prikaz prve stranice dokumenta Neuronske mreže - pogled u blackbox
  Download
PDF 20.07 MB
master's thesis
Neuronske mreže - pogled u blackbox
Osijek: Josip Juraj Strossmayer University of Osijek, School of Applied Mathematics and Informatics, 2024. urn:nbn:hr:126:570998
Golobić, Nika
Josip Juraj Strossmayer University of Osijek
School of Applied Mathematics and Informatics

Institutional repository: Repository of School of Applied Mathematics and Informatics

Cite this document

Golobić, N. (2024). Neuronske mreže - pogled u blackbox (Master's thesis). Osijek: Josip Juraj Strossmayer University of Osijek, School of Applied Mathematics and Informatics. Retrieved from https://urn.nsk.hr/urn:nbn:hr:126:570998

Golobić, Nika. "Neuronske mreže - pogled u blackbox." Master's thesis, Josip Juraj Strossmayer University of Osijek, School of Applied Mathematics and Informatics, 2024. https://urn.nsk.hr/urn:nbn:hr:126:570998

Golobić, Nika. "Neuronske mreže - pogled u blackbox." Master's thesis, Josip Juraj Strossmayer University of Osijek, School of Applied Mathematics and Informatics, 2024. https://urn.nsk.hr/urn:nbn:hr:126:570998

Golobić, N. (2024). 'Neuronske mreže - pogled u blackbox', Master's thesis, Josip Juraj Strossmayer University of Osijek, School of Applied Mathematics and Informatics, accessed 30 November 2024, https://urn.nsk.hr/urn:nbn:hr:126:570998

Golobić N. Neuronske mreže - pogled u blackbox [Master's thesis]. Osijek: Josip Juraj Strossmayer University of Osijek, School of Applied Mathematics and Informatics; 2024 [cited 2024 November 30] Available at: https://urn.nsk.hr/urn:nbn:hr:126:570998

N. Golobić, "Neuronske mreže - pogled u blackbox", Master's thesis, Josip Juraj Strossmayer University of Osijek, School of Applied Mathematics and Informatics, Osijek, 2024. Available at: https://urn.nsk.hr/urn:nbn:hr:126:570998

Metadata
TitleNeuronske mreže - pogled u blackbox
Title (english)Neural Network - Insight Into Blackbox
AuthorNika Golobić
MentorNenad Šuvak (mentor)
Committee memberDanijel Grahovac (predsjednik povjerenstva)
Committee memberNenad Šuvak (član povjerenstva)
Committee memberMirta Benšić (član povjerenstva)
GranterJosip Juraj Strossmayer University of Osijek
School of Applied Mathematics and Informatics
Osijek
Defense date and country2024-10-28, Croatia
Scientific / art field,
discipline and subdiscipline		NATURAL SCIENCES
Mathematics
Applied Mathematics and Mathematical Modeling
Abstract
Ovaj diplomski rad istražuje neuronske mreže i njihovu primjenu u procjeni kreditnog rizika. U radu se objašnjavaju osnovni pojmovi neuronskih mreža, počevši od arhitekture jednoslojnog i višeslojnog perceptrona, te procesa treniranja modela uz korištenje različitih optimizacijskih algoritama. Također se razmatraju složenije mreže, poput konvolucijskih i rekurentnih neuronskih mreža, kao i izazov interpretacije tih modela, poznatih kao blackbox. Rad ističe važnost tehnika interpretacije rezultata, s naglaskom na LIME metodu, koja omogućuje bolje razumijevanje odluka modela. Na kraju, izrađen je model neuronske mreže za procjenu kreditnog rizika, čija je učinkovitost analizirana i interpretirana primjenom LIME metode.
Abstract (english)
This thesis explores neural networks and their application in credit risk assessment. The work explains the basic concepts of neural networks, starting with the architecture of single-layer and multi-layer perceptrons, and the process of training models using various optimization algorithms. Furthermore, it also discusses more complex networks, such as convolutional and recurrent neural networks, as well as the challenge of interpreting these models, known as black box models. The thesis emphasizes the importance of result interpretation techniques, focusing on the LIME method, which allows for a better understanding of model decisions. Finally, a neural network model for credit risk assessment is developed, and its effectiveness is analyzed and interpreted using the LIME method.
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:126:570998
Study programmeTitle: Mathematics; specializations in: Financial and Statistical Mathematics, Mathematics and Computer Science, Industrial and Applied Mathematics
Course: Financial and Statistical Mathematics
Study programme type: university
Study level: graduate
Academic / professional title: magistar/magistra matematike (magistar/magistra matematike)
Type of resourceText
File originBorn digital
Access conditionsOpen access
Terms of useLicence In copyright icon
Created on2024-10-30 12:00:55