Ostvarenje sustava preporučivanja koristeći algoritam suradničkog filtriranja

undergraduate thesis

• Title (english): Recommendation Systems Implementation Using Collaborative Filtering Algorithm
• Author: Fran Vrankić
• Mentor: Klemo Vladimir (mentor)
• Committee member: Klemo Vladimir (predsjednik povjerenstva)
• Committee member: Adrian Satja Kurdija (član povjerenstva)
• Committee member: Davor Vukadin (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2021-09-14, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Rad sadrži implementaciju sustava preporučivanja metodom suradničkog filtriranja i pokriva značajke sustava preporučivanja općenito. Osnovne metode algoritma suradničkog filtriranja nalažu pronalazak ograničenog skupa najsličnijih stavki unutar skupa, ako se radi o korisnik-korisnik tipu algoritma traži se skup korisnika najvećih koeficijenata korelacije, a ako se radi o predmet-predmet pristupu traže se najsličnije stvari za danu stvar. Koeficijent korelacije između dvije stvake se računa Pearsonovom formulom, a zatim se varijacijama težinske metode računaju predikcije ocjena. Iz novonastalog skupa predviđenih ocjena bira se ograničen broj najvećih i preporučuje. Naposlijetku se metodom RMSE vrednuje kvaliteta dobivenog skupa predikcija.

Abstract (english)

The paper contains the implementation of the recommendation system using the method of collaborative filtering and covers features of the recommendation system in general. The basic methods of the collaborative filtering algorithm require finding a limited set of most similar things within the set. In user-user version of the algorithm, set of users with the highest correlation coefficients is required. In item-item case the most common items are sought. The correlation coefficient between two subjects is calculated by Pearson’s formula. The post rating predictions are calculated using variations of the weighted average model. From the newly created set of predicted ratings, a limited number of the highest ones are selected and recommended. Finally, the quality of the obtained set of predictions is evaluated by the RMSE method.

• Keywords: Suradničko filtriranje; sustav preporučivanja; obrada velikog skupa podataka; Pearsonov koeficijent korelacije
• Keywords (english): Collaborative filtering; recommendation system; mining of massive dataset; Pearson correlation coefficient
• Language: croatian
• URN:NBN: urn:nbn:hr:168:962642
• 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 16:28:35