Implementacija sustava za preporuku filmova

undergraduate thesis

• Title (english): Implementation of a Movie Recommender System
• Author: Sanja Krznarić
• Mentor: Damir Pintar (mentor)
• Committee member: Damir Pintar (predsjednik povjerenstva)
• Committee member: Luka Humski (član povjerenstva)
• Committee member: Ljiljana Brkić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2020-07-07, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Preporučiteljski sustavi su softverski sustavi koji koriste algoritme za filtriranje sadržaja
s ciljem da se pojedinom korisniku prikaže samo njemu relevantan dio ponude
odnosno sadržaja sustava. Postoji nekoliko vrsta preporučiteljskih algoritama - neki
od njih korisnicima prikazuju nepersonalizirani najpopularniji sadržaj, neki na temelju
značajki korisnika i sadržaja rade klasifikaciju kako bi predvidjeli hoće li se korisniku
(ili neće) svidjeti određeni sadržaj, a neki koriste kolaborativno filtriranje tj. na temelju
sličnosti između stavki ili korisnika za određenog korisnika traže one stavke koje
bi mu trebale biti najzanimljivije.
U ovom radu analizirani su neki od algoritama Pythonove biblioteke Surprise te je
procijenjena njihova uspješnost nad MovieLens 100k bazom podataka.

Abstract (english)

Recommender systems are software systems that use content filtering algorithms
with the aim of reducing content shown to a user to the items that he is likely to be interested
in. There are several groups of recommender algorithms - the ones that show
all the users the same most popular content without personalisation, the ones that use
classification by user and item features to determine whether the user is going to like
a specific item or not, and those that use collaborative filtering or, in other words, use
similarity between items and users to determine which items are going to be most appealing
to a specific user.
This paper analyses some of the algorithms provided by Python’s Surprise library
and estimates their precision above the MovieLens 100k database.

• Keywords: Preporučiteljski sustav; preporučiteljski algoritmi; korisnik; film; ocjena; sadržaj; filtriranje; baza; MovieLens 100k; program; Python; Surprise; NormalPredictor; BaselineOnly; SVD; KNNBasic; KNNWithMeans; KNNBaseline
• Keywords (english): Recommender system; recommender algorithm; user; movie; rating; content; filtering; dataset; MovieLens 100k; program; Python; Surprise; NormalPredictor; BaselineOnly; SVD; KNNBasic; KNNWithMeans; KNNBaseline
• Language: croatian
• URN:NBN: urn:nbn:hr:168:329787
• Study programme: Title: Computing; Study programme type: university; Study level: undergraduate; Academic / professional title: sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva (sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2023-01-13 15:21:42