Skip to content

vijitVM/Lending-Club

Repository files navigation

Lending-Club

Description

  • Lending Club is a US Peer to Peer Loan Company. It enables the borrowers to create unsecured personal Loans between $1000 and $40,000.
  • The Loan period is for a standard three years. The investors can select the loans which they want to invest into based on information supplied about the borrower, grade of loan, loan purpose etc.
  • Lending club makes money by charging borrowers an origination fee and investors a service Fee.

Motivation

  • People save money in the banks but that offers them with lower interest rates.
  • Lending club is helping in transforming the banking sector by making credit more affordable and investing more rewarding.
  • Here we need to classify each of the borrower as a defaulter or not using the data collected when the loan has been given.

Problem Statement

  • To Classify if a borrower will default the loan using the borrower's financial history.
  • We need to predict the target variable as 1> paid off and 0 > charged off.
  • The metrics we use for the problem:
    • AUC – ROC Curve
    • Accuracy
    • Time Taken

Procedure

Here is the code for the Lending club data analysis where I have performed the following:

  • Preprocessing
  • Exploratory Data Analysis
  • Model evaluation using different Machine learing models which were as follows :
    • Linear Regression
    • Logistic Regression
    • Support Vector Machine
    • K nearest Neighbour
    • XG Boosting
    • Neural Network
  • Comparision of the Models and choose the best one based on three factors.