Sustav preporuke računalnih igara zasnovan na hibridnom suradničkom filtriranju uz primjenu dubokog učenja

master's thesis

• Title (english): Video Game Recommender System Based on Hybrid Collaborative Filtering With Application of Deep Learning
• Author: Roko Marinović
• Mentor: Goran Delač (mentor)
• Committee member: Goran Delač (predsjednik povjerenstva)
• Committee member: Marin Šilić (član povjerenstva)
• Committee member: Klemo Vladimir (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2021-07-06, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

U ovom radu opisani su sustavi preporuke zasnovani na suradničkom filtriranju. Opisano je i primjerima potkrijepljeno suradničko filtriranje zasnovano na memoriji. Opisani su algoritmi suradničkog filtriranja zasnovanog na modelima uz opis strojnog i dubokog učenja. Opisane su metrike vrednovanja sustava preporuke. Implementiran je i opisan sustav preporuke računalnih igara zasnovan na hibridnom suradničkom filtriranju uz primjenu dubokog učenja. Sustav je učen i vrednovan na Amazon Video Games skupu podataka te su opisani dobiveni rezultati vrednovanja.

Abstract (english)

This paper describes recommender systems based on collaborative filtering. Memory-based collaborative filtering is described with the examples. Model-based collaborative filtering algorithms are described with a description of machine and deep learning. The metrics for evaluating the recommender systems are described. A video game recommender system based on hybrid collaborative filtering with application of deep learning has been implemented and described. The system was trained and evaluated on the Amazon Video Games data set and the obtained evaluation results are described.

• Keywords: sustav preporuke; suradničko filtriranje; hibridno suradničko filtriranje; matrica korisnosti; strojno učenje; faktorizacijski strojevi; duboko učenje; unaprijedne neuronske mreže; konvolucijske neuronske mreže; obrada prirodnog jezika; metrike vrednovanja
• Keywords (english): recommender system; collaborative filtering; hybrid collaborative filtering; utility matrix; machine learning; factorization machines; deep learning; feedforward neural networks; convolutional neural networks; natural language processing; evaluation metrics
• Language: croatian
• URN:NBN: urn:nbn:hr:168:635262
• 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: 2023-01-13 15:30:59