NORMA eResearch @NCI Library

Corporate Bankruptcy Prediction using Machine Learning Techniques

Deshpande, Shantanu (2020) Corporate Bankruptcy Prediction using Machine Learning Techniques. Masters thesis, Dublin, National College of Ireland.

[img]
Preview
PDF (Master of Science)
Download (1MB) | Preview
[img]
Preview
PDF (Configuration manual)
Download (1MB) | Preview

Abstract

Corporate bankruptcy has been a cause of concern for business stakeholders including the management, investors etc. since last few decades and has been of interest among the researchers worldwide. It is not enough to rely on a single predictive model due to the vast number of factors responsible for bankruptcy and thus the challenge lies in identifying only the key factors that are more responsible. Another major hurdle is the high class imbalance within the data which hinders the model’s performance. Though there are numerous techniques that have been previously explored which include Decision Trees, SVM, Neural Networks etc. with different pre-processing strategies, we further take this research forward by using a novel combination of a Random Forest feature selection technique along with SMOTEENN hybrid resampling technique on Polish Bankruptcy dataset. Further we implement four classifiers, Random Forest, Decision Tree, KNN and AdaBoost on the transformed data and evaluate each model’s performance. The results show that under our proposed strategy, Random forest classifier showed highest accuracy of 89% whereas overall AdaBoost model performed better with Recall 73% and
Geometric Mean as 80%.

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. > Bankruptcy
Divisions: School of Computing > Master of Science in Data Analytics
Depositing User: Dan English
Date Deposited: 22 Jun 2020 13:09
Last Modified: 23 Jun 2020 09:15
URI: http://trap.ncirl.ie/id/eprint/4312

Actions (login required)

View Item View Item