Heart disease prediction is the major research carried out nowadays to find the disease in the chosen data set of medical records. A number of computer oriented techniques utilized for the identification of heart diseases in order to find whether the diseases available in the chosen data set or not. One such method is used in this research work namely Type 2-Fuzzy technique and it has been concerning as the finest way to decrease this indistinctness. Newly, numerous researches have been available in terms of Coronary Heart Disease (CHD) diagnosis. The approach used in this research work is assist in envisage illness probability which gives extract result. The Type-2 Fuzzy methodology conventions are applied to calculate the chances of Coronary Heart Disease (CHD) as minimum, average or maximum. Usually the doctors consider the following attributes such as diabetes, fatness, agitated anxiety, smolder, deprived fasting, tension, etc to predict Coronary Heart Disease (CHD). The Karinx machine learning model is applied to find the Coronary Heart Disease (CHD) and yields appropriate platform to predict heart diseases in fast and efficient manner. A comparative analysis is carried out in this work to find the performance of the Karinx model with Probabilistic Neural Network (PNN), Support Vector Machines (SVM), Fuzzy Adaptive Resonance Theory (FARTMAP), Adaptive Neuro Fuzzy Inference System (ANFIS), Knuth Morris Pratt (KMP) algorithms. This research work yields the best results with accuracy rate of 98.1% in order to find the coronary heart disease in the chosen data set.

Coronary Heart Disease, Illness Treatments, Disease Imprecision, Machine Learning, Type-2 Fuzzy Method