A Comparison of Two Oversampling Techniques (SMOTE vs MTDF) for Handling Class Imbalance Problem: A Case Study of Customer Churn Prediction.
Adnan AminFaisal RahimImtiaz AliChangez KhanSajid AnwarPublished in: WorldCIST (1) (2015)
Keyphrases
- class imbalance
- customer churn
- class distribution
- prediction model
- churn prediction
- active learning
- minority class
- cost sensitive
- cost sensitive learning
- class imbalanced
- majority class
- imbalanced datasets
- prediction accuracy
- high dimensionality
- imbalanced data
- imbalanced data sets
- sampling methods
- feature selection
- customer relationship management
- concept drift
- regression model
- classification accuracy
- training set
- training data
- high dimensional
- feature extraction
- neural network