Scholarlink

This paper introduces SwAtI, an AI-driven chatbot developed to assist victims of domestic violence under the Protection of Women from Domestic Violence Act, 2005. SwAtI combines Social Emotional Learning (SEL) principles with a fine-tuned LLaMA model to provide both legal guidance and emotional support. Phase 1 of the research focused on identifying key SEL principles—such as empathy, active listening, confidentiality, and non-judgmental support—through semi-structured interviews with 4 legal experts and four focus groups with a total of 25 domestic violence survivors. Phase 2 involved embedding these principles into the chatbot’s conversational flow and testing their impact on user satisfaction and empowerment. During the live research preview, hosted on the SwAtI platform from February 19th to 25th, 2024, over 314 interactions were recorded with 21 unique users. The findings indicated that incorporating SEL principles increased user engagement by 40%, with 85% of users reporting feeling emotionally supported and 95% expressing trust in the confidentiality of their interactions. By addressing both the legal and emotional needs of survivors, SwAtI highlights the potential of AI to provide comprehensive and empathetic support to victims of domestic violence.

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

Artificial Intelligence for Legal Assistance: A Prescriptive Analytics Model Integrating Social Emotional Learning for Assisting Victims of Domestic Violence in India

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Dipali Awasekar,
  • L. M. R. J. Lobo

摘要

This paper introduces SwAtI, an AI-driven chatbot developed to assist victims of domestic violence under the Protection of Women from Domestic Violence Act, 2005. SwAtI combines Social Emotional Learning (SEL) principles with a fine-tuned LLaMA model to provide both legal guidance and emotional support. Phase 1 of the research focused on identifying key SEL principles—such as empathy, active listening, confidentiality, and non-judgmental support—through semi-structured interviews with 4 legal experts and four focus groups with a total of 25 domestic violence survivors. Phase 2 involved embedding these principles into the chatbot’s conversational flow and testing their impact on user satisfaction and empowerment. During the live research preview, hosted on the SwAtI platform from February 19th to 25th, 2024, over 314 interactions were recorded with 21 unique users. The findings indicated that incorporating SEL principles increased user engagement by 40%, with 85% of users reporting feeling emotionally supported and 95% expressing trust in the confidentiality of their interactions. By addressing both the legal and emotional needs of survivors, SwAtI highlights the potential of AI to provide comprehensive and empathetic support to victims of domestic violence.