{"title":"Enhancing neural topic modeling for social media text via semantic bag of word clusters and log-domain Sinkhorn transport","authors":"Yi Sun, Junhao Zhao, Haoran Xu, Ronghua Zhang, Changzheng Liu, Limengzi Yuan","doi":"10.1016/j.ipm.2025.104411","DOIUrl":"10.1016/j.ipm.2025.104411","url":null,"abstract":"<div><div>Topic modeling has been widely applied to analyze text data from social media platforms. Under this scenario, traditional Neural Topic Models (NTMs) encounter three primary challenges: (1) initial text representation; (2) the long-tail nature of topic distributions in social network texts; (3) approximation of Optimal Transport. Motivated by these challenges, we propose an end-to-end solution spanning from text representation to topic modeling.</div><div>First, we propose SBoWC, a novel text representation method that performs dimensionality reduction while absorbing semantic information through base terms, achieved by combining word embeddings with clustering statistics. Subsequently, we propose GSWTM, a Wasserstein-based autoencoder topic model that fits the long-tail topic distribution in social network texts via Gamma priors and innovatively employs log-domain Sinkhorn to approximate Optimal Transport.</div><div>Ablation studies demonstrate the transferability and effectiveness of SBoWC in text representation. GSWTM demonstrates significantly better performance than baselines in TU, <span><math><msub><mrow><mi>C</mi></mrow><mrow><mi>V</mi></mrow></msub></math></span>, and the comprehensive metrics TQ across four real social network datasets of varying sizes. The log-domain Sinkhorn approximation exhibits excellent stability, allowing the regularization parameter <span><math><mi>ϵ</mi></math></span> to be reduced to 0.1–0.01, thereby approaching the original Optimal Transport.</div></div>","PeriodicalId":50365,"journal":{"name":"Information Processing & Management","volume":"63 2","pages":"Article 104411"},"PeriodicalIF":6.9,"publicationDate":"2025-09-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145158146","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Chenrui Mao , Kai Shuang , Jinyu Guo , Bing Qian , Yu Yang , Haoqing Li
{"title":"Cognition-aligned frequency filtering for sentence embeddings","authors":"Chenrui Mao , Kai Shuang , Jinyu Guo , Bing Qian , Yu Yang , Haoqing Li","doi":"10.1016/j.ipm.2025.104415","DOIUrl":"10.1016/j.ipm.2025.104415","url":null,"abstract":"<div><div>Learning better sentence embeddings that capture precise semantic plays an important role in Natural Language Processing (NLP). The Sentence Textual Similarity (STS) of embeddings reflects their semantic precision, as this task requires a direct comparison of semantic meanings in vector space. Thus, we focus on improving the ability of sentence embeddings to capture semantic similarity. From the perspective of human cognition, we identify a critical cognitive gap in frequency-domain semantic representation: while semantic information is distributed across all frequency components of embeddings, the human selective attention mechanism suggests that only specific frequency bands are utilized for semantic processing. This frequency-domain cognitive gap leads to semantic redundancy in machine-learned embeddings, which is particularly detrimental for tasks requiring redundancy-resistant representations. To bridge this gap, we propose a simple <strong>C</strong>ognition-<strong>A</strong>ligned <strong>F</strong>requency <strong>F</strong>iltering (CAFF) method for unsupervised embedding training on <span><math><mrow><mn>1</mn><msup><mrow><mn>0</mn></mrow><mrow><mn>6</mn></mrow></msup></mrow></math></span> sentences from Wikipedia. CAFF introduces a self-adaptive Frequency Filtering Unit (FFU) to modulate the frequency components of embedding. The FFU functions as a filtering mechanism that suppresses irrelevant components in embeddings to mitigate semantic redundancy. Extensive evaluations with SentEval show that our embeddings improve over the initial encoder by 2.33% on the STS task, achieving state-of-the-art performance. Additionally, our results demonstrate improved performance on both transfer and retrieval tasks.</div></div>","PeriodicalId":50365,"journal":{"name":"Information Processing & Management","volume":"63 2","pages":"Article 104415"},"PeriodicalIF":6.9,"publicationDate":"2025-09-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145219872","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Moralization-aware identity fusion for detecting violent radicalization in social media","authors":"Ming Yin , Miao Wan , Zihao Lin , Jijiao Jiang","doi":"10.1016/j.ipm.2025.104413","DOIUrl":"10.1016/j.ipm.2025.104413","url":null,"abstract":"<div><div>Most extremists use social media to spread radical ideologies. Existing methods for detecting violent radicalization primarily depend on superficial radicalization characteristics from user content or interactions, which fail to capture the role of morally-driven emotional shifts that underpin radicalization. These approaches overlook the dynamic convergence of moral emotions—a key facilitator of identity transformation and group allegiance. This paper proposes a novel property-specific method that incorporates moral frame, which is a systematic structure of moral reasoning rooted in foundational values such as care/harm, fairness/cheating, loyalty/betrayal, authority/subversion, and purity/degradation to model and fuse moralized identities for detecting violent radicalization. Specifically, it constructs a moral frame model integrated with user profiles to generate enhanced moralized representations, while an innovative moral identity fusion module employs heterogeneous graph neural networks to capture group moral homogeneity. A self-supervised Relational Graph Convolutional Networks (R-GCNs) clusters similar nodes to improve feature space discrimination, enabling effective violent radicalization detection through multi-module collaboration. Our experiments demonstrate significant performance across all metrics. On the moral foundation dataset, our method achieves 62.40 % accuracy, with 60.97 % macro F1 and 63.89 % weighted F1-scores, outperforming all baselines. Evaluation on datasets (340,310 tweets from Twitter; 188,358 posts from Gab) confirms the method’s detection performance. Case studies further validate its ability through moral convergence and social relationship patterns.</div></div>","PeriodicalId":50365,"journal":{"name":"Information Processing & Management","volume":"63 2","pages":"Article 104413"},"PeriodicalIF":6.9,"publicationDate":"2025-09-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145158149","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Min Teng , Chao Gao , Xianghua Li , Zhen Wang , Kefeng Fan , Vladimir Nekorkin
{"title":"Multi-scale graph contrastive learning for community detection in dynamic graphs","authors":"Min Teng , Chao Gao , Xianghua Li , Zhen Wang , Kefeng Fan , Vladimir Nekorkin","doi":"10.1016/j.ipm.2025.104410","DOIUrl":"10.1016/j.ipm.2025.104410","url":null,"abstract":"<div><div>Community detection in dynamic graphs is crucial for understanding the evolving relationships between entities in complex networks, such as social networks. These relationships change over time, driving the evolution of community structures. However, most existing research mainly focus on static graphs and cannot capture the evolution of community structures. Even when dynamic graphs are considered, they often rely on single-view graph information, leading to potential noise and suboptimal performance. To address these challenges, this paper proposes a new method called Multi-Scale Graph Contrastive Learning (MSGCL) for community detection in dynamic graphs. The MSGCL method first integrates the neighbor overlap similarity and topological structure similarity to enhance the graph features. Then, a multi-view graph representation learning module is proposed to learn local node representations and global graph representations of the graph, thereby addressing potential noise within single views. On this basis, a multi-scale contrastive learning module is proposed to improve the consistency and robustness of the node representations by leveraging both local-local and local-global contrastive learning. Finally, a Long Short-Term Memory (LSTM) module is incorporated to address the smooth transitions across time steps and achieve accurate community detection in dynamic graphs. Extensive experimental results demonstrate that MSGCL achieves superior performance across multiple datasets, with average improvements of 12.62% in NMI and 19.57% in ARI over the suboptimal method, outperforming the existing approaches. The code is available at <span><span>https://anonymous.4open.science/r/MSGCL-340C/</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":50365,"journal":{"name":"Information Processing & Management","volume":"63 2","pages":"Article 104410"},"PeriodicalIF":6.9,"publicationDate":"2025-09-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145158150","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Fudong Li , Yuxiang Zhou , Qinglai Yang , Yong Chen , Dell Zhang , Xuelong Li
{"title":"QuFiH: Hybrid low-bit quantization and block-level parameter efficient fine-tuning for video hashing","authors":"Fudong Li , Yuxiang Zhou , Qinglai Yang , Yong Chen , Dell Zhang , Xuelong Li","doi":"10.1016/j.ipm.2025.104408","DOIUrl":"10.1016/j.ipm.2025.104408","url":null,"abstract":"<div><div>Current video hashing methods lack large-model-based deep feature representation capabilities, resulting in suboptimal retrieval performance on large-scale video databases. To address this, we propose Video-HLM (Video-Hashing Large Model), built upon the pretrained Video-LLaMA. However, its computational demands hinder practical deployment. To balance efficiency and performance, we develop a novel video hashing framework that fuses hybrid low-bit <strong>Qu</strong>antization and block-level parameter-efficient <strong>Fi</strong>ne-tuning (QuFiH). Specifically, QuFiH employs <strong>4-bit quantization for the foundation model</strong> and <strong>2-bit quantization for the hash head</strong>, optimizing storage and performance. Additionally, we propose <strong>block-level LoRA/Propulsion</strong>, reducing redundant parameters while maintaining model expressiveness with lower computational overhead. Furthermore, we explore a <strong>“distill-then-finetune”</strong> strategy, combining knowledge distillation with the downstream task fine-tuning to enhance generalization and retrieval performance. Rich experiments on public datasets demonstrate QuFiH’s superiority. It compresses model parameters by <strong>6.84×</strong> while outperforming state-of-the-art hashing methods in mAP@100, even surpassing the full-precision Video-HLM. QuFiH achieves <strong>dual advancements in efficiency and accuracy</strong>, offering a practical solution for large-scale video retrieval in resource-constrained environments. <em>Code is available at:</em> <span><span><em>https://github.com/kydbj/QuFiH</em></span><svg><path></path></svg></span><em>.</em></div></div>","PeriodicalId":50365,"journal":{"name":"Information Processing & Management","volume":"63 2","pages":"Article 104408"},"PeriodicalIF":6.9,"publicationDate":"2025-09-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145158151","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Yubo Chen , Tong Zhou , Daojian Zeng , Sirui Li , Kang Liu , Jun Zhao
{"title":"ASDE: Low-budget text classification via active semi-supervised learning with debiasing training mechanism","authors":"Yubo Chen , Tong Zhou , Daojian Zeng , Sirui Li , Kang Liu , Jun Zhao","doi":"10.1016/j.ipm.2025.104390","DOIUrl":"10.1016/j.ipm.2025.104390","url":null,"abstract":"<div><div>Semi-supervised learning (SSL) is widely employed in text classification to address the challenges associated with limited labeled data availability. Nevertheless, current SSL methods often exhibit two significant limitations: they typically neglect the crucial process of selecting the initial labeled data effectively, and they fail to adequately mitigate the inherent bias that accumulates due to error propagation during the semi-supervised training phase. To address these shortcomings, we introduce an <strong>A</strong>ctive <strong>S</strong>emi-supervised learning framework with a <strong>DE</strong>biasing training mechanism (<strong>ASDE</strong>). Specifically, ASDE includes a novel task-aware cold-start active data selection component designed to establish a more representative and informative initial labeled set by leveraging task-specific information. Additionally, to combat the detrimental effects of error propagation, we develop a spatial interpolation debiasing mechanism integrated into the training process. Empirical results on four widely used text classification datasets demonstrate the substantial performance gains achieved by our proposed ASDE framework, particularly under low-budget conditions.</div></div>","PeriodicalId":50365,"journal":{"name":"Information Processing & Management","volume":"63 2","pages":"Article 104390"},"PeriodicalIF":6.9,"publicationDate":"2025-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145158148","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Zhonghao Xi , Bengong Yu , Chenyue Li , Haoyu Wang , Shanlin Yang
{"title":"MS-SMF: A probabilistic–causal multi-layer framework for image–text matching","authors":"Zhonghao Xi , Bengong Yu , Chenyue Li , Haoyu Wang , Shanlin Yang","doi":"10.1016/j.ipm.2025.104407","DOIUrl":"10.1016/j.ipm.2025.104407","url":null,"abstract":"<div><div>Image–text matching faces significant challenges due to the complex many-to-many semantic relationships between modalities, and existing methods remain deficient in both uncertainty modeling and causal robustness. To address these issues, we propose a multi-layer structural semantic matching framework (MS-SMF) that enhances cross-modal representations from a probabilistic–causal perspective. First, our method employs graph convolutional networks augmented with pseudo-coordinates and syntactic dependencies to precisely align image regions with text tokens, encoding each aligned pair as a Gaussian distribution to explicitly quantify semantic uncertainty. Next, these local Gaussian distributions are aggregated via structure-aware similarity weights into global Gaussian features, capturing holistic semantic context. Finally, we simulate causal interventions by randomly permuting the global mixture weights to generate counterfactual semantic distributions, and impose a contrastive loss between the original and counterfactual embeddings to enforce both invariance and discriminability. By synergizing the “soft” robustness of Gaussian modeling with the “hard” discriminative power of causal contrast, MS-SMF substantially improves the stability and discriminative capacity of cross-modal representations under structural perturbations. Experimental results on the MSCOCO and Flickr30K datasets demonstrate that our approach outperforms state-of-the-art methods — achieving an rSum of 542.9 on Flickr30K, with image-to-text R@1 and text-to-image R@1 improved by 1.0% and 2.3%, respectively — validating its superior performance in complex and ambiguous matching scenarios.</div></div>","PeriodicalId":50365,"journal":{"name":"Information Processing & Management","volume":"63 2","pages":"Article 104407"},"PeriodicalIF":6.9,"publicationDate":"2025-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145118581","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Deep time-series clustering via evolutionary learning and graph-based manifold learning","authors":"Hossein Abbasimehr , Ali Noshad","doi":"10.1016/j.ipm.2025.104409","DOIUrl":"10.1016/j.ipm.2025.104409","url":null,"abstract":"<div><div>Deep time series clustering (DTC) methods have recently gained attention, but they often suffer from imbalanced clusters, sensitivity to initialization, and local optima due to their reliance on the KL-divergence-based loss. To overcome this, we propose a novel Deep Evolutionary Time Series Clustering (DETC) method, which uses an evolutionary search process, generating diverse candidate solutions in each iteration. These solutions are evaluated using a fitness function based on internal validation metrics in both raw time series and latent spaces. This enhances robustness and avoids sub-optimal solutions, leading to more stable and accurate clustering. DETC employs autoencoders for latent representation, but without proper constraints, models may learn poor representations, resulting in local optima, slow convergence, instability, and sensitivity to noise. To learn discriminative representations, DETC introduces a graph-regularized loss that preserves the topological structure of time series in both latent and reconstructed spaces. We conducted extensive experiments on 15 diverse time series datasets, including varying sample sizes, cluster counts, and sequence lengths for a comprehensive assessment. Experimental results demonstrate that DETC significantly outperforms existing state-of-the-art DTC benchmarks, showing its superior clustering performance and robustness. Among 11 compared methods, DETC obtains an average rank of 1.47 and leads to an average improvement of 11% in NMI compared to the best benchmark model. The code and data are available at: <span><span>https://anonymous.4open.science/r/Deep-Evolutionary-Time-Series-Clustering-DETC-DD2D/README.md</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":50365,"journal":{"name":"Information Processing & Management","volume":"63 2","pages":"Article 104409"},"PeriodicalIF":6.9,"publicationDate":"2025-09-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145109519","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Learning path recommendation based on forgetting factors and knowledge graph awareness","authors":"Yunxia Fan , Mingwen Tong , Duantengchuan Li","doi":"10.1016/j.ipm.2025.104393","DOIUrl":"10.1016/j.ipm.2025.104393","url":null,"abstract":"<div><div>Learning path recommendation involves generating sequences of learning objects that are adapted to learners’ needs, goals, abilities, and other factors through recommendation algorithms. Reinforcement learning (RL) has become an important approach for this task; however, it primarily emphasizes recommending new knowledge concepts while neglecting the necessity of revisiting forgotten ones. To overcome this limitation, FKGRec is introduced as a learning path recommendation framework that incorporates forgetting factors and knowledge graph awareness. To address the forgetting problem, a novel method named MemGNN is proposed, which integrates forgetting and knowledge graph features and employs a graph neural network with a memory gate structure to predict both new and previously learned knowledge concepts at each learning step. To further optimize the sequencing of new and previously learned knowledge concepts, an action space is constructed based on knowledge concept prediction, taking learners’ cognitive states into account. An RL algorithm is then applied to recommend optimal learning paths by balancing new and previously learned knowledge concepts using a designed reward function. Experiments conducted on three datasets demonstrate that FKGRec surpasses existing state-of-the-art frameworks. A case analysis shows that the FKGRec framework can recommend learning paths that integrate new and previously learned knowledge concepts, aligned with learners’ current cognitive state and forgetting factors.</div></div>","PeriodicalId":50365,"journal":{"name":"Information Processing & Management","volume":"63 2","pages":"Article 104393"},"PeriodicalIF":6.9,"publicationDate":"2025-09-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145106101","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Navigating the perceived credibility and adoption of AI-generated review summaries in online shopping: An affordance perspective","authors":"Mingxia Jia , Yuxiang Zhao Chris , Xiaoyu Zhang","doi":"10.1016/j.ipm.2025.104404","DOIUrl":"10.1016/j.ipm.2025.104404","url":null,"abstract":"<div><div>As generative AI (GenAI) advances, e-commerce platforms increasingly leverage AI-generated review summaries to facilitate consumer decision-making. However, given the experience-driven nature of online review consumption, whether consumers perceive these summaries as credible, useful, and adoptable remains a key challenge to their effective implementation. Therefore, using affordance actualization theory, we conducted a scenario-based experiment and survey to analyze the quantitative data from 713 consumers (N_search product = 356, N_experience product = 357) regarding their perceptions of AI-generated review summaries. The findings show that functional affordances (algorithmic transparency, understandability, and convenience) and symbolic expressions (conveyed values and meanings) toward AI-generated review summaries play important roles in shaping consumers’ perceived credibility. Among them, algorithmic transparency, meaning conveyed, and understandability were identified as strong predictors. Perceived credibility further predicts perceived helpfulness, which, in turn, motivates users’ intentions to adopt AI-generated review summaries and contribute to consumer reviews. Interestingly, these influence pathways differ significantly depending on whether the product is a search or an experience product. This study provides an empirical investigation into the pathway from affordance to actualized belief and behavioral intention in the AI-generated review summaries context and offers practical insights for its effective application in AI-powered marketing.</div></div>","PeriodicalId":50365,"journal":{"name":"Information Processing & Management","volume":"63 2","pages":"Article 104404"},"PeriodicalIF":6.9,"publicationDate":"2025-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145106100","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}