Comora: An Agenda-First Social Gathering Platform for Community Building

Abstract

In this paper, we examine Comora, a social interaction platform that connects individuals based on shared interests (i.e. commonality), rather than using traditional matching methods (i.e. demographic-based). Event hosts will be able to use the platform to connect with individuals who share a mutual interest; therefore, individuals looking for community engagement can find events based on that mutual interest; additionally, hosts can review the attendees of events to plan and execute new events in the future based on those event attendees. Some unique or novel characteristics of the platform pertain to the use of Mistral AI for conversational onboarding; Groq with Llama 3.3 70B for automatically generated content drawn from input provided by users; and a conversational chatbot interface that is consistently aware of the context in which conversations occur. The serverless architectural configuration of the platform developed consists of the following components: React 19 framework for the client side, Postgres with Row Level Security (Or Object Level Security) for the database tier, Free-tier cloud infrastructure provided by Supabase Cloud Infrastructure, and additional elements that illustrate the "Production Ready" quality attributes of Comora (i.e., Role-Based Access Control; Four-Axis Peer Review Process, Contribution/ Capture Mechanism that supports both qualitative and quantitative evaluations). Our study demonstrates how the combined application of Structured Interest Matching, coupled with Artificial Intelligence support, allow Hosts to create/organise events with lesser impediments in event creation/event attendance process for societies who are part of large metropolitan areas.