Design and Development of a RAG-Based Chatbot System for Web Content Exploration

This thesis presents the design and development of an AI-based chatbot built on the Retrieval-Augmented Generation (RAG) framework and deployed within the Microsoft Azure environment. Developed in collaboration with the consultancy firm ICONSULTING S.p.A., the chatbot is intended to assist users in navigating a section of the Data Visualization Hub (DVH) of a Regional Public Administration, helping them retrieve information about healthcare performance indicators. As an experimental real-world application, the system is designed in compliance with current legal and ethical regulations, particularly the provisions of the EU AI Act, to ensure responsible AI integration in the public sector. We implemented a robust pipeline to support the development of the RAG-based chatbot and defined an evaluation process in collaboration with experts from the Regional Healthcare System. This project offers a concrete example of how generative AI tools can support the digital transformation of public healthcare services in Italy, paving the way for future research and development in this field.