Handling-Imbalance-data

This repository will compare the performance of different classification algorithms on various imbalanced datasets using multiple balancing techniques.

Classification Algorithms:

• Logistic Rgression
• Support Vector Machines
• Decision Trees
• Random Forest
• Naive Bayes

Balancing Techniques

• SMOTE: We balanced the data in two ways a) Whole data b) Train data only
• ADASYN same as in SMOTE
• UNDER SAMPLING same as in SMOTE
• ALGORITHMIC WEIGHTAGE
• GAN same as in SMOTE

Datasets

• Fraud
• Churn
• BankMarketing