Sustav za generiranje preporuka temeljen na analizi javno dostupnih podataka

master's thesis

• Title (english): Recommender System Based on the Publicly Available Data
• Author: Goran Antolić
• Mentor: Ljiljana Brkić (mentor)
• Committee member: Ljiljana Brkić (predsjednik povjerenstva)
• Committee member: Boris Vrdoljak (član povjerenstva)
• Committee member: Igor Mekterović (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2017-02-06, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Ovaj rad opisuje dizajn i razvoj preporučiteljskog sustava koji gradi profil korisnika na temelju podataka preuzetih s društvenih mreža. Ti podaci sadrže informacije o korisnikovim interesima i služe kako bi se svladao čest problem preporučiteljskih sustava zvan cold start. Izgrađeni korisnički profil koristi se kod pretraživanja sadržaja koji bi korisniku mogao biti relevantan. Preporučiteljski sustavi uobičajeno su vezani samo uz jednu domenu, dok se u ovom radu opisuje arhitektura sustava koja podržava preporučivanje sadržaja iz različitih domena. U web-aplikaciji korisniku se preporučuju filmovi, glazbeni izvođači i web-članci. Kako bi se za korisnika u budućnosti mogle generirati bolje preporuke, razvijen je modul za ocjenjivanje preporuka kojim se ažurira korisnički profil nakon što korisnik ocijeni preporučeni sadržaj.

Abstract (english)

This paper describes the design and development of a recommender system that generates the user's profile based on data extracted from social networks. These data include information about the user's preferences and are used to overcome a common problem of recommender systems called "cold start". Based on the generated user's profile, system is capable of finding relevant content for the user. While recommender systems are commonly associated with only one domain, this paper proposes a system architecture that supports content recommendation from more than one domain. The recommender system described in this paper is able to recommend three types of items: a movie, a musician/band and a web article. The user can rate recommended items. Ratings are used to refresh the user's profile which leads to better recommendation process in the future.

• Keywords: preporučiteljski sustav; cold start; društvena mreža; API; ASP.NET MVC; AlchemyAPI
• Keywords (english): recommender system; cold start; social network; API; ASP.NET MVC; AlchemyAPI
• Language: croatian
• URN:NBN: urn:nbn:hr:168:746607
• 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: 2019-03-13 18:38:32