Performance evaluation of classifiers for non communicable diseases

Non-Communicable Diseases (NCD) has been drastically increased across the world in the recent years. Diseases that are not transmissible from one person to another are known as Non-Communicable Diseases (NCDs). Parkinson's disease, Autoimmune Disorders, Strokes, Heart Diseases, Cancers, Diabetes are examples of NCDs. It is found that certain Non-Communicable Diseases are believed to be exacerbated by lifestyle and climate which cause of death in the world. Many Machine Learning methods such as Nave Bayes, Support Vector Machine have been applied to predict NCDs. The aim of this paper is to predict the Non-Communicable Diseases (NCDs) and evaluate the performance of the key classifiers in terms of Prediction accuracy, Kappa and F-Measure. The progress in computer training and AI can remarkably help the analysis of Non-Communicable Diseases (NCDs). Nevertheless, the intrinsic complication of black-box samples controls the understanding of the sample. Consequently, likely regulative concerns arise. Furthermore, the lack of faith inside the therapeutic society is obvious due to the absence of understanding of how and why a sample prophesies. This research illustrates how sample-cynical ways of explainable AI (XAI) can assist give descriptions to assume black-box samples on NCDs information collection better.