prikaz prve stranice dokumenta Modeli dubokog učenja za upareno rangiranje
No public access
master's thesis
Modeli dubokog učenja za upareno rangiranje
2019. urn:nbn:hr:168:360726
Ljubičić, Karmela
University of Zagreb
Faculty of Electrical Engineering and Computing

Cite this document

Ljubičić, K. (2019). Modeli dubokog učenja za upareno rangiranje (Master's thesis). Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing. Retrieved from https://urn.nsk.hr/urn:nbn:hr:168:360726

Ljubičić, Karmela. "Modeli dubokog učenja za upareno rangiranje." Master's thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, 2019. https://urn.nsk.hr/urn:nbn:hr:168:360726

Ljubičić, Karmela. "Modeli dubokog učenja za upareno rangiranje." Master's thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, 2019. https://urn.nsk.hr/urn:nbn:hr:168:360726

Ljubičić, K. (2019). 'Modeli dubokog učenja za upareno rangiranje', Master's thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, accessed 04 November 2024, https://urn.nsk.hr/urn:nbn:hr:168:360726

Ljubičić K. Modeli dubokog učenja za upareno rangiranje [Master's thesis]. Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing; 2019 [cited 2024 November 04] Available at: https://urn.nsk.hr/urn:nbn:hr:168:360726

K. Ljubičić, "Modeli dubokog učenja za upareno rangiranje", Master's thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, Zagreb, 2019. Available at: https://urn.nsk.hr/urn:nbn:hr:168:360726

Please login to the repository to save this object to your list.
Metadata
TitleModeli dubokog učenja za upareno rangiranje
Title (english)Deep Learning Models for Pairwise Ranking
AuthorKarmela Ljubičić
MentorZvonko Kostanjčar (mentor)
Committee memberZvonko Kostanjčar (predsjednik povjerenstva)
Committee memberMarko Subašić (član povjerenstva)
Committee memberTomislav Petković (član povjerenstva)
GranterUniversity of Zagreb
Faculty of Electrical Engineering and Computing
Zagreb
Defense date and country2019-07-12, Croatia
Scientific / art field,
discipline and subdiscipline		TECHNICAL SCIENCES
Computing
Abstract
U ovom radu predstavljena je alternativna metoda za upareno rangiranje ostvarena modelima dubokog učenja. Implementirana je uparena duboka neuronska mreža. Za treniranje ovog modela ulazni parovi primjera kreirani su iz skupova podataka za procjenu kreditnog rizika. Opisano je pet različitih načina uparivanja primjera iz skupa podataka. Predloženi modeli za upareno rangiranje trenirani su za niz različitih hiperparametara. Rezultati ovih modela evaluirani su mjerom Somers' D i uspoređeni su s rezultatima uobičajenih klasifikacijskih modela binarne logističke regresije i duboke neuronske mreže.
Abstract (english)
This paper presents an alternative method for pairwise ranking achieved by deep learning models. A deep neural network for pairwise ranking was implemented. Datasets for credit risk assessment were used to produce input pairs for the implemented model. In this thesis, five different methods for data pairing were described. Proposed models for pairwise ranking were trained with a variety of hyperparameters. Their evaluated Somers' D measures were compared with those of usual binary logistic regression and deep neural network classification algorithms.
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:168:360726
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 conditionsClosed access
Terms of useLicence In copyright icon
Public note
Created on2020-01-07 19:54:19