prikaz prve stranice dokumenta Sustavi preporuka glazbe prema sadržaju
  Download
PDF 1.29 MB
master's thesis
Sustavi preporuka glazbe prema sadržaju
Split: University of Split, Faculty of Science, 2024. urn:nbn:hr:166:122377
Klarin, Danijel
University of Split
Faculty of Science
Department of Informatics

Institutional repository: Repository of Faculty of Science

Cite this document

Klarin, D. (2024). Sustavi preporuka glazbe prema sadržaju (Master's thesis). Split: University of Split, Faculty of Science. Retrieved from https://urn.nsk.hr/urn:nbn:hr:166:122377

Klarin, Danijel. "Sustavi preporuka glazbe prema sadržaju." Master's thesis, University of Split, Faculty of Science, 2024. https://urn.nsk.hr/urn:nbn:hr:166:122377

Klarin, Danijel. "Sustavi preporuka glazbe prema sadržaju." Master's thesis, University of Split, Faculty of Science, 2024. https://urn.nsk.hr/urn:nbn:hr:166:122377

Klarin, D. (2024). 'Sustavi preporuka glazbe prema sadržaju', Master's thesis, University of Split, Faculty of Science, accessed 07 July 2024, https://urn.nsk.hr/urn:nbn:hr:166:122377

Klarin D. Sustavi preporuka glazbe prema sadržaju [Master's thesis]. Split: University of Split, Faculty of Science; 2024 [cited 2024 July 07] Available at: https://urn.nsk.hr/urn:nbn:hr:166:122377

D. Klarin, "Sustavi preporuka glazbe prema sadržaju", Master's thesis, University of Split, Faculty of Science, Split, 2024. Available at: https://urn.nsk.hr/urn:nbn:hr:166:122377

Metadata
TitleSustavi preporuka glazbe prema sadržaju
Title (english)Music content-based recommendation systems
AuthorDanijel Klarin
MentorSaša Mladenović (mentor)
Committee memberSaša Mladenović (predsjednik povjerenstva)
Committee memberDivna Krpan (član povjerenstva)
Committee memberGoran Zaharija (član povjerenstva)
GranterUniversity of Split
Faculty of Science
(Department of Informatics)
Split
Defense date and country2024-04-12, Croatia
Scientific / art field,
discipline and subdiscipline		TECHNICAL SCIENCES
Computing
Abstract
U ovom diplomskom radu korišteni su algoritmi strojnog učenja – k-NN, K-Means i DBSCAN u sustavu preporuka prema sadržaju na primjerima pjesama. Korišten je skup podataka istrugan s vlastito kreiranog popisa pjesama sa glazbenog servisa Spotify. Korištene su tehnike pri obradi podataka: min-max skaliranje i dimenzioniranje uz pomoć algoritama PCA i t-SNE. Kod k-NN provedeni su testovi uz različite udaljenosti, algoritme pretraživanja i vremensku komponentu, dok su kod K-Means i DBSCAN algoritma korišteni testovi svojstveni za nenadzirano učenje. Sve kako bi se ocijenilo koji algoritmi, tehnike i parametri dovode do najboljih rezultata preporuke.
Abstract (english)
In this master's thesis, machine learning algorithms - k-NN, K-Means, and DBSCAN - were used in a content-based recommendation system using examples of songs. The dataset was extracted from a self-created list of songs from the Spotify music service. Data processing techniques included min-max scaling and dimensionality reduction using PCA and t-SNE algorithms. For k-NN, tests were conducted with different distances, search algorithms, and a time component, while for K-Means and DBSCAN algorithms, tests specific to unsupervised learning were used. All in order to evaluate which algorithms, techniques, and parameters lead to the best recommendation results.
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:166:122377
Study programmeTitle: Data Science and Engineering
Study programme type: university
Study level: graduate
Academic / professional title: sveučilišni/a magistar/magistra podatkovne znanosti i inženjerstva (sveučilišni/a magistar/magistra podatkovne znanosti i inženjerstva)
Type of resourceText
File originBorn digital
Access conditionsOpen access
Terms of useLicence In copyright icon
Created on2024-05-03 13:12:29