Usporedba dubokog učenja i konvencionalnih metoda u modernim preporučiteljskim sustavima

master's thesis

• Title (english): Comparison Between Conventional and Deep Learning Methods in Modern Recommender Systems
• Author: Matteo Miloš
• Mentor: Damir Pintar (mentor)
• Committee member: Damir Pintar (predsjednik povjerenstva)
• Committee member: Zoran Skočir (član povjerenstva)
• Committee member: Mihaela Vranić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2019-07-09, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Preporučiteljski sustavi su područje koje u posljednje vrijeme ostvaruje velik napredak, zbog gotovo neograničene dostupnosti interneta. Tako su velike svjetske firme orijentirane na pružanje personaliziranog sadržaja svakom korisniku, kako bi njegovo iskustvo korištenja sadržaja bilo što pozitivnije, ali i kako bi istovremeno povisile profit. Donedavno, preporučiteljski sustavi su se oslanjali na klasične metode suradničkog filtriranja i filtriranja zasnovanog na sadržaju, uz manje modifikacije, međutim u posljednje vrijeme taj se trend mijenja te se u metode preporuke uvode i koncepti dubokog učenja. U ovom radu uspoređene su performanse klasičnih metoda i metoda dubokog učenja, odnosno neuronskih mreža.

Abstract (english)

Recommender systems are the field that has gained great improvement in recent times, mostly because of the almost unlimited availability of internet. Consequently, great companies are trying to give a personalized content to each user, so that his experience is as positive as possible, but also to simultaneously raise their profit. Until recently, recommender systems were using traditional methods as collaborative filtering and content-based filtering, but in recent times, that trend is changing, and deep learning concepts are being used more and more. In this thesis we have compared performances of traditional methods and methods of deep learning, more precisely neural networks.

• Keywords: nadzirano učenje; preporučiteljski sustavi; duboko učenje; neuronske mreže
• Keywords (english): supervised learning; recommender systems; deep learning; neural networks
• Language: croatian
• URN:NBN: urn:nbn:hr:168:850799
• 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: 2020-01-07 20:32:28