Social recommendation utilizes social relationships to alleviate data sparsity and improve recommendation quality. Nevertheless, existing approaches still encounter several limitations. First, user preference structures in social networks are often hierarchical and non-Euclidean, whereas most existing models rely on Euclidean embeddings that are insufficient for capturing such complex geometries. Second, noisy social connections may propagate unreliable information and degrade representation quality. Third, commonly used multi-view fusion strategies based on multilayer perceptrons (MLPs) often lack the expressive capability required to effectively integrate heterogeneous user signals. To address these challenges, this paper proposes HGKAN (Hyperbolic Graph-Based Kolmogorov–Arnold Network), a unified framework that combines hyperbolic representation learning, graph neural networks, and KAN-based nonlinear fusion for social recommendation. Specifically, a Hyperbolic Intent Attention module is introduced to learn fine-grained and hierarchical user intent representations in hyperbolic space. In addition, a Hyperbolic Attention GCN is designed to denoise and aggregate social information while preserving complex relational structures. Furthermore, a KAN-based Multi-View Fusion Layer is developed to replace conventional MLP-based fusion, enabling more expressive and interpretable integration of preference, social, and intent representations. Extensive experiments conducted on three public benchmark datasets, including Yelp, Ciao, and Douban, demonstrate that HGKAN consistently outperforms several state-of-the-art recommendation methods. Experimental results show that the proposed framework achieves substantial improvements in both Recall@5 and NDCG@5, with all performance gains being statistically significant (p < 0.05). These findings verify the effectiveness of integrating hyperbolic geometric learning with KAN-based multi-view fusion for robust and accurate social recommendation.

Sequential-social recommendation systems are essential for understanding users’ evolving interests and predicting their future behaviors. While existing methods employing bidirectional graph modeling have shown promising results in capturing user interactions, they face significant challenges in effectively leveraging bidirectional information and addressing dynamic user preferences. To overcome these limitations, we propose Bidirectional Transformer Graphs for Sequential-Social Recommendation (BiTG4Rec), a novel framework that integrates both supervised and self-supervised learning to analyze dynamic user preferences at sequential and social levels through bidirectional graphs. Our approach comprises three core components: 1) a Bidirectional Dynamic Graph Convolutional Network (BiDGCN) that models long-term sequential preferences, 2) a Bidirectional Dynamic Graph Attention Network (BiDGAT) that captures short-term sequential preferences, and 3) a Bidirectional Hyperbolic Graph Contrastive Learning module (BiHGCL) that extracts social preferences. To unify these diverse signals, we introduce a TransformerGated mechanism that dynamically integrates long-term, short-term, and social preferences. Extensive experiments on multiple real-world datasets demonstrate that BiTG4Rec consistently outperforms state-of-the-art methods, validating its effectiveness for sequential-social recommendation tasks.