Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Explainable Session-based Recommendation via Path Reasoning

Yang Cao, Shuo Shang, Jun Wang, Wei Zhang

TL;DR

A generalized hierarchical reinforcement learning framework for SR, which improves the explainability of existing SR models via Path Reasoning, namely PR4SR, which is instantiate in five state-of-the-art SR models and compared with other explainable SR frameworks to demonstrate the effectiveness.

Abstract

This paper explores providing explainability for session-based recommendation (SR) by path reasoning. Current SR models emphasize accuracy but lack explainability, while traditional path reasoning prioritizes knowledge graph exploration, ignoring sequential patterns present in the session history. Therefore, we propose a generalized hierarchical reinforcement learning framework for SR, which improves the explainability of existing SR models via Path Reasoning, namely PR4SR. Considering the different importance of items to the session, we design the session-level agent to select the items in the session as the starting point for path reasoning and the path-level agent to perform path reasoning. In particular, we design a multi-target reward mechanism to adapt to the skip behaviors of sequential patterns in SR, and introduce path midpoint reward to enhance the exploration efficiency in knowledge graphs. To improve the completeness of the knowledge graph and to diversify the paths of explanation, we incorporate extracted feature information from images into the knowledge graph. We instantiate PR4SR in five state-of-the-art SR models (i.e., GRU4REC, NARM, GCSAN, SR-GNN, SASRec) and compare it with other explainable SR frameworks, to demonstrate the effectiveness of PR4SR for recommendation and explanation tasks through extensive experiments with these approaches on four datasets.

Explainable Session-based Recommendation via Path Reasoning

TL;DR

A generalized hierarchical reinforcement learning framework for SR, which improves the explainability of existing SR models via Path Reasoning, namely PR4SR, which is instantiate in five state-of-the-art SR models and compared with other explainable SR frameworks to demonstrate the effectiveness.

Abstract

This paper explores providing explainability for session-based recommendation (SR) by path reasoning. Current SR models emphasize accuracy but lack explainability, while traditional path reasoning prioritizes knowledge graph exploration, ignoring sequential patterns present in the session history. Therefore, we propose a generalized hierarchical reinforcement learning framework for SR, which improves the explainability of existing SR models via Path Reasoning, namely PR4SR. Considering the different importance of items to the session, we design the session-level agent to select the items in the session as the starting point for path reasoning and the path-level agent to perform path reasoning. In particular, we design a multi-target reward mechanism to adapt to the skip behaviors of sequential patterns in SR, and introduce path midpoint reward to enhance the exploration efficiency in knowledge graphs. To improve the completeness of the knowledge graph and to diversify the paths of explanation, we incorporate extracted feature information from images into the knowledge graph. We instantiate PR4SR in five state-of-the-art SR models (i.e., GRU4REC, NARM, GCSAN, SR-GNN, SASRec) and compare it with other explainable SR frameworks, to demonstrate the effectiveness of PR4SR for recommendation and explanation tasks through extensive experiments with these approaches on four datasets.
Paper Structure (14 sections, 14 equations, 11 figures, 9 tables, 1 algorithm)

This paper contains 14 sections, 14 equations, 11 figures, 9 tables, 1 algorithm.

Table of Contents

  1. Introduction
  2. Related Work
  3. Session-based recommendation
  4. Path reasoning approaches
  5. Preliminary
  6. Knowledge Graph Construction
  7. Proposed Frameworks
  8. Mixed state encoder
  9. Hierarchical Reinforcement Learning
  10. Definition of the MDP for PR4SR
  11. Experiments
  12. Experimental Setup
  13. Results and Analysis
  14. Conclusion

Figures (11)

  • Figure 1: Three examples of path reasoning: (a) starting with a user, (b) starting with the last item, (c) starting with the most relevant item.
  • Figure 2: Details of knowledge graph construction: (a) extract image features; (b) split relations based on different product domains; (c) merge duplicate entities.
  • Figure 3: The overview of the proposed framework.
  • Figure 4: An example explaining the design of Multi-Target Reward.
  • Figure 5: An example explaining the design of Path Midpoint Reward. Entity 7 is the goal point.
  • ...and 6 more figures