Sustavi za preporučivanje zasnovani na suradničkom filtriranju

undergraduate thesis

• Title (english): Recommender Systems Based on Collaborative Filtering
• Author: Stipe Babić
• Mentor: Marin Šilić (mentor)
• Committee member: Marin Šilić (predsjednik povjerenstva)
• Committee member: Helena Čeović (član povjerenstva)
• Committee member: Adrian Satja Kurdija (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2022-07-01, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Rad se sastoji od opisa sustava za preporučivanje s naglaskom na sustave temeljene suradničkima filtriranje te na njihovim primjenama i izazovima. Objašnjena je podjela na sustave zasnovane na memoriji i modelu, a zatim se objašnjena i dva različita pogleda na sustave zasnovane na memoriji (item-item i user-user).
Implementirana su tri sustava za preporuku. Jedan od sustava temeljen je na user-user paradigmi, drugi na item-item paradigmi, a treći na modelu koji koristi dekompoziciju na singularne vrijednosti. Za sve modele generirane su preporuke referentnom korisniku.
Naposljetku je napravljen eksperiment koji testira različite algoritme i vrednuje njihovu točnost i brzinu pri treniranju i testiranju.
Ključne riječi: sustavi za preporučivanje, sustavi bazirani na suradničkom filtriranju, sustavi bazirani na memoriji, sustavi bazirani na modelu, generiranje preporuka.

Abstract (english)

The thesis consists of describing recommender systems with emphasis on collaborative filtering and its challenges and applications. It describes memory based and model-based systems and then presents two different views on memory-based systems (item-item and user-user).
Three recommender systems were implemented. One on user-user, the second is based on item-item similarity and the third model is based using singular value decomposition. Referent user got recommendations for all of three implemented models.
Finally, experiment was made on same dataset to test different algorithms and to grade its accuracy and speed while fitting model and testing it.
Keywords: recommender systems, collaborative filtering, memory-based recommender systems, model-based recommender systems, generating recommendations.

• Keywords: sustavi za preporučivanje; sustavi bazirani na suradničkom filtriranju; sustavi bazirani na memoriji; sustavi bazirani na modelu; generiranje preporuka
• Keywords (english): recommender systems; collaborative filtering; memory-based recommender systems; model-based recommender systems; generating recommendations
• Language: croatian
• URN:NBN: urn:nbn:hr:168:928001
• 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: 2023-01-15 09:42:09