Towards an Online Empathetic Chatbot with Emotion Causes

TL;DR

This paper presents an empathetic chatbot that leverages emotion causes to improve understanding and response quality, using counseling strategies and real-world data to enhance online conversational empathy.

Contribution

It introduces a novel approach to incorporate emotion causes into empathetic chatbots, advancing beyond response content control to understanding underlying emotional triggers.

Findings

The proposed model outperforms SOTA methods on automatic metrics.

Expert human judgments favor the chatbot's empathetic responses.

Online user evaluations show improved engagement and satisfaction.

Abstract

Existing emotion-aware conversational models usually focus on controlling the response contents to align with a specific emotion class, whereas empathy is the ability to understand and concern the feelings and experience of others. Hence, it is critical to learn the causes that evoke the users' emotion for empathetic responding, a.k.a. emotion causes. To gather emotion causes in online environments, we leverage counseling strategies and develop an empathetic chatbot to utilize the causal emotion information. On a real-world online dataset, we verify the effectiveness of the proposed approach by comparing our chatbot with several SOTA methods using automatic metrics, expert-based human judgements as well as user-based online evaluation.