Abstract:Empowering conversational recommender systems (CRSs) with knowledge facilitates the generation of high-quality human-like recommendation proposals to users. Despite substantial endeavors in developing knowledge-based CRSs, they mainly focus on modeling entity knowledge by introducing external data, which cannot sufficiently contribute to understanding complex contextual semantics. Additionally, the source of knowledge relies on external data, ignoring the effective use of original training data, significantly increasing the cost of building efficient CRSs.To tackle the issues above, we explore how to leverage raw dialogue data for modeling and exploiting contextual knowledge to compensate for existing knowledge's shortcomings. Specifically, we first train knowledge generators by constructing multiple standard hybrid prompt templates and answer spaces, which is conducive to perceiving the reasons behind contextual connections from multiple aspects and storing corresponding contextual knowledge. Then, we design answer-guided prompt templates based on the answer space, closing the gap between inference and training of knowledge generators and facilitating the generation of aspect-specific contextual knowledge. Next, we extend our approach by harnessing external knowledge graphs to construct entity knowledge. We propose a local-to-global semantic fusion strategy to fuse knowledge representations from different sources better. Finally, based on the fused knowledge representations, we tailor contextual knowledge-enhanced prompt templates for the CRS to stimulate the reasoning ability of pre-trained language models on dialogue and recommendation tasks. We evaluate the performance of our method on separate evaluation datasets, including the ReDial and the INSPIRED datasets. Based on the Recall, MRR, NDCG, and Distinct evaluation indicators, we demonstrate the effectiveness of the proposed method in dialogue tasks and recommendation tasks. At the same time, we demonstrate that by constructing an interpretable multi-aspect hybrid prompt template, we can effectively mine contextual knowledge from multiple aspects of dialogue.

Conversation and Recommendation: Knowledge-Enhanced Personalized Dialog System.

KB4ConvRec: Knowledge Base Enhanced Memory Network for Conversational Recommendation

Towards Knowledge-Based Recommender Dialog System

KERS: A Knowledge-Enhanced Framework for Recommendation Dialog Systems with Multiple Subgoals

A Personalized System for Conversational Recommendations

Improving Conversation Recommendation System Through Personalized Preference Modeling and Knowledge Graph

Improving Conversational Recommender Systems via Knowledge-Enhanced Temporal Embedding.

Knowledge-Based Conversational Recommender Systems Enhanced by Dialogue Policy Learning

Dual Dynamic Memory Network for End-to-End Multi-turn Task-oriented Dialog Systems.

Enhancing User Personalization in Conversational Recommenders

Towards Conversational Search and Recommendation: System Ask, User Respond.

Dialog History Management for End-to-end Task-Oriented Dialog Systems

Improving Conversational Recommender Systems via Knowledge Graph based Semantic Fusion

IRF: Interactive Recommendation through Dialogue

Estimation-Action-Reflection: Towards Deep Interaction Between Conversational and Recommender Systems

Conversational Recommendation: Formulation, Methods, and Evaluation

Dialoging Resonance: How Users Perceive, Reciprocate and React to Chatbot's Self-Disclosure in Conversational Recommendations

Dialogue History Matters! Personalized Response Selection in Multi-Turn Retrieval-Based Chatbots

MCKP: Multi-aspect Contextual Knowledge-enhanced Prompting for Conversational Recommender Systems

Interactive Social Recommendation

Less is More: Learning to Refine Dialogue History for Personalized Dialogue Generation