Prediction Model for Loan Applications

Introduction:

This is a binary classification problem where we have information about a sample of applicants and we need to predict whether or not to grant a loan based on that data.

Table of contents

• Data Visualization .
• Feature selection and feature engineering
• Some techniques for data processing
• Handling missing data .
• Handling of categorical data and numerical data .
• outlier detection
• model evaluation

What we'll be using?

• The following libraries: sklearn, matplotlib, numpy, pandas, seaborn, scipy

• For handling missing data, we'll be using the backward 'bfill' method for numerical data and the most frequent value for categorical data.

• 4 different models:

  a) Logistic Regression

  b) KNeighborsClassifier

  c) SVC

  d) DecisionTreeClassifier

Here we go!

See the code