NORMA eResearch @NCI Library

Minimizing Credit Risk In Peer-to-Peer Lending Business Using Supervised Machine Learning Techniques

Ayantola, Akeem (2020) Minimizing Credit Risk In Peer-to-Peer Lending Business Using Supervised Machine Learning Techniques. Masters thesis, Dublin, National College of Ireland.

[thumbnail of Master of Science]
Preview
PDF (Master of Science)
Download (1MB) | Preview

Abstract

One of the major challenges facing the retail finance market including banks is the issue of credit risk. This process involves the evaluation of individual customers' historical transaction data to determine their credit worthiness. Hence, the decision-making process is largely influenced by the outcome of the credit risk evaluation. Data mining techniques have been applied on financial dataset and this has shown significant results. In this report, the dataset from a peer-to-peer company (LendingClub Corporation) has been utilized. An oversampling Technique (Resample) was applied to overcome the class imbalance in the dataset. Similarly, five machine learning algorithms have been implemented to train and build an efficient classification model for evaluating credit risk. These models include Logistic regression, Random Forest, Decision Tree, Naive-Bayes, and Ada-Boost. A feature importance plot was implemented where a new dataset containing only the top ten important features were obtained . All five models were retrained using this new dataset and resulted obtained. The results were evaluated on their Accuracy,Recall,precision,F1-Score and Auc-Score.The highest proportion of Charged-Off clients were correctly classified by Decision Tree and the result showed that Decision Tree performed best with an accuracy score of 87% and 96% recall respectively.There was improvement in the computation time and the model's performance especially with Random forest in the second experiment. This research will further enhance the operations of peer-to-peer lenders to identify potential defaulters and further minimize the risk associated with such decisions.

Item Type: Thesis (Masters)
Subjects: Q Science > QA Mathematics > Electronic computers. Computer science
T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science
Q Science > QA Mathematics > Computer software
T Technology > T Technology (General) > Information Technology > Computer software
H Social Sciences > HG Finance > Credit. Debt. Loans.
Divisions: School of Computing > Master of Science in Data Analytics
Depositing User: Dan English
Date Deposited: 23 Jun 2020 12:26
Last Modified: 23 Jun 2020 12:26
URI: https://norma.ncirl.ie/id/eprint/4317

Actions (login required)

View Item View Item