Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Item Cluster-aware Prompt Learning for Session-based Recommendation

Wooseong Yang, Chen Wang, Zihe Song, Weizhi Zhang, Philip S. Yu

TL;DR

CLIP-SBR addresses the limitation of relying solely on intra-session item relations in session-based recommendation by introducing a two-module framework: a global item relationship graph with Leiden-based clustering to capture inter-session connections, and cluster-aware soft prompts that are integrated into SBR training. By unifying intra- and inter-session interactions and injecting cluster-specific cues, it provides a universal, efficient mechanism to enhance diverse SBR models. Empirical results across three real-world datasets and eight baselines show consistent performance gains and reduced training time, validating the approach's practicality. This work advances SBR by enabling scalable cross-session knowledge transfer through prompt learning on item communities.

Abstract

Session-based recommendation (SBR) aims to capture dynamic user preferences by analyzing item sequences within individual sessions. However, most existing approaches focus mainly on intra-session item relationships, neglecting the connections between items across different sessions (inter-session relationships), which limits their ability to fully capture complex item interactions. While some methods incorporate inter-session information, they often suffer from high computational costs, leading to longer training times and reduced efficiency. To address these challenges, we propose the CLIP-SBR (Cluster-aware Item Prompt learning for Session-Based Recommendation) framework. CLIP-SBR is composed of two modules: 1) an item relationship mining module that builds a global graph to effectively model both intra- and inter-session relationships, and 2) an item cluster-aware prompt learning module that uses soft prompts to integrate these relationships into SBR models efficiently. We evaluate CLIP-SBR across eight SBR models and three benchmark datasets, consistently demonstrating improved recommendation performance and establishing CLIP-SBR as a robust solution for session-based recommendation tasks.

Item Cluster-aware Prompt Learning for Session-based Recommendation

TL;DR

CLIP-SBR addresses the limitation of relying solely on intra-session item relations in session-based recommendation by introducing a two-module framework: a global item relationship graph with Leiden-based clustering to capture inter-session connections, and cluster-aware soft prompts that are integrated into SBR training. By unifying intra- and inter-session interactions and injecting cluster-specific cues, it provides a universal, efficient mechanism to enhance diverse SBR models. Empirical results across three real-world datasets and eight baselines show consistent performance gains and reduced training time, validating the approach's practicality. This work advances SBR by enabling scalable cross-session knowledge transfer through prompt learning on item communities.

Abstract

Session-based recommendation (SBR) aims to capture dynamic user preferences by analyzing item sequences within individual sessions. However, most existing approaches focus mainly on intra-session item relationships, neglecting the connections between items across different sessions (inter-session relationships), which limits their ability to fully capture complex item interactions. While some methods incorporate inter-session information, they often suffer from high computational costs, leading to longer training times and reduced efficiency. To address these challenges, we propose the CLIP-SBR (Cluster-aware Item Prompt learning for Session-Based Recommendation) framework. CLIP-SBR is composed of two modules: 1) an item relationship mining module that builds a global graph to effectively model both intra- and inter-session relationships, and 2) an item cluster-aware prompt learning module that uses soft prompts to integrate these relationships into SBR models efficiently. We evaluate CLIP-SBR across eight SBR models and three benchmark datasets, consistently demonstrating improved recommendation performance and establishing CLIP-SBR as a robust solution for session-based recommendation tasks.
Paper Structure (24 sections, 9 equations, 3 figures, 3 tables, 1 algorithm)

This paper contains 24 sections, 9 equations, 3 figures, 3 tables, 1 algorithm.

Table of Contents

  1. Introduction
  2. Related Works
  3. Session-based Recommendation
  4. Prompt Learning
  5. Proposed Method
  6. Problem Statement
  7. Item Relationship Mining
  8. Graph Construction
  9. Session Graph Construction
  10. Global Graph Construction
  11. Item Cluster Detection
  12. Item Cluster-aware Prompt Learning
  13. Experiments
  14. Experimental Setup
  15. Datasets
  16. ...and 9 more sections

Figures (3)

  • Figure 1: The proposed CLIP-SBR framework consists of two main modules. The Item Relationship Mining module constructs session graphs and combines them into a global graph to capture intra- and inter-session item relationships, followed by community detection to identify item clusters. The Item Cluster-aware Prompt Learning module enhances SBR models by integrating learnable soft prompts tailored to these clusters, embedding cluster-specific information to improve recommendation accuracy and efficiency.
  • Figure 2: Improvement (%) comparison of CLIP-SBR variants on the three datasets. The blue bar represents the original CLIP-SBR, while the other colors indicate various CLIP-SBR variants. "Avg. Improv." refers to the average of improvement percentage calculated from MRR@5 and Recall@5.
  • Figure 3: Performance comparison of CLIP-GRU4Rec with different resolution parameters on the three datasets. The MRR@5 scale for Last.fm and Xing is shown on the left y-axis, while the scale for Reddit is on the right y-axis.