Different Oversampling techniques to handle imbalance data in machine learning | SMOTE | Part3
Different Techniques to deal with Imbalanced Dataset (Imbalanced Classes) in Machine Learning using Oversampling
00:10 Random Over-Sampling and Random over-sampling with imblearn
01:37 SMOTE -Synthetic Minority Oversampling Technique
02:41 SMOTE-NC
03:35 Borderline-SMOTE
09:57 Borderline-SMOTE SVM
11:13 KMeansSMOTE
14:22 Oversampling with Adaptive Synthetic sampling(ADASYN)
16:03 Creating logistic Regression Model before Over-Sampling
16:17 Creating logistic Regression Model using Over-Sampling DataSet
17:20 Combining Oversampling and Undersampling to handle imbalance datasets
18:06 Resampling technique using right performance metric
19:10 Resampling technique using Penalize Algorithms (Cost-Sensitive Training)
21:32 Collect more and more Data to handle imbalance datasets
21:48 Treat the problem as anomaly detection to handle imbalance datasets
21:56 Model-based approach or weights to the cases that get misclassified to handle imbalance datasets | Random Forest Classifier Model and XG Boost Classifier Model
22:34 Use Cross Validation(CV) with Sampling to handle imbalance datasets
27:02 Advantage and Disadvantages of Over-sampling
27:22 Advantage and Disadvantages of Under-sampling
Please refer below link to access the code :
https://github.com/atulpatelDS/Youtub...