Improving Conversational Recommender Systems via Knowledge Graph based Semantic Fusion

Abstract

Conversational recommender systems (CRS) aim to recommend high-quality items to users through interactive conversations. Although several efforts have been made for CRS, two major issues still remain to be solved. First, the conversation data itself lacks of sufficient contextual information for accurately understanding users' preference. Second, there is a semantic gap between natural language expression and item-level user preference. To address these issues, we incorporate both word-oriented and entity-oriented knowledge graphs (KG) to enhance the data representations in CRSs, and adopt Mutual Information Maximization to align the word-level and entity-level semantic spaces. Based on the aligned semantic representations, we further develop a KG-enhanced recommender component for making accurate recommendations, and a KG-enhanced dialog component that can generate informative keywords or entities in the response text. Extensive experiments have demonstrated the effectiveness of our approach in yielding better performance on both recommendation and conversation tasks.

References in corpus (12)

• Semi-Supervised Classification with Graph Convolutional Networks
• A Neural Conversational Model
• ConceptNet 5.5: An Open Multilingual Graph of General Knowledge
• Incorporating Copying Mechanism in Sequence-to-Sequence Learning
• A Diversity-Promoting Objective Function for Neural Conversation Models
• Estimation-Action-Reflection: Towards Deep Interaction Between Conversational and Recommender Systems
• An Information Retrieval Approach to Short Text Conversation
• Towards Deep Conversational Recommendations
• Deep Conversational Recommender in Travel
• Towards Knowledge-Based Recommender Dialog System
• Sequential Matching Network: A New Architecture for Multi-turn Response Selection in Retrieval-based Chatbots
• QAInfomax: Learning Robust Question Answering System by Mutual Information Maximization