Explainable recommendation has attracted great attention due to its capability of enhancing user trust and satisfaction. Users’ curiosities highly affect the recommendation accuracy and the effectiveness of explanations. Different target users have different levels of curiosities, while the curiosity of the same user changes dynamically. However, existing techniques cannot capture users’ dynamic curiosities from the historical user-item interactions for effective explainable recommendation. In this article, we propose a novel explainable recommendation approach for effective D ynamic C uriosity-aware E xplainable R ecommendation (DCER). Specifically, we first propose a novel multi-view representation learning to model the temporal user-item interactions. Then, we propose a new curiosity-enhanced recommendation to dynamically capture users’ curiosities, which improves the recommendation quality in a mutual promotion manner. Finally, we propose an adaptive rule-guided hybrid explanation generation strategy that enables more personalized explanations and well reflects the users’ dynamic psychological states behind the transactions. The experimental results demonstrate the high effectiveness of our proposed model.

Fairness-aware recommendation has emerged as a pivotal research area in recent years. Current fairness studies primarily examine two independent dimensions: user-side fairness and item-side fairness. However, most approaches address each side's fairness in isolation while neglecting their complex interdependencies. In this paper, we propose an LLM-Enhanced DiscriminAtive Counterfactual Debiasing Model for Two-sided Fairness in Recommendation (LeadFairRec). Specifically, we first design a two-sided causal graph that jointly models provider-customer fairness interactions through their causal relationships. Then we propose a discriminative counterfactual debiasing method, which effectively removes spurious correlations while maintaining true user-item interactions. Finally, we propose an LLM-enhanced counterfactual inference method to derive noise-resistant user/item representations from interaction data, enhancing the robustness of causal debiasing. The experimental results demonstrate the high effectiveness of our proposed model. We provide our code at https://github.com/houyimin660/LeadFairRec.

Most existing knowledge graph completion methods fail to model the fine-grained interactions among elements within triples, such as dependencies between entity attributes or contextual relationships involving predicates and entities. This limitation weakens their ability to infer implicit knowledge and hinders overall reasoning performance. To address this issue, we define a three-level classification of element interactions: Interactions between Elements at the Head Entity (IEH), Interactions between Elements at the Relationship (IER), and Interactions between Elements at the Tail Entity (IET), that systematically models the influence propagation patterns among knowledge graph triples at element level. Based on these interaction types, we propose a novel Knowledge Graph Completion Model Based on Fine-Grained Element Interactions (EI-KGC). Our model captures both global structural patterns and semantic dependencies within triples by combining GNN propagation with fine-grained interaction modeling. Experimental results show that the EI-KGC consistently outperforms traditional baseline models, demonstrating the high effectiveness of our proposed model.

The goal of document-level relation extraction is to extract semantic information from multiple sentences within a document and identify the relations between entities across sentences. However, effectively representing the document's content and reasoning about cross-sentence entities presents a formidable challenge. In this paper, we propose an efficient Document-Level Relation Extraction Model based on Heterogeneous Graph Reasoning (HGR-DREM), which enables relation extraction more accurate. Specifically, we first construct a document-level heterogeneous graph to comprehensively capture the semantic relations between entities. Then, we design a meta-path attention-based reasoning mechanism to enhance the mutual influence among graph nodes. Furthermore, we utilize an extended adjacency matrix to represent the heterogeneous graph and leverage graph convolutional neural networks (GCNs) to extract high-dimensional features. The experiments on a real-world dataset demonstrate the effectiveness of our proposed model. All codes have been released at https://github.com/NuyoaH-code/HGR-DREM.

Data imputation has increasingly gained attention due to its critical role in enhancing data quality and accuracy. However, traditional imputation methods often lack the ability to leverage the underlying category information and employ static denoising strategies, leading to suboptimal results. In this paper, we propose a novel data imputation method based on generative adversarial denoise network to predict and fill in the missing values. Our approach first employs a dataset partitioning scheme to divide the dataset into several subsets based on potential data categories. We then propose a Generative Adversarial Denoise Imputation Network (GAD IN) to combine dynamic noise reduction with generative adversarial networks to enhance the model's adaptability and robustness. Extensive experiments on real-world datasets validate the superior performance of our proposed method in comparison to existing techniques.

Finding a team that is both competent in performing the task and compatible in working together has been extensively studied. However, most methods for team formation tend to rely on a set of skills only. In order to solve this problem, we present an efficient team formation method based on Constrained Pattern Graph (called CPG). Unlike traditional methods, our method takes into account both structure constraints and communication constraints on team members, which can better meet the requirements of users. First, a CPG preprocessing method is proposed to normalize a CPG and represent it as a CoreCPG in order to establish the basis for efficient matching. Second, a Communication Cost Index (called CCI) is constructed to speed up the matching between a CPG and its corresponding social network. Third, a CCI-based node matching algorithm is proposed to minimize the total number of intermediate results. Moreover, a set of incremental maintenance strategies for the changes of social networks are proposed. We conduct experimental studies based on two real-world social networks. The experiments demonstrate the effectiveness and the efficiency of our proposed method in comparison with traditional methods.