An Innovative Approach to Predict Loan Eligibility of a Customer in Bank by Comparing Random Forest Algorithm over Logistic Regression in terms of Accuracy

Authors

  • Ch.Venkata Sandeep
  • Dr.T. Devi

DOI:

https://doi.org/10.47750/pnr.2022.13.S04.210

Keywords:

Logistic Regression, Novel Random Forest, Machine Learning, Mean Accuracy, Loan Prediction, Analysis.

Abstract

Aim: To predict the loan of the person using Random Forest Algorithm (RF) over Logistic Regression Algorithm (LR). Materials and Methods: The existing model uses a logistic regression algorithm. The 20 sample values are used to find out the mean, std. deviation, std. error means. The proposed Novel Random forest algorithm uses 20 sample values where various statistical metrics are evaluated per group. A total of 40 samples are used to find out the mean, Std. deviation, std. error means between the groups. The Random Forest is a supervised learning model, it constructs solutions for different regression problems. It provides a high rate of accuracy by cross-validation. The sample size was measured as 20 per group using G power (80%). Results: The graph explains the comparison of the mean accuracy value with algorithms Novel Random Forest and Logistic Regression where the mean accuracy of the decision tree is about 70.5% and the mean accuracy value of the Logistic Regression is about 69.5%. The statistical significance p>0.05, since p=1.0 previls insignificance based on independent sample T-Test. Conclusion: The mean accuracy rate of the Novel Random Forest algorithm has been improved to 70.5% compared to Logistic Regression which is having around 69.5% mean accuracy. This suggests the
proposed system provides an accurate analysis for loan approval.

Downloads

Published

2022-10-07

Issue

Section

Articles

How to Cite

An Innovative Approach to Predict Loan Eligibility of a Customer in Bank by Comparing Random Forest Algorithm over Logistic Regression in terms of Accuracy. (2022). Journal of Pharmaceutical Negative Results, 1741-1747. https://doi.org/10.47750/pnr.2022.13.S04.210