Scholarlink

This paper presents the identification of emotions using Watson, a natural language processing tool, applied to multi-agent systems. In this approach, Watson acts as an intermediary between agents and identifies the emotions expressed in exchanged messages to assist in the agents’ decision-making. The identification of emotions was based on the OCC model using tokens – words related to each emotion – which are useful for identifying and indicating the intensity of emotions. We built a prototype using the Jason platform to test communication between two agents, with Watson acting as an intermediary to identify emotions based on the OCC model. As an experiment, we present a study in Prisoner’s Dilemma and how emotions change agents’ decision-making.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Emotions Identification in Exchanges of Messages Between Agents

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Thiago Dantas,
  • Giovani Farias,
  • Cleo Billa,
  • Eder Gonçalves,
  • Diana Adamatti

摘要

This paper presents the identification of emotions using Watson, a natural language processing tool, applied to multi-agent systems. In this approach, Watson acts as an intermediary between agents and identifies the emotions expressed in exchanged messages to assist in the agents’ decision-making. The identification of emotions was based on the OCC model using tokens – words related to each emotion – which are useful for identifying and indicating the intensity of emotions. We built a prototype using the Jason platform to test communication between two agents, with Watson acting as an intermediary to identify emotions based on the OCC model. As an experiment, we present a study in Prisoner’s Dilemma and how emotions change agents’ decision-making.