Skip to content

The project predicts the probability of loan default using various financial features of customer. I applied SMOTENN by combining SMOTE and Edited Nearest Neighbor (ENN) to handle class imbalance. Logistic Regression, Random Forest and CATBOOST models have been apllied and evaluated based on accuray, F1 score, ROC-AUC score.

Notifications You must be signed in to change notification settings

muhammadfhaider12/loan-default-prediction

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

3 Commits
 
 
 
 

Repository files navigation

loan-default-prediction

Objective:

The project focuses one one of the major business problem the financial sector is facing is fraud detection especially loan default prediction. One of fact says that the typical financial organization losses 5% of their revenue because of the fraud. Moreover, the estimation of frauds hampering the UK econonmy is of £53 Billion per annum. Therefore, in this project financial data has been analysed and trained using the machine learning algorithm to predict the loan default based on the various financial features.

Data Source:

Lending Club

Models:

  1. Logistic Regression: 97%
  2. Random Forest Algorithm: 96%
  3. CATBOOST Model: 98%

About

The project predicts the probability of loan default using various financial features of customer. I applied SMOTENN by combining SMOTE and Edited Nearest Neighbor (ENN) to handle class imbalance. Logistic Regression, Random Forest and CATBOOST models have been apllied and evaluated based on accuray, F1 score, ROC-AUC score.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published