Loan default prediction using ML

Keywords: Loan Default, Prediction,Machine Learning,Algorithms

Background

Nowadays loan default prediction had become a serious area for all the researchers in the financial industry, the study aims to predict the probability of loan default. It plays important role in the financial world over the years. Although it is quite gainful and useful for both the moneylenders and the debtors. It does, however, carry a great risk, which in the area of loan lending is referred to as Credit risk Throughout the year machine learning algorithms have been used to calculate and predict credit risk by assessing an individual’s historical data. In this journal, the author presents the analysis and predictions on the Loan Defaulters dataset. The author used a dataset consisting of many historical records of loan defaulters from a bank at different periods. The author aims to train a machine-learning model which may predict bank customers cum loan defaulters, based on multiple factors

Aim & Objectives

Aim
The main aim of this project is to find out whether the customer is a loan defaulter or not.
Objective
• Data Cleaning and Pre-processing
• Simple imputer for imputing the null values
• Data Visualization
• SMOTE technique for balancing the dataset
• Build the required ML Model

Research Methodology

Various classification algorithms like SVM,DT,KNN,Logistic Regression and Xgboost classifier

Software & Tools

Jupyter, Google Collab

— Preview truncated. The complete document (full chapters, references, diagrams, and appendices) is shared with clients as part of project delivery. —