Bank Credit Risk Analysis & Machine Learning Preparation

Bank Credit Risk Prediction System

A data analysis and machine learning preparation project focused on exploring customer financial data and preparing it for credit risk prediction.

The project covers the complete data preparation workflow, starting with data cleaning and preprocessing, followed by exploratory data analysis to understand customer financial characteristics and relationships between different variables.

The analysis includes univariate, bivariate, and multivariate analysis, along with outlier detection and visualization of important financial features.

Feature engineering was then applied to create meaningful financial indicators such as Loan Count, Debt-to-Income Ratio, and EMI-to-Income Ratio, in addition to creating binary indicators for different loan types.

Categorical variables were transformed using Ordinal Encoding and One-Hot Encoding, resulting in a fully numerical dataset ready for Machine Learning models.

Tools & Technologies:

Python

Pandas

NumPy

Matplotlib

Seaborn

Data Cleaning & Preprocessing

Exploratory Data Analysis (EDA)

Feature Engineering

Categorical Encoding

Machine Learning Preparation

Project Workflow:

Data Cleaning → Exploratory Data Analysis → Outlier Detection → Feature Engineering → Encoding → Machine Learning Ready Dataset