본문 바로가기
  • Home

The Effects of Empathy Regulation Strategies on Negative Emotions in Place Recommender Systems

  • Journal of The Korea Society of Computer and Information
  • Abbr : JKSCI
  • 2025, 30(8), pp.41~52
  • DOI : 10.9708/jksci.2025.30.08.041
  • Publisher : The Korean Society Of Computer And Information
  • Research Area : Engineering > Computer Science
  • Received : July 17, 2025
  • Accepted : August 6, 2025
  • Published : August 29, 2025

Hae-Ryung Lee 1 Ju-Hye Ha 1 Chang-Hoon Oh 1

1연세대학교

Accredited

ABSTRACT

This study empirically analyzed the impact of conversational agents applying empathy regulation strategies (congruent, divergent) on user experience when recognizing negative emotional states (anger, sadness) in place recommendation contexts. Negative emotions impair users' judgment and decision-making, highlighting the need for empathetic responses. Yet little research exists on strategy effectiveness. A 2 (anger/sadness) × 2 (congruent/divergent) within-subject experiment was conducted with 44 participants using location recommendation scenarios. Results showed the 'sadness × congruent' condition achieved the highest trust ratings, with congruent strategy showing marginally significant advantage in recommendation acceptance for sadness states. Additionally, 'anger × congruent' strategy received higher evaluations, contrary to the change-oriented behavioral characteristics of anger. Through these findings, this study confirmed that acceptance-centered empathy strategies are generally more effective in emotion-based recommender systems, providing theoretical contributions by applying emotion regulation theory to AI-human interaction and offering guidelines for emotion-based recommender system design.

Citation status

* References for papers published after 2024 are currently being built.