FIT-ConferenceHub: From Design to Real-World Application of an LLM-Based System for Academic Conference Discovery in the ICT Domain
摘要
This paper presents FIT-ConferenceHub, an intelligent system developed for the specific needs of researchers and students at the Faculty of Information Technology, University of Science, Vietnam National University Ho Chi Minh City (FIT-HCMUS) in discovering, managing, and tracking scientific conferences in the field of Information and Communication Technology (ICT). The system was designed following a comprehensive analysis of actual workflows, conference selection criteria, and challenges faced by the FIT HCMUS community. Leveraging Large Language Models (LLMs), FIT-ConferenceHub automates web crawling, extracts structured conference information, and maintains a continuously updated knowledge base. It integrates advanced multi-criteria search, personalized recommendations, tracking tools, and a conversational chatbot interface. The system has been fully deployed at FIT-HCMUS. Deployment data and user feedback at FIT-HCMUS confirm that the system improves the conference selection process, underscoring its practical value for the intended users.