Abstract:
The online social media facilitates easy communication between people from various cultures, but language varieties,
emotional expression differences and cultural norms confusion could cause misunderstandings and conflicts. Conventional
moderation practices have been oriented more towards elimination of toxic material than assistance in conversing with
mutual respect. In this paper, we suggest a conceptual model of an AI-based chatbot mediator used to accommodate
intercultural conflict resolution among online communities. Based on a qualitative and content analysis research method,
1,000 publicly accessible comments from social media sites such as Facebook, X (Twitter), and Instagram were collected
to identify popular conflict initiators, emotions fed through during the conflicts and escalation strategies. Building upon
these learnings, we propose a chatbot mediation model leveraging Natural Language Processing along with sentiment
analysis to infer emotion-driven distress and suggest calm, empathetic and culturally appropriate responses. The
framework proposed here brings attention to the capacity for AI to support healthier communication dynamics, mitigate
conflict escalation and enhance intercultural understanding in digital spaces.
