Design and Implementation of ChatGBG: An AI-Driven FAQ Chatbot for Tourists in Gothenburg - Exploring a Practical Application of Conversational AI by Leveraging Open Source Technologies for Municipal services
Download
Date
Type
Examensarbete för masterexamen
Master's Thesis
Master's Thesis
Model builders
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
This thesis presents the design and implementation of an AI-driven FAQ chatbot, ChatGBG, aimed at enhancing the tourist experience in Gothenburg. This project combines advanced AI technologies with intuitive design to provide immediate, reliable, and accessible information to tourists, thus facilitating an enjoyable exploration of the city. It explores the practical application and challenges of integrating conversational AI within a municipal context, highlighting the chatbot’s ability to handle real-time user queries effectively while maintaining a simple interface. Addressing the dual research questions, this study first identifies the key considerations necessary for designing and implementing an effective AI-driven service in a governmental or municipal setting. These include user interaction design, data sourcing, management, and the integration of multilingual support to cater to a diverse user base. Secondly, the thesis evaluates various open-source large language models to determine the most suitable model for this application, focusing on language capabilities, ease of integration, and ethical considerations. Moreover, the project serves as a proof of concept for broader applications of AIdriven information systems in public services, demonstrating potential extensions beyond tourism. Ethical considerations, particularly in data handling, bias mitigation, and user privacy, are critically examined to ensure the chatbot adheres to ethical AI development standards. ChatGBG not only aims to enhance the tourist experience by providing a novel way to explore Gothenburg but also sets a benchmark for future AI applications in urban settings. The insights gained from this project contribute to the ongoing discourse on the responsible use of AI in public services, emphasizing the need for continuous improvement and adaptation to new technological advancements and user expectations.
Description
Keywords
Interaction Design, AI-driven chatbot, Large Language Model, Applied IT, Software Engineering, Open source, Municipal