Optimized stacked classifier to predict Hepatitis B disease

Ensemble machine learning approaches are being used for classifying many tasks like sentiment analysis, stock price prediction, breast cancer prediction etc. Stacked classifier is a type of ensemble learning which used multiple classifiers for generating best classification. The paper presents optimized stacked classifier to predict Hepatitis B disease. Synthetic Minority Over-sampling Technique. (SMOTE) is used to handle class imbalance issue in the dataset. To optimize the parameters of Meta classifier grid search technique is used. The performance of stacked classifier is compared with performance of random forest, gradient boost and support vector machine classifier. Results show that the best accuracy is presented by stacked classifier which is 90% as compare to the accuracy presented by random forest, support vector machine and gradient boost algorithm.