Ostvarenje sustava preporučivanja koristeći algoritam matrične faktorizacije

master's thesis

• Title (english): Recommendation Systems Implementation Using Matrix Factorization Algorithm
• Author: Luka Pavlović
• Mentor: Klemo Vladimir (mentor)
• Committee member: Klemo Vladimir (predsjednik povjerenstva)
• Committee member: Marin Šilić (član povjerenstva)
• Committee member: Goran Delač (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2022-02-18, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

U ovom radu bavili smo se sustavima preporučivanja i metodom matrične faktorizacije kod izrade jednog takvog sustava. U prvom poglavlju čitatelju smo približili temu i opisali sustave preporučivanja te njihove primjene. Pokazali smo koje sve vrste sustava preporučivanja poznajemo i načine na koje se oni izrađuju, kao i tehnike koje nam pomažu pri izradi. U drugom poglavlju bavili smo se vlastitom implementacijom koja koristi matričnu faktorizaciju kao metodu unutar sustava preporučivanja u C++ statičkom programskom jeziku. Predstavili smo skup podataka nad kojim smo evaluirali implementaciju, a onda smo dublje ušli u detalje implementacije cijelog jednog sustava i validacije istog. Na kraju smo usporedili implementirani sustav s postojećim implementacijama sustava preporučivanjima i napravili zaključak o uspješnosti rada.

Abstract (english)

In this paper, we have described the recommendation systems and the matrix factorization method in the development of such a system. In the first chapter, we introduced a reader to the topic and described the recommendation systems and their applications. We have shown which types of recommender systems are there and the ways in which they are made, as well as the techniques we use in order to make them. In the second chapter, we described our own implementation that uses matrix factorization as a method within the recommendation system in the C++ static programming language. We presented a dataset on which to evaluate the implementation, and then described the details of the implementation of the entire system and its validation. Finally, we compared the implemented system with the existing recommender systems and made a conclusion about the success of the work

• Keywords: Sustav preporučivanja; matrica korisnosti; matrična faktorizacija; suradničko filtriranje; stohastički gradijentni spust; C++; Python; surprise; MovieLens
• Keywords (english): Recommender system; utility matrix; matrix factorization; collaborative filtering; stochastic gradient descent; C++; Python; surprise; MovieLens
• Language: croatian
• URN:NBN: urn:nbn:hr:168:933795
• 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 15:50:32