Sustav preporuke filmova zasnovan na primjeni više modela suradničkog filtriranja

master's thesis

• Title (english): Movie Recommender System Based on Application of Multiple Collaborative Filtering Models
• Author: Leo Miloš
• 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 objašnjeni su algortimi suradničkog filtriranja zasnovani na dekompoziciji singularne vrijednosti i nenegativne matrične faktorizacije i njihova implementacija u sustavu preporuka za filmove. Opisan je osnovni model suradničkog filtriranja, potrebni skup podataka za suradničko filtriranje i modeli suradničkog filtriranja zasnovani na metodama latentnih faktora. Ostvareno je programsko rješenje suradničkog filtriranja filmova za skup podataka koji sadrži ocjene korisnika za filmove. Opisane su metode procjene sustava preporuka i ostvarena je procjena programskog rješenja suradničkog filtriranja, s naglaskom na metrike točnosti.

Abstract (english)

This master thesis explains collaborative filtering algorithms based on singular value decomposition and non-negative matrix factorization and their implementation in movie recommender system. Basic model of collaborative filtering, dataset needed for collaborative filtering and collaborative filtering models based on latent factors are described. Collaborative filtering software solution for dataset which contains ratings of users for movies is realized. Evaluation methods for recommender systems are described and evaluation of collaborative filtering software solution is realized, with a focus on accuracy metrics.

• Keywords: sustav preporuka; suradničko filtriranje; skup podataka; latentni faktori; dekompozicija singularne vrijednosti; nenegativna matrična faktorizacija; procjena; metrike točnosti
• Keywords (english): recommender system; collaborative filtering; dataset; latent factors; singular value decomposition; non-negative matrix factorization; evaluation; accuracy metrics
• Language: croatian
• URN:NBN: urn:nbn:hr:168:773737
• 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 13:14:16