Default-Credit-Card-Prediction
Machine Learning Project for predicting Credit Card Defaults.
Features:
- Dataset Loading
- Feature Assessment/Visualization:
- Normalized Histogram Distribution
- Box Plots
- Pairwise Relationships
- Empirical Cumulative/Standard Density Functions
- Pearson Correlation
- 2D PCA
- 2D LDA
- Preprocessing:
- Standardization
- Scaling
- Normalization
- Dataset Balancing
- Feature Selection:
- (Filter) Information Gain
- (Filter) Gain Ratio
- (Filter) Chi-squared Test
- (Filter) Kruskal-Wallis Test
- (Filter) Fisher Score
- (Filter) Pearson Correlation (Feature-Feature, Feature-Class)
- (Filter) mRMR
- (Filter) Area Under the Curve (AUC)
- (Wrapper) Sequential Forward/Backward Selection
- (Wrapper) Recursive Feature Elimination
- Feature Reduction:
- Principal Component Analysis (PCA)
- Fisher's Linear Discriminant Analysis (LDA)
- Classification:
- Minimum Distance Classifier
- k-Nearest-Neighbors (kNN)
- Naive Bayes
- Support Vector Machines (SVM)
- Decision Tree (CART)
- Random Forest
- Evaluation:
- Stratified K-folds Cross Validation
- Receiver Operating Characteristic (ROC) Curves
- Precision-Recall Curves
Requirements:
Usage:
usage: python default_predictor.py
Examples: