Published Paper Details:

Data mining and machine learning play an extensively crucial role in the application of medical diagnosis. Consequently, several forms of analysis are ongoing to better predict the diseases and improving the quality of diagnosis. The existing analysis suffers from high dimensionality and imbalanced class distribution problem. Due to this issue, various classification algorithms prejudge over the majority classes, while ignoring the minority classes. This leads to the misprediction while predicting some rarely possible diseases. This research paper aims to increase the classification accuracy of the various classifiers over healthcare data. This research proposed a hybrid approach of both Principle Component Analysis (PCA) and Synthetic Minority Oversampling Technique (SMOTE) to reduce the high dimensional imbalanced class distribution problem. Then the resultant dataset is applied to the various classifier and compared based on the evaluation metrics.