prikaz prve stranice dokumenta Razvoj i analiza izvedbe sustava za preporuku
Access restricted to authenticated users
master's thesis
Razvoj i analiza izvedbe sustava za preporuku
2021. urn:nbn:hr:179:217867
Čikotić, Ante
University of Split
Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture
Department of Electronics and Computing

Cite this document

Čikotić, A. (2021). Razvoj i analiza izvedbe sustava za preporuku (Master's thesis). Split: University of Split, Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture. Retrieved from https://urn.nsk.hr/urn:nbn:hr:179:217867

Čikotić, Ante. "Razvoj i analiza izvedbe sustava za preporuku." Master's thesis, University of Split, Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture, 2021. https://urn.nsk.hr/urn:nbn:hr:179:217867

Čikotić, Ante. "Razvoj i analiza izvedbe sustava za preporuku." Master's thesis, University of Split, Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture, 2021. https://urn.nsk.hr/urn:nbn:hr:179:217867

Čikotić, A. (2021). 'Razvoj i analiza izvedbe sustava za preporuku', Master's thesis, University of Split, Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture, accessed 14 November 2024, https://urn.nsk.hr/urn:nbn:hr:179:217867

Čikotić A. Razvoj i analiza izvedbe sustava za preporuku [Master's thesis]. Split: University of Split, Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture; 2021 [cited 2024 November 14] Available at: https://urn.nsk.hr/urn:nbn:hr:179:217867

A. Čikotić, "Razvoj i analiza izvedbe sustava za preporuku", Master's thesis, University of Split, Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture, Split, 2021. Available at: https://urn.nsk.hr/urn:nbn:hr:179:217867

Please login to the repository to save this object to your list.
Metadata
TitleRazvoj i analiza izvedbe sustava za preporuku
Title (english)The development and analysis of recommender system
AuthorAnte Čikotić
MentorAna Kuzmanić Skelin (mentor)
GranterUniversity of Split
Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture
(Department of Electronics and Computing)
Split
Defense date and country2021-07-23, Croatia
Scientific / art field,
discipline and subdiscipline		TECHNICAL SCIENCES
Computing
Abstract
Ovaj diplomski rad bavi se razvojem i analiza računalnog sustava za preporuke. Dan je pregled metoda i algoritama koji se koriste u sustavima za preporuke („recommender systems“) s naglaskom na trima glavnim pristupima: sustav temeljen na pristupu filtriranja sadržaja, sustav temeljen na suradničkom filtriranju, te hibridni model. U praktičnom dijelu rada, obrađena su sve tri kategorije sustava. Za sustav temeljen na sadržaju korišteno je filtriranje po tekstualnim i numeričkim metapodacima. Za sustav temeljen na suradnji korištene su metode grupa sličnih korisnika i grupa sličnih predmeta, a za hibridni sustav napravljen je model koji u nekoliko navrata izmjenjuje filtriranje po sadržaju pa suradnji, kako bi se dobile što preciznije preporuke te izbjegao problem hladnog starta. U praktičnom dijelu rada obrađene su tri javno dostupne baze podataka: „MovieLens Dataset“, „IMDb movies extensive dataset“ i „MPST: Movie Plot Synopses with Tags“, pri čemu su korištene Pythonove biblioteke Scikit-learna i Plotlya.
Abstract (english)
This master thesis deals with the development and the analysis of a movie recommender system. An overview of methods and algorithms used in recommender systems are given, with an emphasis on three main approaches: content-based filtering system, collaborative filtering system, and a hybrid approach using collaborative filtering and content-based filtering. In the practical part of the thesis, all the three approaches are addressed. Content-based filtering by textual and numerical metadata was used for the content-based system. For a system based on collaborative filtering, the methods of a group of similar users and a group of similar objects were used, and in a hybrid approach a model was developed that alternates content based filtering and collaborative filtering, in order to obtain more precise recommendations and to avoid the problem of a cold start. Three publicly available databases were used in the analysis: "MovieLens Dataset", "IMDb Movie Extensive Dataset" and "MPST: Movie Plot Synopses with Tags". Recommender systems were built upon Python libraries: Scikit-Learn and Plotly.
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:179:217867
Study programmeTitle: Computing
Study programme type: university
Study level: graduate
Academic / professional title: magistar/magistra inženjer/inženjerka računarstva (magistar/magistra inženjer/inženjerka računarstva)
Type of resourceText
File originBorn digital
Access conditionsAccess restricted to authenticated users
Terms of useLicence In copyright icon
Created on2021-12-01 11:22:21