Institutional Repository of Vilnius University
Finansinės kredito rizikos analizė naudojant mašininio mokymosi metodus /
Abstract
dc:descriptionOne of the most important responsibilities of financial organisations is the assessment of credit risk, which takes into account the likelihood that borrowers will not be able to repay their loans. In recent years, interest in artificial intelligence and its use in the financial sector has grown rapidly. The main advantages of machine learning and artificial intelligence are the ability to analyse large amounts of data and to identify recurring behavioural patterns in the data, thus overcoming the shortcomings of conventional credit risk assessment methods, and the ability to predict credit risk with greater accuracy. This paper explores the use of machine learning techniques, neural network models for financial credit risk assessment. The applicability of the topic is discussed, highlighting the value of consistent risk assessment in the financial sector. The main problem of the paper is whether neural network models can accurately predict credit risk, borrower solvency and probability of default using a variety of financial variables. The aim of this study is to develop a machine learning model that can adequately identify complex relationships and correlations in credit data, including elements such as income and employment status. The challenge is to develop a neural network that can quickly process and analyse large volumes of data while accurately determining the probability of solvency. The conclusions of the model evaluation and sensitivity analysis are reviewed, providing details on the functionality and robustness of the developed models. The implications of the findings for financial risk management are explored, highlighting the promising advantages and disadvantages of applying machine learning methods for credit risk assessment.
Degree
thesis:*- Grantor dc:publisher
- Institutional Repository of Vilnius University
- Year dc:date
- 2023
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Lutz, Filip Aleksandr,
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:ELABAETD192842916&prefLang=en_US