Loan Approval Prediction using Exploratory Da...

Loan Approval Prediction using Exploratory Data Analysis and Machine Learning

Author Name : Gourav Bhattacharjee, Tarash Budhrani , Tushar Khatri , Vishwam Gopal valiyaveettil , Arshdeep Singh , Divyanshi Uppal , Prakruthi N

ABSTRACT Predicting mortgages is beneficial for both bank employees and applicants. The Mortgage Prediction machine automates the calculation of the weight of each feature involved in the loan processing. On new test data, similar features are processed based on their respective weights. An applicant can set a deadline to check whether their loan will be sanctioned. The Mortgage Prediction system allows prioritized checking of specific applications. This project provides insights into solving real business problems through Exploratory Data Analysis (EDA). It delves into risk analytics in banking and financial services, showcasing how data minimizes the risk of financial loss during lending. In India, the increasing number of loan applications poses a challenge for bank employees to predict whether a customer is likely to repay the amount. This paper aims to explore the characteristics of customers applying for personal loans using exploratory data analysis. The dataset undergoes normalization, missing value treatment, column selection, filtering, derivation of new columns, identification of target variables, and visualization through graphical representations. Python, specifically the Pandas library, is employed for efficient data processing. Matplotlib is utilized for creating graphs, providing a visual representation of the results. The organization under consideration is a leading online loan marketplace, offering personal loans, business loans, and medical procedure financing. Identifying risky applicants through EDA is crucial to reduce credit loss caused by defaults. The project aims to identify driving factors or variables that strongly indicate loan default, providing valuable insights for portfolio management and risk assessment