PeriodicalIF":4.6,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149097508","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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":"Production-Distribution Planning Considering Work Overtime and Multiuncertainties in Cloud Manufacturing","authors":"Zhongyi Chen;Xiaoping Li","doi":"10.1109/TCSS.2026.3711589","DOIUrl":"https://doi.org/10.1109/TCSS.2026.3711589","url":null,"abstract":"In cloud manufacturing (CMfg), participating enterprises lease several distributed warehouses in different locations to fulfil the demands of geographically distributed customers and schedule production to replenish the warehouses or directly fulfil customers. The wide distribution of customers and warehouses, along with the limited production and inventory capacities, pose significant challenges for cost-effective demand fulfilment and inventory control. In this article, we study a production-distribution planning problem considering work overtime to maximize the total profit for a CMfg manufacturing enterprise with several distributed warehouses. Work overtime can increase daily production with stochastic product quality decline. Stochastic demands from distributed customers are considered along with the consideration of stochastic delivery time. To describe the stochastic dynamic process of the considered problem, a Markov decision process (MDP)-based model is formulated. By integrating proximal policy optimization (PPO) with the rolling horizon method (RHM), a reinforcement learning (RL)-based method ReROP is proposed to maximize the total profit, where PPO is used for dynamic and intelligent decision-making, and RHM is employed for robust and stable stochastic optimization. The proposed method is evaluated by comparing it to other RL methods and existing methods for similar problems over a comprehensive set of numerical experiments. Experimental results demonstrate that the proposed method ReROP achieves higher convergence rewards and fewer convergence iterations than other RL methods, and it gains the highest profit as compared to existing methods for similar problems.","PeriodicalId":13044,"journal":{"name":"IEEE Transactions on Computational Social Systems","volume":"13 5","pages":"5919-5931"},"PeriodicalIF":4.6,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149097854","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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":"Blunt Giants, Sharp Specialists: Decoding Stress and Triggers Beyond General-Purpose LLMs","authors":"Soumitra Ghosh;Gopendra Vikram Singh;Khanjan Pathak;Asif Ekbal","doi":"10.1109/TCSS.2026.3686713","DOIUrl":"https://doi.org/10.1109/TCSS.2026.3686713","url":null,"abstract":"The United Nations sustainable development goals emphasize mental health as essential for inclusive and sustainable development, making <italic>stress</i>, an important factor affecting well-being and societal participation, a critical area of study. Leveraging social media data, we propose the emotion-aware stress cause analysis framework (ESCAF), which jointly models stress detection (SD) and stress cause identification (SCI). ESCAF employs a contextual attention block (CAB) to prioritize stress-relevant context and incorporates sentence-level emotion information for enriched contextual understanding. A key contribution is the stress and cause recognition dataset, derived from the Dreaddit corpus, providing utterance-level annotations for emotion and stress-triggering sentences within each post. ESCAF surpasses in-domain state-of-the-art methods with average F1 gains of +5.23% for SD and +6.57% for SCI. For comprehensive evaluation, we also assess fine-tuned and zero-shot performances of leading large language models (LLMs); ESCAF outperforms larger models including Llama, Mistral, Qwen, and Phi, while maintaining a smaller parameter footprint. These findings highlight the need for task-specific architectures requiring nuanced emotional understanding and caution against over-reliance on generic LLMs in sensitive applications.","PeriodicalId":13044,"journal":{"name":"IEEE Transactions on Computational Social Systems","volume":"13 5","pages":"6747-6757"},"PeriodicalIF":4.6,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149097877","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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":"Pairwise Uncertainty-Based Graph Contrastive Learning","authors":"Chengcheng Xu;Tianfeng Wang;Bin Sun;Xiao Han;Jiaxuan Geng;Zhisong Pan","doi":"10.1109/TCSS.2026.3713233","DOIUrl":"https://doi.org/10.1109/TCSS.2026.3713233","url":null,"abstract":"Currently, vanilla graph contrastive learning (GCL) has become the mainstream paradigm for learning discriminative node representations in unsupervised settings. However, its performance is limited by two issues. First, traditional graph augmentation strategies lack awareness of edge semantic relationships, which may cause uncertain signals to propagate along the graph structure. Second, the model pushes away all negative samples equally, ignoring the rich discriminative information embedded in highly uncertain negative pairs. To address these issues, we propose a pairwise uncertainty-based GCL method named PUGCL. Specifically, PUGCL divides affinity samples and estrangement samples for each anchor by fitting similarity distributions, and an uncertainty estimator is trained in a pairwise manner. Subsequently, the estimated uncertainty is used to guide the design of graph augmentation and contrastive loss. In this process, PUGCL adopts a differentiated treatment. Edges with high uncertainty are removed to purify the information propagation signal, while negative pairs with high uncertainty are assigned higher weights to refine the decision boundary. Extensive experiments on node classification and node clustering across six datasets are conducted, and the results demonstrate that PUGCL outperforms state-of-the-art methods.","PeriodicalId":13044,"journal":{"name":"IEEE Transactions on Computational Social Systems","volume":"13 5","pages":"5891-5906"},"PeriodicalIF":4.6,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149097892","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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":"Meta-Adaptive Continual Learning for Dynamic Social Communication Graphs","authors":"Monika Varshney;Tanima Dutta","doi":"10.1109/TCSS.2026.3671632","DOIUrl":"https://doi.org/10.1109/TCSS.2026.3671632","url":null,"abstract":"The rapid and continuous evolution of user interactions in social communication systems such as social media platforms, messaging applications, and collaborative content networks creates highly dynamic graph structures. These evolving environments present a significant challenge for accurately modeling and understanding user behaviors, which is essential for tasks such as personalized recommendations, community detection, and social influence analysis. Traditional graph neural networks (GNNs) are designed for static graphs and struggle in dynamic settings, suffering from catastrophic forgetting as new information overwrites previously learned knowledge. Given the critical need for adaptive models that can capture the complexities of real-time, evolving social graphs, we propose Comm-CLGNN, a novel communication-aware continual learning GNN. Comm-CLGNN introduces a dual-strategy approach: a meta-adaptive task modeling module that enables fast adaptation to evolving graph patterns, and a competitive memory selection mechanism that preserves task-specific knowledge over time without relying on replay buffers or complete retraining. Additionally, we apply a lightweight L2 regularization on task-specific parameters to discourage abrupt changes to useful components, further reducing interference with previously learned representations. We evaluate <sc>Comm-CLGNN</small> on multiple real-world benchmarks and show that it achieves consistently stronger average performance and competitive forgetting behavior across datasets. For example, <sc>Comm-CLGNN</small> attains <inline-formula> <tex-math>$94.89%$</tex-math> </inline-formula> AP with 0.03 average forgetting on CoraFull, and on Email-EU-core it outperforms strong baselines with backward transfer of <inline-formula> <tex-math>$-0.3$</tex-math> </inline-formula> and forward transfer of <inline-formula> <tex-math>$+2.0$</tex-math> </inline-formula>.","PeriodicalId":13044,"journal":{"name":"IEEE Transactions on Computational Social Systems","volume":"13 5","pages":"6228-6240"},"PeriodicalIF":4.6,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149097002","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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":"Double Whammy: Stealthy Data Manipulation-Aided Reconstruction Attack on Graph Federated Learning","authors":"Jinyin Chen;Chengyi Wang;Minying Ma;Haibin Zheng;Qi Xuan","doi":"10.1109/TCSS.2026.3709286","DOIUrl":"https://doi.org/10.1109/TCSS.2026.3709286","url":null,"abstract":"Graph federated learning (GFL) is one of the effective distributed learning paradigms for training graph neural network (GNN) on isolated graph data. It perfectly addresses the issue that GNN requires a large amount of labeled graph data without original data sharing. Unfortunately, recent research has constructed a successful graph reconstruction attack (GRA) on GFL. However, these attacks are still challenged in aspects of effectiveness and stealth. To address the issues, we propose the first <underline>D</u>ata <underline>Man</u>ipulation-aided <underline>Rec</u>onstruction attack on GFL, dubbed as <italic>DMan4Rec</i>. The malicious client is born to manipulate its locally collected data to enhance graph stealing privacy from benign ones, so as to construct a double whammy on GFL. It differs from previous work in three terms: 1) <italic>effectiveness</i> - to fully utilize the sparsity and feature smoothness of the graph, novel penalty terms are designed adaptive to diverse similarity functions for connected and unconnected node pairs, as well as incorporation label smoothing on top of the original cross-entropy loss; 2) <italic>scalability</i> - DMan4Rec is capable of both white-box and black-box attacks via training a supervised model to infer the posterior probabilities obtained from limited queries; 3) <italic>stealthiness</i> - by manipulating the malicious client’s node features, it can maintain the overall graph structure’s invariance and conceal the attack. Comprehensive experiments on four real datasets and three GNN models demonstrate that DMan4Rec achieves the state-of-the-art (SOTA) performance, e.g., the attack AUC and precision improved by 9.2% and 10.5%, respectively, compared with the SOTA. Particularly, DMan4Rec achieves an AUC score and a precision score of up to 99.59% and 99.56%, respectively, in black-box setting. Nevertheless, the complete overlap of the distribution graphs supports the stealthiness of the attack. Besides, DMan4Rec still beats the defensive GFL, which alarms a new threat to GFL.","PeriodicalId":13044,"journal":{"name":"IEEE Transactions on Computational Social Systems","volume":"13 5","pages":"5674-5688"},"PeriodicalIF":4.6,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149097086","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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":"Rationality Meets Irrationality: Incentivizing Dual Cognitive Loop for LLM-Based Agent in Social Simulation","authors":"Qun Ma;Xiao Xue;Xuwen Zhang;Zihan Zhao;Hui Yu;Juanjuan Li","doi":"10.1109/TCSS.2026.3710148","DOIUrl":"https://doi.org/10.1109/TCSS.2026.3710148","url":null,"abstract":"Social simulation in artificial intelligence for social science requires agents that capture both rational decision-making and human-like behavioral dynamics. However, existing large language model (LLM)-based agents often suffer from an emotional integration gap, a separation between rational and irrational processes, and memory systems lacking emotional context. To address these challenges, we propose dual cognitive loop (DCL)-Agent, a DCL framework that integrates emotion-driven (system 1) and utility-based (system 2) processes via dynamic weighting, supported by a three-tier memory architecture. Emotions modulate goal selection, plan evaluation, and memory retrieval, enabling more adaptive and socially plausible behavior. Experiments across multiple simulated scenarios demonstrate that DCL-Agent improves success rates by 5%–25% and reduces completion time by up to 30% compared with strong baselines. Ablation studies further highlight the complementary roles of emotion, memory, and rationality.","PeriodicalId":13044,"journal":{"name":"IEEE Transactions on Computational Social Systems","volume":"13 5","pages":"5539-5550"},"PeriodicalIF":4.6,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149097156","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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":"A More General Lightweight Network for Urban Regional Traffic Prediction","authors":"Jie Yin;Yue Hou;KeKe Lu","doi":"10.1109/TCSS.2025.3630657","DOIUrl":"https://doi.org/10.1109/TCSS.2025.3630657","url":null,"abstract":"With the increasing complexity of metropolitan management, the importance of urban regional traffic prediction has become increasingly prominent. However, existing traffic prediction models are often limited to specific regional traffic scenarios, lacking sufficient generalization ability. To address this issue, we propose a more generalizable lightweight spatiotemporal neural network (GL-STNN) for urban regional traffic prediction. By leveraging a fully convolutional design, GL-STNN circumvents the scalability issues of graph convolutional networks and the quadratic complexity of Transformers in dense spatiotemporal modeling. This network designs a novel deep convolutional architecture that adaptively integrates and captures global and local spatiotemporal correlations and heterogeneities by combining dilated convolutions and gating mechanisms. Furthermore, a multilevel grouped fusion convolutional architecture is designed to effectively fuse and align feature dimensions. To further enhance prediction accuracy, we also design a global masking mechanism to mark blank areas in the city. Experimental evaluations on a comprehensive benchmark of eight public datasets, spanning international cities and diverse traffic domains, demonstrate that GL-STNN outperforms existing baseline methods in terms of both performance and computational efficiency.","PeriodicalId":13044,"journal":{"name":"IEEE Transactions on Computational Social Systems","volume":"13 5","pages":"6571-6583"},"PeriodicalIF":4.6,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149097228","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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":"CEE-NeRF: Dynamic Neural Radiance Fields With Contrastive Enhanced Expression Representation for 4-D Face Reconstruction","authors":"Haowen Yu;Hui Yu;Rachael Jack;Yiming Wang;Philippe G. Schyns","doi":"10.1109/TCSS.2026.3715316","DOIUrl":"https://doi.org/10.1109/TCSS.2026.3715316","url":null,"abstract":"The reconstruction of photorealistic 4-D faces from monocular video holds significant importance for applications such as virtual reality and digital human-robot interaction. However, achieving reconstructions with structurally reasonable facial geometry and natural expressions remains challenging. Existing approaches typically combine dynamic neural radiance fields (NeRFs) with low-dimensional parametric models, such as 3-D morphable model (3DMM), for facial reconstruction and expression control. Nevertheless, 3DMM-based expression features often lack the ability to preserve high-frequency details and subtle individual variations. To address this issue, we propose contrastive expression enhancement (CEE)-NeRF, a novel framework that integrates geometry-aware 3DMM features with high-level semantic features through contrastive learning for complementary feature fusion. This design enhances the representation of fine-grained facial details while mitigating expression artifacts. In addition, unlike previous studies, we propose reformulating monocular 4-D face reconstruction as a multimodal learning problem and replacing the conventional multilayer perceptron used in NeRF with a mixture-of-experts Transformer, which enables modality-specific expert processing for accurate density and color estimation. This multimodal framework is also highly extensible and can be easily adapted to additional modalities. Extensive experiments demonstrate that CEE-NeRF achieves high-fidelity reconstructions from challenging monocular videos, consistently outperforming state-of-the-art approaches in rendering quality.","PeriodicalId":13044,"journal":{"name":"IEEE Transactions on Computational Social Systems","volume":"13 5","pages":"5608-5619"},"PeriodicalIF":4.6,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149097595","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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":"Guest Editorial: Special Issue on Mental Disorder Detection on Social Media","authors":"Qing Li;Nancy Xiaonan Yu;Haoyang Li;Xiangmeng Wang","doi":"10.1109/TCSS.2026.3731006","DOIUrl":"https://doi.org/10.1109/TCSS.2026.3731006","url":null,"abstract":"","PeriodicalId":13044,"journal":{"name":"IEEE Transactions on Computational Social Systems","volume":"13 5","pages":"6700-6703"},"PeriodicalIF":4.6,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11717831","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149097767","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}