Prediction of myocardial infarction using K-medoid clustering algorithm

Data Mining is a technique for obtaining hidden information from various databases. Data mining method is helpful for analyzing a data from various proportions and for determining the relationships. Now a day's Myocardial Infarction (MI) is a very familiar disease in the world. The main goal of this research is to predict the myocardial infarction using data mining clustering techniques. This system can recognize and select hidden intelligence from historical myocardial infarction datasets. Clustering is the method of grouping a data object that is corresponding to one another within the cluster. They even grouped different objects into another cluster. It is otherwise named as data distribution in some application because it divides enormous data set into groups according to the resemblance. This paper has been examined the prediction of heart attack with more quantity of input attributes. The organization exploits medical terms like age, gender, blood pressure, cholesterol and so on for predicting the patients who obtained heart disease. Feature Selection in data mining is an approach to minimizing the number of input attributes underestimation. Originally the number of attributes was 14 and after feature selection, the number of attributes was reduced to 8.