Pub Date : 2025-11-28DOI: 10.1109/TKDE.2025.3638343
Miaomiao Cai;Lei Chen;Yifan Wang;Zhiyong Cheng;Min Zhang;Meng Wang
Popularity bias is a common challenge in recommender systems. It often causes unbalanced item recommendation performance and intensifies the Matthew effect. Due to limited user-item interactions, unpopular items are frequently constrained to the embedding neighborhoods of only a few users, leading to representation collapse and weakening the model’s generalization. Although existing supervised alignment and reweighting methods can help mitigate this problem, they still face two major limitations: (1) they overlook the inherent variability among different Graph Convolutional Networks (GCNs) layers, which can result in negative gains in deeper layers; (2) they rely heavily on fixed hyperparameters to balance popular and unpopular items, limiting adaptability to diverse data distributions and increasing model complexity. To address these challenges, we propose Graph-Structured Dual Adaptation Framework (GSDA), a dual adaptive framework for mitigating popularity bias in recommendation. Our theoretical analysis shows that supervised alignment in GCNs is hindered by the over-smoothing effect, where the distinction between popular and unpopular items diminishes as layers deepen, reducing the effectiveness of alignment at deeper levels. To overcome this limitation, GSDA integrates a hierarchical adaptive alignment mechanism that counteracts entropy decay across layers together with a distribution-aware contrastive weighting strategy based on the Gini coefficient, enabling the model to adapt its debiasing strength dynamically without relying on fixed hyperparameters. Extensive experiments on three benchmark datasets demonstrate that GSDA effectively alleviates popularity bias while consistently outperforming state-of-the-art methods in recommendation performance.
{"title":"Graph-Structured Driven Dual Adaptation for Mitigating Popularity Bias","authors":"Miaomiao Cai;Lei Chen;Yifan Wang;Zhiyong Cheng;Min Zhang;Meng Wang","doi":"10.1109/TKDE.2025.3638343","DOIUrl":"https://doi.org/10.1109/TKDE.2025.3638343","url":null,"abstract":"Popularity bias is a common challenge in recommender systems. It often causes unbalanced item recommendation performance and intensifies the Matthew effect. Due to limited user-item interactions, unpopular items are frequently constrained to the embedding neighborhoods of only a few users, leading to representation collapse and weakening the model’s generalization. Although existing supervised alignment and reweighting methods can help mitigate this problem, they still face two major limitations: (1) they overlook the inherent variability among different Graph Convolutional Networks (GCNs) layers, which can result in negative gains in deeper layers; (2) they rely heavily on fixed hyperparameters to balance popular and unpopular items, limiting adaptability to diverse data distributions and increasing model complexity. To address these challenges, we propose <italic><b><u>G</u>raph-<u>S</u>tructured <u>D</u>ual <u>A</u>daptation Framework (GSDA)</b></i>, a dual adaptive framework for mitigating popularity bias in recommendation. Our theoretical analysis shows that supervised alignment in GCNs is hindered by the over-smoothing effect, where the distinction between popular and unpopular items diminishes as layers deepen, reducing the effectiveness of alignment at deeper levels. To overcome this limitation, <italic>GSDA</i> integrates a hierarchical adaptive alignment mechanism that counteracts entropy decay across layers together with a distribution-aware contrastive weighting strategy based on the Gini coefficient, enabling the model to adapt its debiasing strength dynamically without relying on fixed hyperparameters. Extensive experiments on three benchmark datasets demonstrate that <italic>GSDA</i> effectively alleviates popularity bias while consistently outperforming state-of-the-art methods in recommendation performance.","PeriodicalId":13496,"journal":{"name":"IEEE Transactions on Knowledge and Data Engineering","volume":"38 2","pages":"1129-1143"},"PeriodicalIF":10.4,"publicationDate":"2025-11-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145898182","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}
Spatio-temporal traffic data imputation is a fundamental component in intelligent transportation systems, which can significantly improve data quality and enhance the accuracy of downstream data mining tasks. Recently, low-rank tensor representation has shown great potential for spatio-temporal traffic data imputation. However, the low-rank assumption focuses on the global structure, neglecting the critical spatial topology and local temporal dependencies inherent in spatio-temporal data. To address these issues, we propose a topology-induced low-rank tensor representation (TILR), which can accurately capture the underlying low-rankness of the spatial multi-scale features induced by topology knowledge. Moreover, to exploit local temporal dependencies, we suggest a learnable convolutional regularization framework, which not only includes some classical convolution-based regularizers but also leads to the discovery of new convolutional regularizers. Equipped with the suggested TILR and convolutional regularizer, we build a unified low-rank tensor model harmonizing spatial topology and temporal dependencies for traffic data imputation, which is expected to deliver promising performance even under extreme and complex missing scenarios. To solve the proposed nonconvex model, we develop an efficient alternating direction method of multipliers (ADMM)-based algorithm and analyze its computational complexity. Extensive experiments demonstrate that the proposed model outperforms state-of-the-art baselines for various missing scenarios. These results reveal the critical synergy between topology-aware low-rank constraint and temporal dynamic modeling for spatio-temporal data imputation.
{"title":"Topology-Induced Low-Rank Tensor Representation for Spatio-Temporal Traffic Data Imputation","authors":"Zhi-Long Han;Ting-Zhu Huang;Xi-Le Zhao;Ben-Zheng Li;Meng Ding","doi":"10.1109/TKDE.2025.3638633","DOIUrl":"https://doi.org/10.1109/TKDE.2025.3638633","url":null,"abstract":"Spatio-temporal traffic data imputation is a fundamental component in intelligent transportation systems, which can significantly improve data quality and enhance the accuracy of downstream data mining tasks. Recently, low-rank tensor representation has shown great potential for spatio-temporal traffic data imputation. However, the low-rank assumption focuses on the global structure, neglecting the critical spatial topology and local temporal dependencies inherent in spatio-temporal data. To address these issues, we propose a topology-induced low-rank tensor representation (TILR), which can accurately capture the underlying low-rankness of the spatial multi-scale features induced by topology knowledge. Moreover, to exploit local temporal dependencies, we suggest a learnable convolutional regularization framework, which not only includes some classical convolution-based regularizers but also leads to the discovery of new convolutional regularizers. Equipped with the suggested TILR and convolutional regularizer, we build a unified low-rank tensor model harmonizing spatial topology and temporal dependencies for traffic data imputation, which is expected to deliver promising performance even under extreme and complex missing scenarios. To solve the proposed nonconvex model, we develop an efficient alternating direction method of multipliers (ADMM)-based algorithm and analyze its computational complexity. Extensive experiments demonstrate that the proposed model outperforms state-of-the-art baselines for various missing scenarios. These results reveal the critical synergy between topology-aware low-rank constraint and temporal dynamic modeling for spatio-temporal data imputation.","PeriodicalId":13496,"journal":{"name":"IEEE Transactions on Knowledge and Data Engineering","volume":"38 2","pages":"1349-1363"},"PeriodicalIF":10.4,"publicationDate":"2025-11-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145898220","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}
Traffic prediction is essential for modern transportation systems, enhancing traffic management and urban planning. Accurate predictions of traffic flow and speed are crucial for understanding road usage, mitigating congestion, and providing real-time traffic monitoring and dynamic route guidance, thus improving road safety and infrastructure efficiency. Traditional research has often focused on predicting traffic flow or speed independently, leading to higher resource consumption due to the need for separate models. Few studies have explored the simultaneous prediction of both metrics, with recent attempts failing to account for spatial correlations, resulting in suboptimal performance. To address these challenges, we propose MTNet, a multi-task learning framework for joint traffic flow and speed prediction. MTNet employs a Transformer-like Encoder-Decoder architecture to process and enhance feature representations, capturing complex spatio-temporal correlations. Specifically, MTNet extracts intra-task dependencies using a cross-task interaction module and models task-specific spatiotemporal dependencies using spatial and temporal-aware modules with cascaded residual structures. Additionally, spatio-temporal positional encoding is integrated to increase awareness of long-term and long-distance dependencies. Extensive experiments on three diverse traffic datasets—Manchester, PeMSD4, and PeMSD8—demonstrate that MTNet significantly outperforms state-of-the-art methods in both traffic flow and speed prediction. MTNet achieves substantial improvements in prediction accuracy and efficiency, striking an optimal balance between performance and computational resource usage.
{"title":"MTNet: A Multi-Task Learning Framework That Integrates Intra-Task and Task-Specific Dependencies for Traffic Forecasting","authors":"Shaokun Zhang;Rui Wang;Hongjun Tang;Kaizhong Zuo;Peng Jiang;Peng Hu;Wenjie Li;Biao Jie;Peize Zhao","doi":"10.1109/TKDE.2025.3638147","DOIUrl":"https://doi.org/10.1109/TKDE.2025.3638147","url":null,"abstract":"Traffic prediction is essential for modern transportation systems, enhancing traffic management and urban planning. Accurate predictions of traffic flow and speed are crucial for understanding road usage, mitigating congestion, and providing real-time traffic monitoring and dynamic route guidance, thus improving road safety and infrastructure efficiency. Traditional research has often focused on predicting traffic flow or speed independently, leading to higher resource consumption due to the need for separate models. Few studies have explored the simultaneous prediction of both metrics, with recent attempts failing to account for spatial correlations, resulting in suboptimal performance. To address these challenges, we propose MTNet, a multi-task learning framework for joint traffic flow and speed prediction. MTNet employs a Transformer-like Encoder-Decoder architecture to process and enhance feature representations, capturing complex spatio-temporal correlations. Specifically, MTNet extracts intra-task dependencies using a cross-task interaction module and models task-specific spatiotemporal dependencies using spatial and temporal-aware modules with cascaded residual structures. Additionally, spatio-temporal positional encoding is integrated to increase awareness of long-term and long-distance dependencies. Extensive experiments on three diverse traffic datasets—Manchester, PeMSD4, and PeMSD8—demonstrate that MTNet significantly outperforms state-of-the-art methods in both traffic flow and speed prediction. MTNet achieves substantial improvements in prediction accuracy and efficiency, striking an optimal balance between performance and computational resource usage.","PeriodicalId":13496,"journal":{"name":"IEEE Transactions on Knowledge and Data Engineering","volume":"38 2","pages":"1206-1220"},"PeriodicalIF":10.4,"publicationDate":"2025-11-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145898241","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}
Pub Date : 2025-11-27DOI: 10.1109/TKDE.2025.3634839
Zhengwei Tao;Xiancai Chen;Zhi Jin;Xiaoying Bai;Haiyan Zhao;Wenpeng Hu;Chongyang Tao;Shuai Ma
Event reasoning is to reason with events and certain inter-event relations. These cutting-edge techniques possess crucial and fundamental capabilities that underlie various applications. Large language models (LLMs) have made advances in event reasoning owing to their wealth of training. However, the LLMs commonly used today still do not consistently demonstrate proficiency in managing event reasoning as humans. This discrepancy arises from not explicitly modeling events and their relations and insufficient knowledge of event relations. In addition, the different reasoning paradigms of the LLMs are trained in an imbalanced way. In this paper, we propose $textsc {WizardEvent}$, to synthesize data from the unlabeled corpus with the proposed hybrid event-aware instruction tuning. Specifically, we first represent the events and their relation in a novel structure and then extract the knowledge from the raw text. Second, we introduce hybrid event reasoning paradigms with four reasoning formats. Lastly, we wrap our constructed event relational knowledge with the paradigms to create the instruction tuning dataset. We fine-tune the model with this enriched dataset, significantly improving the event reasoning. The performance of $textsc {WizardEvent}$ is rigorously evaluated through extensive experiments. The results demonstrate that $textsc {WizardEvent}$ substantially outperforms baselines, indicating the effectiveness of our approach.
{"title":"WizardEvent: Empowering Event Reasoning by Hybrid Event-Aware Data Synthesizing","authors":"Zhengwei Tao;Xiancai Chen;Zhi Jin;Xiaoying Bai;Haiyan Zhao;Wenpeng Hu;Chongyang Tao;Shuai Ma","doi":"10.1109/TKDE.2025.3634839","DOIUrl":"https://doi.org/10.1109/TKDE.2025.3634839","url":null,"abstract":"Event reasoning is to reason with events and certain inter-event relations. These cutting-edge techniques possess crucial and fundamental capabilities that underlie various applications. Large language models (LLMs) have made advances in event reasoning owing to their wealth of training. However, the LLMs commonly used today still do not consistently demonstrate proficiency in managing event reasoning as humans. This discrepancy arises from not explicitly modeling events and their relations and insufficient knowledge of event relations. In addition, the different reasoning paradigms of the LLMs are trained in an imbalanced way. In this paper, we propose <inline-formula><tex-math>$textsc {WizardEvent}$</tex-math></inline-formula>, to synthesize data from the unlabeled corpus with the proposed hybrid event-aware instruction tuning. Specifically, we first represent the events and their relation in a novel structure and then extract the knowledge from the raw text. Second, we introduce hybrid event reasoning paradigms with four reasoning formats. Lastly, we wrap our constructed event relational knowledge with the paradigms to create the instruction tuning dataset. We fine-tune the model with this enriched dataset, significantly improving the event reasoning. The performance of <inline-formula><tex-math>$textsc {WizardEvent}$</tex-math></inline-formula> is rigorously evaluated through extensive experiments. The results demonstrate that <inline-formula><tex-math>$textsc {WizardEvent}$</tex-math></inline-formula> substantially outperforms baselines, indicating the effectiveness of our approach.","PeriodicalId":13496,"journal":{"name":"IEEE Transactions on Knowledge and Data Engineering","volume":"38 2","pages":"1412-1426"},"PeriodicalIF":10.4,"publicationDate":"2025-11-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145898236","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}
Large-scale social networks can be modeled as decentralized graphs, where each node holds a part of the overall network. Local differential privacy (LDP) has been widely adopted in decentralized graph analysis to ensure privacy for individual nodes. However, existing LDP-based methods often fail to accommodate personalized privacy requirements due to their uniform encoding and equal perturbation mechanisms. To address this issue, we propose PEGS, a novel privacy-preserving decentralized graph synthesis approach that significantly improves utility while respecting user-specific privacy preferences. Specifically, we introduce interactive local differential privacy (iLDP), a new edge-level definition of LDP that relaxes the constraints of node-independent perturbation, thereby enabling the fulfillment of individual privacy needs. Furthermore, we develop a decentralized graph perturbation framework offering three levels of privacy settings. To optimize the balance between information preservation and privacy, we design encoding and perturbation mechanisms leveraging information entropy tailored to different privacy levels. Extensive experimental evaluations and rigorous theoretical analysis demonstrate that our method produces high-quality synthetic graphs while adhering to iLDP guarantees.
{"title":"PEGS: A Graph Synthesis Approach Based on Local Differential Privacy Preference","authors":"Lihe Hou;Weiwei Ni;Nan Fu;Dongyue Zhang;Ruyu Zhang","doi":"10.1109/TKDE.2025.3637324","DOIUrl":"https://doi.org/10.1109/TKDE.2025.3637324","url":null,"abstract":"Large-scale social networks can be modeled as decentralized graphs, where each node holds a part of the overall network. Local differential privacy (LDP) has been widely adopted in decentralized graph analysis to ensure privacy for individual nodes. However, existing LDP-based methods often fail to accommodate personalized privacy requirements due to their uniform encoding and equal perturbation mechanisms. To address this issue, we propose PEGS, a novel privacy-preserving decentralized graph synthesis approach that significantly improves utility while respecting user-specific privacy preferences. Specifically, we introduce interactive local differential privacy (iLDP), a new edge-level definition of LDP that relaxes the constraints of node-independent perturbation, thereby enabling the fulfillment of individual privacy needs. Furthermore, we develop a decentralized graph perturbation framework offering three levels of privacy settings. To optimize the balance between information preservation and privacy, we design encoding and perturbation mechanisms leveraging information entropy tailored to different privacy levels. Extensive experimental evaluations and rigorous theoretical analysis demonstrate that our method produces high-quality synthetic graphs while adhering to iLDP guarantees.","PeriodicalId":13496,"journal":{"name":"IEEE Transactions on Knowledge and Data Engineering","volume":"38 2","pages":"1236-1248"},"PeriodicalIF":10.4,"publicationDate":"2025-11-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145898205","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}
Utilizing pre-trained generative models for sentiment element extraction has recently significantly enhanced aspect-based sentiment analysis benchmarks. Nonetheless, these models have two significant drawbacks: 1) high-computational cost in both the inference time and hardware requirement. 2) Lack of explicit modeling as they model the connections between sentiment elements with fragile natural or notational language target sequence. To overcome these challenges, we present a novel opinion tree parsing model designed to swiftly parse sentiment elements from an opinion tree. This approach not only accelerates the process but also explicitly unveils a more comprehensive and fully articulated aspect-level sentiment structure. Our method begins by introducing a pioneering context-free opinion grammar to standardize the opinion tree structure. Subsequently, we leverage a neural chart-based opinion tree parser to thoroughly explore the interconnections among sentiment elements and parse them into a structured opinion tree. Extensive experiments underscore the effectiveness of our proposed model and the capability of the opinion tree parser, particularly when coupled with the introduced context-free opinion grammar. Crucially, the results confirm the superior speed of our model compared to the SOTA baselines.
{"title":"Exploring Context-Free Opinion Grammar for Aspect-Based Sentiment Analysis","authors":"Xiaoyi Bao;Jinghang Gu;Zhongqing Wang;Xiaotong Jiang;Chu-Ren Huang","doi":"10.1109/TKDE.2025.3632628","DOIUrl":"https://doi.org/10.1109/TKDE.2025.3632628","url":null,"abstract":"Utilizing pre-trained generative models for sentiment element extraction has recently significantly enhanced aspect-based sentiment analysis benchmarks. Nonetheless, these models have two significant drawbacks: 1) high-computational cost in both the inference time and hardware requirement. 2) Lack of explicit modeling as they model the connections between sentiment elements with fragile natural or notational language target sequence. To overcome these challenges, we present a novel opinion tree parsing model designed to swiftly parse sentiment elements from an opinion tree. This approach not only accelerates the process but also explicitly unveils a more comprehensive and fully articulated aspect-level sentiment structure. Our method begins by introducing a pioneering context-free opinion grammar to standardize the opinion tree structure. Subsequently, we leverage a neural chart-based opinion tree parser to thoroughly explore the interconnections among sentiment elements and parse them into a structured opinion tree. Extensive experiments underscore the effectiveness of our proposed model and the capability of the opinion tree parser, particularly when coupled with the introduced context-free opinion grammar. Crucially, the results confirm the superior speed of our model compared to the SOTA baselines.","PeriodicalId":13496,"journal":{"name":"IEEE Transactions on Knowledge and Data Engineering","volume":"38 2","pages":"1070-1083"},"PeriodicalIF":10.4,"publicationDate":"2025-11-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145898240","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}
Graph pre-training has been concentrated on graph-level tasks involving small graphs (e.g., molecular graphs) or learning node representations on a fixed graph. Extending graph pre-trained models to web-scale graphs with billions of nodes in industrial scenarios, while avoiding negative transfer across graphs or tasks, remains a challenge. We aim to develop a general graph pre-trained model with inductive ability that can make predictions for unseen new nodes and even new graphs. In this work, we introduce a scalable transformer-based graph pre-training framework called PGT (Pre-trained Graph Transformer). Based on the masked autoencoder architecture, we design two pre-training tasks: one for reconstructing node features and the other for reconstructing local structures. Unlike the original autoencoder architecture where the pre-trained decoder is discarded, we propose a novel strategy that utilizes the decoder for feature augmentation. Our framework, tested on the publicly available ogbn-papers100 M dataset with 111 million nodes and 1.6 billion edges, achieves state-of-the-art performance, showcasing scalability and efficiency. We have deployed our framework on Tencent’s online game data, confirming its capability to pre-train on real-world graphs with over 540 million nodes and 12 billion edges and to generalize effectively across diverse static and dynamic downstream tasks.
{"title":"Generalizing Graph Transformers Across Diverse Graphs and Tasks via Pre-Training","authors":"Yufei He;Zhenyu Hou;Yukuo Cen;Jun Hu;Feng He;Xu Cheng;Jie Tang;Bryan Hooi","doi":"10.1109/TKDE.2025.3632394","DOIUrl":"https://doi.org/10.1109/TKDE.2025.3632394","url":null,"abstract":"Graph pre-training has been concentrated on graph-level tasks involving small graphs (e.g., molecular graphs) or learning node representations on a fixed graph. Extending graph pre-trained models to web-scale graphs with billions of nodes in industrial scenarios, while avoiding negative transfer across graphs or tasks, remains a challenge. We aim to develop a general graph pre-trained model with inductive ability that can make predictions for unseen new nodes and even new graphs. In this work, we introduce a scalable transformer-based graph pre-training framework called PGT (Pre-trained Graph Transformer). Based on the masked autoencoder architecture, we design two pre-training tasks: one for reconstructing node features and the other for reconstructing local structures. Unlike the original autoencoder architecture where the pre-trained decoder is discarded, we propose a novel strategy that utilizes the decoder for feature augmentation. Our framework, tested on the publicly available ogbn-papers100 M dataset with 111 million nodes and 1.6 billion edges, achieves state-of-the-art performance, showcasing scalability and efficiency. We have deployed our framework on Tencent’s online game data, confirming its capability to pre-train on real-world graphs with over 540 million nodes and 12 billion edges and to generalize effectively across diverse static and dynamic downstream tasks.","PeriodicalId":13496,"journal":{"name":"IEEE Transactions on Knowledge and Data Engineering","volume":"38 2","pages":"1114-1128"},"PeriodicalIF":10.4,"publicationDate":"2025-11-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145898268","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}
Time series forecasting faces significant challenges due to non-stationary components that obscure underlying patterns. While Transformer-based models are effective at capturing stationary components, they struggle with non-stationary dynamics and multivariate dependencies. In this paper, we propose FreqEvo, a lightweight Frequency Domain Feature Enhancement module for time series forecasting. FreqEvo progressively filters frequency components from high to low amplitude, ensuring the preservation of informative features while reducing noise. By integrating recursive Fourier-based residual modeling and cross-domain attention, FreqEvo effectively refines low-amplitude frequency features and stabilizes the embeddings, outperforming traditional low-pass filtering and random frequency selection methods in capturing both short-term and long-term dependencies. Experimental results on benchmark datasets demonstrate that FreqEvo outperforms state-of-the-art (SOTA) models and serves as a plug-and-play module to enhance existing Long-Term Sequence Forecasting (LSTF) models.
{"title":"FreqEvo: Enhancing Time Series Forecasting With Multi-Level Frequency Domain Feature Extraction","authors":"Guohong Wang;Xianhan Tan;Zengming Lin;Binli Luo;Shangjian Zhong;Kele Xu","doi":"10.1109/TKDE.2025.3632365","DOIUrl":"https://doi.org/10.1109/TKDE.2025.3632365","url":null,"abstract":"Time series forecasting faces significant challenges due to non-stationary components that obscure underlying patterns. While Transformer-based models are effective at capturing stationary components, they struggle with non-stationary dynamics and multivariate dependencies. In this paper, we propose <italic>FreqEvo</i>, a lightweight Frequency Domain Feature Enhancement module for time series forecasting. <italic>FreqEvo</i> progressively filters frequency components from high to low amplitude, ensuring the preservation of informative features while reducing noise. By integrating recursive Fourier-based residual modeling and cross-domain attention, <italic>FreqEvo</i> effectively refines low-amplitude frequency features and stabilizes the embeddings, outperforming traditional low-pass filtering and random frequency selection methods in capturing both short-term and long-term dependencies. Experimental results on benchmark datasets demonstrate that <italic>FreqEvo</i> outperforms state-of-the-art (SOTA) models and serves as a plug-and-play module to enhance existing Long-Term Sequence Forecasting (LSTF) models.","PeriodicalId":13496,"journal":{"name":"IEEE Transactions on Knowledge and Data Engineering","volume":"38 2","pages":"1099-1113"},"PeriodicalIF":10.4,"publicationDate":"2025-11-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145898267","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}
Pub Date : 2025-11-13DOI: 10.1109/TKDE.2025.3631909
Yawen Li;Xiaobao Wang;Bin Wen;Di Jin;Junping Du
The COVID-19 pandemic not only triggered a global health crisis but also amplified public panic through the rapid spread of misinformation. Understanding public sentiment and identifying the causes of sudden sentiment spikes is therefore critical for ensuring accurate information dissemination and guiding effective policymaking. However, mining such causes from social media remains challenging. Tweets collected during sentiment spike periods are often short, noisy, and dominated by repetitive background topics, making it difficult for existing topic models to separate emerging issues from long-standing discussions. To address these challenges, we propose the Sentiment Variation-aware Emerging Topics Mining Model (SVETM), a probabilistic graphical framework that leverages user sentiment variation between adjacent time windows as a guiding signal to distinguish emerging topics from background content. We further reformulate inference as a maximum a posteriori (MAP) problem and develop an efficient variational inference algorithm for scalable learning. Extensive experiments on a large-scale COVID-19 Twitter dataset demonstrate that SVETM outperforms strong baselines in terms of topic coherence, interpretability, and its ability to uncover the underlying causes of sentiment spikes.
{"title":"Sentiment Variation-Aware Sentiment Spike Explanation During COVID-19 Epidemic","authors":"Yawen Li;Xiaobao Wang;Bin Wen;Di Jin;Junping Du","doi":"10.1109/TKDE.2025.3631909","DOIUrl":"https://doi.org/10.1109/TKDE.2025.3631909","url":null,"abstract":"The COVID-19 pandemic not only triggered a global health crisis but also amplified public panic through the rapid spread of misinformation. Understanding public sentiment and identifying the causes of sudden sentiment spikes is therefore critical for ensuring accurate information dissemination and guiding effective policymaking. However, mining such causes from social media remains challenging. Tweets collected during sentiment spike periods are often short, noisy, and dominated by repetitive background topics, making it difficult for existing topic models to separate emerging issues from long-standing discussions. To address these challenges, we propose the Sentiment Variation-aware Emerging Topics Mining Model (SVETM), a probabilistic graphical framework that leverages user sentiment variation between adjacent time windows as a guiding signal to distinguish emerging topics from background content. We further reformulate inference as a maximum a posteriori (MAP) problem and develop an efficient variational inference algorithm for scalable learning. Extensive experiments on a large-scale COVID-19 Twitter dataset demonstrate that SVETM outperforms strong baselines in terms of topic coherence, interpretability, and its ability to uncover the underlying causes of sentiment spikes.","PeriodicalId":13496,"journal":{"name":"IEEE Transactions on Knowledge and Data Engineering","volume":"38 2","pages":"1306-1318"},"PeriodicalIF":10.4,"publicationDate":"2025-11-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145898219","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}
Pub Date : 2025-11-12DOI: 10.1109/TKDE.2025.3632233
Runlin Lei;Haipeng Ding;Zhewei Wei
Graph Neural Networks (GNNs) have become widely popular across various applications, with their vulnerability to adversarial attacks being a key concern. Among the different types of graph attacks, Restricted Black-box Attacks (RBAs) present the most strict constraints, as attackers have limited access only to node features and graph structure. Existing RBAs rely on homophily assumptions or shift-based losses as their objectives to conduct structural perturbations, but we demonstrate that all the approaches fail on heterophilic graphs. To address this challenge, we introduce node-wise distance metrics as the objective to fundamentally quantify the quality of the graph structure after perturbations. Our theoretical results show that the proposed objective allows RBAs to effectively handle graphs beyond homophily. Leveraging this objective, we propose HetAttack, a scalable method that significantly reduces the distinguishability of nodes on the victim graph. Experiments on both synthetic and real-world graphs confirm the efficacy of HetAttack across varying levels of homophily, achieving performance comparable to split-unknown white-box attacks without prior knowledge of labels or the target model.
{"title":"Restricted Black-Box Attack on Graphs Beyond Homophily","authors":"Runlin Lei;Haipeng Ding;Zhewei Wei","doi":"10.1109/TKDE.2025.3632233","DOIUrl":"https://doi.org/10.1109/TKDE.2025.3632233","url":null,"abstract":"Graph Neural Networks (GNNs) have become widely popular across various applications, with their vulnerability to adversarial attacks being a key concern. Among the different types of graph attacks, Restricted Black-box Attacks (RBAs) present the most strict constraints, as attackers have limited access only to node features and graph structure. Existing RBAs rely on homophily assumptions or shift-based losses as their objectives to conduct structural perturbations, but we demonstrate that all the approaches fail on heterophilic graphs. To address this challenge, we introduce node-wise distance metrics as the objective to fundamentally quantify the quality of the graph structure after perturbations. Our theoretical results show that the proposed objective allows RBAs to effectively handle graphs beyond homophily. Leveraging this objective, we propose HetAttack, a scalable method that significantly reduces the distinguishability of nodes on the victim graph. Experiments on both synthetic and real-world graphs confirm the efficacy of HetAttack across varying levels of homophily, achieving performance comparable to split-unknown white-box attacks without prior knowledge of labels or the target model.","PeriodicalId":13496,"journal":{"name":"IEEE Transactions on Knowledge and Data Engineering","volume":"38 2","pages":"1292-1305"},"PeriodicalIF":10.4,"publicationDate":"2025-11-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145898257","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}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：