CHATBOT RETRIEVAL-AUGMENTED GENERATION (RAG) AND ACCESS TO INSTITUTIONAL INFORMATION: A DIAGNOSTIC AND EVALUATION CASE STUDY AT THE SERIDÓ HIGHER EDUCATION CENTER (UFRN)
Chatbot; Retrieval-Augmented Generation (RAG); Information Access; Information Overload; Organizational Communication.
Access to information in higher education institutions (HEIs) is often hindered by channel fragmentation and communication overload. At the Centro de Ensino Superior do Seridó (CERES/UFRN), this scenario is reflected in the dispersion of procedural and administrative information across multiple repositories, generating inconsistencies, rework, and challenges for the academic community. This research aims to analyze the extent to which a conversational artificial intelligence system based on Retrieval-Augmented Generation (RAG) can support institutional information access at CERES/UFRN. The methodology adopts a mixed-methods approach structured into four phases: (1) qualitative and quantitative diagnosis through focus groups and surveys; (2) construction of the documentary corpus; (3) development of a functional RAG-based prototype (MVP); and (4) qualitative evaluation of its effectiveness, utility, and acceptance, grounded in TAM/UTAUT models. Expected outcomes include a systematic diagnosis of informational flows and the development of a prototype capable of mitigating informational overload and centralizing access to official data in an accurate and traceable manner. The research contributes a transferable socio-technical methodology for other HEIs facing similar fragmentation challenges.