Ostvarenje sustava preporučivanja koristeći algoritam suradničkog filtriranja

undergraduate thesis

• Title (english): Recommendation Systems Implementation Using Collaborative Filtering Algorithm
• Author: Andrija Raguž
• Mentor: Klemo Vladimir (mentor)
• Committee member: Klemo Vladimir (predsjednik povjerenstva)
• Committee member: Adrian Satja Kurdija (član povjerenstva)
• Committee member: Petar Afrić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2020-07-03, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

U ovom radu objašnjeni su sustavi preporučivanja. Sustavi preporučivanja služe online uslugama za bolju i personaliziraniju ponudu. Dijele se na tri osnovne skupine: sustave temeljene na sadržaju, sustave temeljene na suradničkom filtriranju i hibridne sustave koji su mješavina prethodna dva. Ovdje je pobliže opisan i implementiran algoritam suradničkog filtriranja. Sustavi suradničkog filtriranja dijele se na one temeljene na korisnik – korisnik odnosu i one temeljene na predmet – predmet... More odnosu. Suradničko filtriranje djeluje tako što izračuna sličnost između korisnika ili predmeta u bazi podataka pomoću neke matematičke funkcije npr. kosinus funkcija, Jaccardova funkcija itd. i na osnovu te sličnosti nađe određeni broj najsličnijih korisnika ili predmeta. Zatim koristeći njihove ocjene generira ocjene predmeta koje aktivni korisnik nije ocijenio i na osnovu toga stvara preporuke. Jednostavni oblik suradničkog filtriranja je implementiran koristeći programski jezik Python. Također, implementirana je mjera za evaluaciju sustava preporučivanja RMSE. RMSE uspoređuje stvarne i procijenjene vrijednosti ocjena i na osnovu toga vrednuje sustav preporučivanja. Less

Abstract (english)

This paper explains recommendation systems. Recommendation systems are used by online services to provide a better and more personalized offer. They are divided into three basic groups: content-based systems, collaborative filtering-based systems, and hybrid systems that are a mixture of the previous two. The collaborative filtering algorithm is described and implemented in more detail here. Collaborative filtering systems are divided into those based on user-user relationship and those... More based on item-item relationship. Collaborative filtering works by calculating a similarity between a user or an item in a database using a mathematical function such as a cosine function, a Jaccard function, etc., and based on that similarity finds a number of the most similar users or items. It then uses their grades to generate item grades that the active user has not graded and creates recommendations based on that. A simple form of collaborative filtering is implemented using the Python programming language. Also, a measure was implemented to evaluate the RMSE recommendation system. RMSE compares actual and estimated scores and evaluates the recommendation system accordingly. Less

• Keywords: sustavi preporučivanja; suradničko filtriranje; kosinus funkcija; korisnik – korisnik odnos; sličnost; generiranje ocjena; Python; RMSE
• Keywords (english): recommendation systems; collaborative filtering; cosine function; user - user relationship; similarity; rating generation; Python; RMSE
• Language: croatian
• URN:NBN: urn:nbn:hr:168:590465
• 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 16:00:27