Scholarlink

<p>Session-based recommendation (SBR) focuses on forecasting the next item a user is likely to select using brief and anonymous sequences of interactions. Existing methods face three key challenges: (1) difficulty in distinguishing noisy transitions within sessions, (2) absence of explicit modeling for target intent, and (3) misalignment between intra- and inter-session information. We propose <b>TiDGRec</b> (<b>T</b>arget-intention aware <b>D</b>ual-<b>G</b>raph <b>Rec</b>ommender), a framework designed to address these limitations through hierarchical denoising and target-guided dual-graph learning. A <b>Target Proxy Node (TPN)</b> is introduced into the <b>Sequential Transition Graph (STG)</b> to capture user intent representations. An <b>Adaptive Target-aware Sparsifier (ATS)</b> based on dynamic <InlineEquation ID="IEq1"> <EquationSource Format="TEX">\(\alpha _s\)</EquationSource> </InlineEquation>, adaptively filters irrelevant transitions. The learned target representation and item embeddings from the <b>Cross-session Co-occurrence Graph (CCG)</b> are jointly input to the <b>Target-guided Cross-graph Filter (TCF)</b> to enhance target-aware global relations. By connecting STG and CCG through shared target signals, TiDGRec forms a dual-graph, dual-target architecture that enhances intent alignment, suppresses semantic noise, and improves overall recommendation quality. A comprehensive evaluation across various benchmark datasets demonstrates that TiDGRec achieves superior performance compared to existing SBR methods.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

TiDGRec: dual-graph modeling with target-intention filtering for session-based recommendation

  • 期刊论文

  • 发表日期：

    2025年10月10日

查看全文
  • Junnan Zhuo,
  • Bohan Li,
  • Sujie Yu,
  • Shuai Xu,
  • Yicong Li,
  • Xinzhe Zhao,
  • Guan Yuan

摘要

Session-based recommendation (SBR) focuses on forecasting the next item a user is likely to select using brief and anonymous sequences of interactions. Existing methods face three key challenges: (1) difficulty in distinguishing noisy transitions within sessions, (2) absence of explicit modeling for target intent, and (3) misalignment between intra- and inter-session information. We propose TiDGRec (Target-intention aware Dual-Graph Recommender), a framework designed to address these limitations through hierarchical denoising and target-guided dual-graph learning. A Target Proxy Node (TPN) is introduced into the Sequential Transition Graph (STG) to capture user intent representations. An Adaptive Target-aware Sparsifier (ATS) based on dynamic \(\alpha _s\) , adaptively filters irrelevant transitions. The learned target representation and item embeddings from the Cross-session Co-occurrence Graph (CCG) are jointly input to the Target-guided Cross-graph Filter (TCF) to enhance target-aware global relations. By connecting STG and CCG through shared target signals, TiDGRec forms a dual-graph, dual-target architecture that enhances intent alignment, suppresses semantic noise, and improves overall recommendation quality. A comprehensive evaluation across various benchmark datasets demonstrates that TiDGRec achieves superior performance compared to existing SBR methods.