prikaz prve stranice dokumenta Sustav preporučivanja filmova zasnovan na sadržaju
No public access
undergraduate thesis
Sustav preporučivanja filmova zasnovan na sadržaju
Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing, 2021. urn:nbn:hr:168:889643
Vinković, Bartol
University of Zagreb
Faculty of Electrical Engineering and Computing

Institutional repository: FER Repository

Cite this document

Vinković, B. (2021). Sustav preporučivanja filmova zasnovan na sadržaju (Undergraduate thesis). Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing. Retrieved from https://urn.nsk.hr/urn:nbn:hr:168:889643

Vinković, Bartol. "Sustav preporučivanja filmova zasnovan na sadržaju." Undergraduate thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, 2021. https://urn.nsk.hr/urn:nbn:hr:168:889643

Vinković, Bartol. "Sustav preporučivanja filmova zasnovan na sadržaju." Undergraduate thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, 2021. https://urn.nsk.hr/urn:nbn:hr:168:889643

Vinković, B. (2021). 'Sustav preporučivanja filmova zasnovan na sadržaju', Undergraduate thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, accessed 26 June 2024, https://urn.nsk.hr/urn:nbn:hr:168:889643

Vinković B. Sustav preporučivanja filmova zasnovan na sadržaju [Undergraduate thesis]. Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing; 2021 [cited 2024 June 26] Available at: https://urn.nsk.hr/urn:nbn:hr:168:889643

B. Vinković, "Sustav preporučivanja filmova zasnovan na sadržaju", Undergraduate thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, Zagreb, 2021. Available at: https://urn.nsk.hr/urn:nbn:hr:168:889643

Metadata
TitleSustav preporučivanja filmova zasnovan na sadržaju
Title (english)Content-based Movie Recommendation System
AuthorBartol Vinković
MentorKlemo Vladimir (mentor)
Committee memberKlemo Vladimir (predsjednik povjerenstva)
Committee memberAdrian Satja Kurdija (član povjerenstva)
Committee memberLucija Šikić (član povjerenstva)
GranterUniversity of Zagreb
Faculty of Electrical Engineering and Computing
Zagreb
Defense date and country2021-06-29, Croatia
Scientific / art field,
discipline and subdiscipline		TECHNICAL SCIENCES
Computing
Abstract
Sustav preporučivanja na temelju sadržaja koristi mjere udaljenosti i sličnosti nad atributa objekta da izračuna sveukupnu sličnost dvaju objekata. Pojmovi sličnosti i udaljenosti su međusobno povezani u nekim slučajevima s izravnom relacijom. Kako bismo mogli napraviti takav sustav potrebno je pronaći prikladan skup podataka i filtrirati korisne podatke i urediti u smislenu cjelinu. Kako bismo bili sigurni da je naš sustav dobro izgrađen moramo provesti pokuse. Najdostupniji pokusi u offline pokusi za koje je samo potrebno imati korisničke podatke prema kojima možemo modelirati korisnika. Najčešći način evaluacije sustava preporučivanja je usporedba s drugim sustavom. Usporedbe radimo testove kojima povlačimo zaključke koji sustav ima bolje performanse, te statistički obradimo rezultate kako bismo bili sigurni da su rezultati pouzdani. Prilikom offline pokusa moramo modelirati korisnika, ali pripaziti da naš model ne odstupa previše od pravog ponašanja korisnika.
Abstract (english)
Content-based recommendation systems give recommendations using distance and similarity measures to determine the similarity of attributes of two objects and the total similarity. Similarity and distance are related terms, and in some cases have a direct relation with each other. In order to make such a system we need an appropriate dataset which we need to filter and organize. We need to conduct experiments to ensure that our system is working properly. The most available experiments are offline experiments for which we only need user-data that we use to model the users. The most common way of evaluating a recommender system is to compare it to another. The comparison is made with tests which determine which system better predicts user behavior. We conduct statistical analysis to determine that our results are repeatable. During the evaluation we need to make sure that our user model doesn't stray too much from real user behavior.
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:168:889643
Study programmeTitle: 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 resourceText
File originBorn digital
Access conditionsClosed access
Terms of useLicence In copyright icon
Public note
Created on2023-01-14 10:12:15