Mobilna aplikacija za analizu osobne potrošnje na temelju slika maloprodajnih računa

master's thesis

• Title (english): A mobile application for personal expenses analysis based on images of retail receipts
• Author: Petar Miličević
• Mentor: Alan Jović (mentor)
• Committee member: Alan Jović (predsjednik povjerenstva)
• Committee member: Nikolina Frid (član povjerenstva)
• Committee member: Jan Šnajder (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2024-07-08, Croatia

Abstract

Ovim radom demonstrirana je praktična primjena tri modela transformera (Seamless,
Donut i BERTić) u zadacima obrade prirodnog jezika. Prikazana je upotreba modela za
obradu slika računa i klasifikaciju pojedinih stavki u jednu od pet kategorija, s klasifika-
cijskom točnošću od oko 97% koristeći model BERTić. Izrađena je mobilna aplikacija za
operacijski sustav Android koja koristi izgrađene modele.

Abstract (english)

This paper demonstrates the practical application of three transformer models (Seam-
less, Donut, and BERTić) in natural language processing tasks. It shows the use of these
models for processing receipt images and classifying individual items into one of five
categories, achieving a classification accuracy of approximately 97% using the BERTić
model. A mobile application for the Android operating system, which utilizes the devel-
oped models, has been created.

• Keywords: Android; NLP; Seamless; Donut; BERT
• Keywords (english): Android; NLP; Seamless; Donut; BERT
• Language: croatian
• URN:NBN: urn:nbn:hr:168:503002
• 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: 2024-10-12 21:36:58