Year: 2024
Type: dissertation
Abstract: [Abstract]: The search for information is a fundamental part of daily university life and can be timeconsuming in some cases. With the rise in the use of applications that leverage advances in Information Retrieval (IR) and Natural Language Processing (NLP), it is possible to create not only an efficient search system but also a fully interactive one. Large Language Models (LLMs) have become much more accessible to developers, allowing them to take advantage of text generation functions for various purposes. This work presents a solution to facilitate Information Retrieval within the University of A Coruña’s environment: a Virtual Assistant in the form of a Chatbot that can respond to users, especially students from the Faculty of Computer Science. The goal is to develop a tool that can be integrated with the faculty’s website, allowing it to access the relevant files and either answer students’ questions or redirect them to the appropriate documents. This approach will enhance the user experience by eliminating the need to search across different platforms and reduce the burden on the faculty’s information services. The final product has been evaluated using different Large Language Models, both to ensure its proper functioning and to conduct a small comparative study to determine the best fit for the system.
Source: RUC (Universidade Da Coruña)
Author: Río Rico, Martín do
Language: English
Subfield: Information Systems
Field: Computer Science
Domain: Physical Sciences
SDG: Quality education
Open Access status: green
OpenAlex ID: W7044133066