Authors :
Himanshi Sharma; Ishika Tyagi; Gauri Agarwal; Deeksha Gupta
Volume/Issue :
Volume 8 - 2023, Issue 3 - March
Google Scholar :
https://bit.ly/3TmGbDi
Scribd :
https://bit.ly/3znD0C0
DOI :
https://doi.org/10.5281/zenodo.7783909
Abstract :
Although there are various items that
banking systems can sell to make money, their primary
source of income is their credit card system. Now that
the banking industry is doing better, but because banks
only have so many assets to lend to, choosing who will
be a safer option for the bank and to whom the loan can
be provided is usually a procedure. The banking sector
still desires a more rigorous predictive modelling
framework for a number of issues. Predicting loan
defaulters is a difficult task for the banking sector. The
loan status, which is the first stage of the loan lending
procedure, is one of the quality metrics of the loan.
Using machine learning, it is possible to automate the
process of determining whether a loan should be
authorized or not to the loan asker. This is done in more
efficient way by searching through available data for
prior loan recipients, after which machine learning
methods are used to train the system based on the
histories and experiences on available data. There are
several ways to analyze the former mentioned issues on
loan prediction in accordance with the research
conducted by many researchers in this era. In this
research paper we basically conducted the exhaustive
investigation on DGHI dataset for analyzing the
customer eligibility whether he is eligible for loan or not
using LRD machine learning algorithms (i.e. Logistic
Regression, Random Forest and Decision Trees). The
experimental study conducted has been divided into two
phases: Training and Testing of the available data. On
the basis of investigation conducted we decided to
choose Logistic Regression as the best technique for
probability of loan prediction for the customer. The
results obtained and selection of Logistic Regression as
the suitable technique for the given approach has been
done on the basis of parameters such as: Loan_id,
Gender, Married, Education, Self-employed and so on.
For future work it has been decided to improve the
accuracy and precision of Logistic Regression.
Keywords :
Decision Tree, Logistic Regression, Machine Learning algorithms, Medical Insurance, Prediction, Random Forest.
Although there are various items that
banking systems can sell to make money, their primary
source of income is their credit card system. Now that
the banking industry is doing better, but because banks
only have so many assets to lend to, choosing who will
be a safer option for the bank and to whom the loan can
be provided is usually a procedure. The banking sector
still desires a more rigorous predictive modelling
framework for a number of issues. Predicting loan
defaulters is a difficult task for the banking sector. The
loan status, which is the first stage of the loan lending
procedure, is one of the quality metrics of the loan.
Using machine learning, it is possible to automate the
process of determining whether a loan should be
authorized or not to the loan asker. This is done in more
efficient way by searching through available data for
prior loan recipients, after which machine learning
methods are used to train the system based on the
histories and experiences on available data. There are
several ways to analyze the former mentioned issues on
loan prediction in accordance with the research
conducted by many researchers in this era. In this
research paper we basically conducted the exhaustive
investigation on DGHI dataset for analyzing the
customer eligibility whether he is eligible for loan or not
using LRD machine learning algorithms (i.e. Logistic
Regression, Random Forest and Decision Trees). The
experimental study conducted has been divided into two
phases: Training and Testing of the available data. On
the basis of investigation conducted we decided to
choose Logistic Regression as the best technique for
probability of loan prediction for the customer. The
results obtained and selection of Logistic Regression as
the suitable technique for the given approach has been
done on the basis of parameters such as: Loan_id,
Gender, Married, Education, Self-employed and so on.
For future work it has been decided to improve the
accuracy and precision of Logistic Regression.
Keywords :
Decision Tree, Logistic Regression, Machine Learning algorithms, Medical Insurance, Prediction, Random Forest.
