Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Towards Empathetic Conversational Recommender Systems

Xiaoyu Zhang, Ruobing Xie, Yougang Lyu, Xin Xin, Pengjie Ren, Mingfei Liang, Bo Zhang, Zhanhui Kang, Maarten de Rijke, Zhaochun Ren

TL;DR

An empathetic conversational recommender (ECR) framework that employs user emotions to refine user preference modeling for accurate recommendations and enlarges its empathetic data using emotion labels annotated by large language models and emotional reviews collected from external resources.

Abstract

Conversational recommender systems (CRSs) are able to elicit user preferences through multi-turn dialogues. They typically incorporate external knowledge and pre-trained language models to capture the dialogue context. Most CRS approaches, trained on benchmark datasets, assume that the standard items and responses in these benchmarks are optimal. However, they overlook that users may express negative emotions with the standard items and may not feel emotionally engaged by the standard responses. This issue leads to a tendency to replicate the logic of recommenders in the dataset instead of aligning with user needs. To remedy this misalignment, we introduce empathy within a CRS. With empathy we refer to a system's ability to capture and express emotions. We propose an empathetic conversational recommender (ECR) framework. ECR contains two main modules: emotion-aware item recommendation and emotion-aligned response generation. Specifically, we employ user emotions to refine user preference modeling for accurate recommendations. To generate human-like emotional responses, ECR applies retrieval-augmented prompts to fine-tune a pre-trained language model aligning with emotions and mitigating hallucination. To address the challenge of insufficient supervision labels, we enlarge our empathetic data using emotion labels annotated by large language models and emotional reviews collected from external resources. We propose novel evaluation metrics to capture user satisfaction in real-world CRS scenarios. Our experiments on the ReDial dataset validate the efficacy of our framework in enhancing recommendation accuracy and improving user satisfaction.

Towards Empathetic Conversational Recommender Systems

TL;DR

An empathetic conversational recommender (ECR) framework that employs user emotions to refine user preference modeling for accurate recommendations and enlarges its empathetic data using emotion labels annotated by large language models and emotional reviews collected from external resources.

Abstract

Conversational recommender systems (CRSs) are able to elicit user preferences through multi-turn dialogues. They typically incorporate external knowledge and pre-trained language models to capture the dialogue context. Most CRS approaches, trained on benchmark datasets, assume that the standard items and responses in these benchmarks are optimal. However, they overlook that users may express negative emotions with the standard items and may not feel emotionally engaged by the standard responses. This issue leads to a tendency to replicate the logic of recommenders in the dataset instead of aligning with user needs. To remedy this misalignment, we introduce empathy within a CRS. With empathy we refer to a system's ability to capture and express emotions. We propose an empathetic conversational recommender (ECR) framework. ECR contains two main modules: emotion-aware item recommendation and emotion-aligned response generation. Specifically, we employ user emotions to refine user preference modeling for accurate recommendations. To generate human-like emotional responses, ECR applies retrieval-augmented prompts to fine-tune a pre-trained language model aligning with emotions and mitigating hallucination. To address the challenge of insufficient supervision labels, we enlarge our empathetic data using emotion labels annotated by large language models and emotional reviews collected from external resources. We propose novel evaluation metrics to capture user satisfaction in real-world CRS scenarios. Our experiments on the ReDial dataset validate the efficacy of our framework in enhancing recommendation accuracy and improving user satisfaction.
Paper Structure (42 sections, 7 equations, 3 figures, 14 tables)

This paper contains 42 sections, 7 equations, 3 figures, 14 tables.

Table of Contents

  1. Introduction
  2. Related Works
  3. Preliminaries
  4. Problem Formulation
  5. Backbone Framework
  6. Method
  7. Empathetic Data Enlargement
  8. User Emotion Extraction
  9. Emotional Response Construction.
  10. Emotion-aware Item Recommendation
  11. Emotion-aware Entity Representation
  12. Feedback-aware Item Reweighting
  13. Emotion-aligned Response Generation
  14. Emotion-aligned Generation Prompt
  15. Emotion-aligned Generator
  16. ...and 27 more sections

Figures (3)

  • Figure 1: An example of a conversation on movie recommendation between a user and the system. Text conveying user emotions is highlighted in red font. System responses expressing emotions are marked in green font.
  • Figure 2: Overview of ECR. ECR has two key modules: (i) emotion-aware item recommendation for better user preference understanding, and (ii) emotion-aligned response generation for engaging conversations.
  • Figure 3: Performance comparison of item recommendation w.r.t. different weight scalars for user feedback.