Implementacija jezičnog asistenta za statističke analize korištenjem velikih jezičnih modela

master's thesis

• Title (english): Implementation of a language assistant for statistical analysis using large language models
• Author: Filip Krišković
• Mentor: Marina Bagić Babac (mentor)
• Committee member: Marina Bagić Babac (predsjednik povjerenstva)
• Committee member: Gordan Gledec (član povjerenstva)
• Committee member: Boris Vrdoljak (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2024-07-11, Croatia

Abstract

Ovaj diplomski rad istražuje potrebu i proces izrade jezičnog asistenta koji pruža odgovore na upite korisnika vezane uz CSV datoteke. Implementacija asistenta se oslanja na korištenje dostupnih velikih jezičnih modela otvorenog koda, prilagođenih zahtjevima korisnika. Korisnički upiti se transformiraju iz prirodnog u programski jezik kako bi se omogućilo izvođenje statističkih izračuna nad unesenim podacima te korisniku pružili odgovori u prirodnom jeziku. Odgovarajućim metrikama provest će se evaluacija performansi jezičnog asistenta kako bi se procijenila njegova učinkovitost i pouzdanost u pružanju odgovora na postavljena pitanja.

Abstract (english)

This thesis investigates the need and process of creating a language assistant that provides answers to user queries related to CSV files. The implementation of the assistant relies on the use of available open source large language models adapted to the user's requirements. User queries are transformed from natural to programming language in order to make it possible to perform statistical calculations on the entered data and provide the user with answers in natural language. The language assistant's performance will be evaluated using appropriate metrics in order to assess its effectiveness and reliability in providing answers to the questions posed.

• Keywords: Obrada Prirodnog Jezika; chatbotovi; generiranje koda; transformeri; Statistička analiza; Prostorna analiza
• Keywords (english): Natural Language Processing; chatbots; code generation; transformers; Statistical Analysis; Spatial Analysis
• Language: english
• URN:NBN: urn:nbn:hr:168:792771
• 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: Closed access
• Created on: 2024-10-12 21:12:12