Published Paper Details:

Loan prediction is a very common real-life problem that each retail bank faces at least once in its lifetime. If done correctly, it can save a lot of man hours at the end of a retail bank. Customers first apply for a home loan after that company validates the customer eligibility for the loan. The loan eligibility process (real time) can be automated based on customer details provided through for example filling an online application form. These details can be Gender, Marital Status, Education, Number of Dependents, Income, Loan Amount, Credit History and many more. Hence the goal is to identify the customer segments that are eligible and in fact germane for loan amounts so that they can be specifically targeted. In this first Phase, some preliminary operations were carried out to understand all the features clearly. Visualization of all the variables for instance Married, to determine whether the married tend to repay the loans back or the unmarried was done. Similarly Gender, Self Employed, Applicant Income, Credit History, and on Independent Variables like Education etc.. plotting Box and Bar Plots. All the variables were Normalized Before Plotting and cross checked if it is Normal using graphs. Some of the Bivariate Analysis Include Loan_Status and Education etc. Loan_status is our Target Variable. Data cleaning includes dealing with missing values Mean (without outliers) and Mode (with Outliers) Replacement Method. Also Correlation is visualized using Heatmap. In the Upcoming Phases, the plan is to progress further by performing feature engineering, Model Building using regression etc..