Automatizacija procesa u turističkom sektoru korištenjem RAG sustava i agenata umjetne inteligencije

master's thesis

• Title (english): Process automation in the tourism sector using RAG systems and AI agents
• Author: Tin Popović
• Mentor: Damir Pintar (mentor)
• Committee member: Damir Pintar (predsjednik povjerenstva)
• Committee member: Mihaela Vranić (član povjerenstva)
• Committee member: Ana Šarčević (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2025-02-28, Croatia

Abstract

Rad istražuje primjenu umjetne inteligencije u automatizaciji procesa u turističkom sektoru s naglaskom na RAG (engl. Retrieval-Augmented Generation) sustave i AI agente. Cilj rada je razviti sustav koji će unaprijediti korisničko iskustvo turista kroz inteligentnu podršku, personalizirane preporuke i poboljšanu obradu podataka. Razmatraju se veliki jezični modeli (LLM), njihova ograničenja i mogućnosti proširenja putem RAG arhitektura. Eksperimentalni dio uključuje razvoj i testiranje... More različitih RAG pristupa – naivni, napredni i modularni , s ciljem optimizacije pretraživanja i generiranja odgovora. Evaluacija sustava provedena je na turističkim upitima, a rezultati pokazuju značajnu prednost naprednih tehnika poput RAG-Fusion modela. Zaključuje se da kombinacija AI agenata i RAG sustava omogućuje poboljšanje korisničke podrške u turizmu kroz preciznije, relevantnije i personalizirane odgovore. Less

Abstract (english)

This paper explores the application of artificial intelligence in automating processes in the tourism sector, with a focus on RAG (Retrieval-Augmented Generation) systems and AI agents. The objective is to develop a system that enhances the tourist experience through intelligent support, personalized recommendations, and improved data processing. Large language models (LLMs) are analyzed, along with their limitations and potential extensions through RAG architectures. The experimental... More part includes the development and testing of different RAG approaches—naïve, advanced, and modular , with the goal of optimizing information retrieval and response generation. The system evaluation was conducted on tourism-related queries, and the results demonstrate a significant advantage of advanced techniques such as the RAG-Fusion model. The study concludes that the combination of AI agents and RAG systems enhances customer support in tourism by providing more precise, relevant, and personalized responses. Less

• Keywords: automatizacija; umjetna inteligencija; RAG sustavi; AI agenti; veliki jezični modeli; personalizirane preporuke; turizam; korisnička podrška
• Keywords (english): automation; artificial intelligence; RAG systems; AI agents; large language models; personalized recommendations; tourism; customer support
• Language: croatian
• URN:NBN: urn:nbn:hr:168:070462
• Study programme: Title: Computing; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra inženjer/inženjerka računarstva (magistar/magistra inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Open access
• Created on: 2025-03-12 11:16:58