-
Language: Python​
-
Libraries: NumPy, Pandas, Matplotlib, Scikit-learn
-
Platform: Jupyter Notebook​
Credit Risk Detection
Purpose
Predict the credit risk (good or bad) of individual based on credit history, gender, marital status, annual salary, existing credits and much more
Models
Logistic Regression
Decision Tree Classifier
Multi Layer Perceptron
Applications
FinTech - as a rapidly evolving area, largely uses credit risk analysis systems to detect the risk involved with the customers when they apply for a credit card.
Tasks Done
Data Preprocessing
Data Normalization
Correlation Analysis
Feature Selection
Train-Test Split
Model Training and Testing
Model Evaluation
Dataset
German credit dataset of 1000 records with 20 attributes each with 'status' representing 1 for Good and 2 for Bad credit risk