Institutional Repository of Vilnius University
Kredito rizikos vertinimas mašininio mokymosi klasifikavimo modeliais /
Abstract
dc:descriptionIn this bachelor thesis, we will examine five different machine learning classification models (logistic regression, decision tree, random forest, k - nearest neighbours method and XGBoost method), which will be used to classify loan data on whether they default or not. In the beginning we will review methods’ theoretical principles, their advantages and shortages. Besides that, we will also review models’ evaluation criteria. In the second part of this thesis, we will use "Python" programming language and it’s machine learning, statistics and data manipulation libraries. We will review and process simulated data about clients and their loans, in order to make the models even more accurate. After that we will construct the models in two phases. Firstly, we will build models with their default hyperparameters and later we will use purposefully chosen hyperparameters to achieve as much accuracy as we can. Lastly we will we will use evaluation criteria to evaluate models both individually and with each other.
Degree
thesis:*- Grantor dc:publisher
- Institutional Repository of Vilnius University
- Year dc:date
- 2022
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Kaukas, Kasparas,
Rights
dc:rights- Statement dc:rights
-
- info:eu-repo/semantics/openAccess
- Language dc:language
- lit
Identifiers
dc:identifier.*- Repository record dc:identifier
- https://repository.vu.lt/VU:ELABAETD146236521&prefLang=en_US