An optimal criterion feature selection method for prediction and effective analysis of heart disease

In healthcare, there are vast areas wherein the prediction and analysis have been carried out for all the disease. Nowadays, the most common disease for the human being under risk is of cardiac disease. The idea here is to analyze the data set of variety of patients and to predict the chance of getting the heart attack is due to high blood pressure, the gene in the family circle, age factor. Among all the other disease, heart disease is the hazardous disease which leads to death. The heart attack occurs when the flow of blood to the heart is blocked, which contains fat, cholesterol and other substances in the arteries that feed the heart. The heart attack is the permanent damage or destroys part of the heart muscle; it creates a permanent scar on the heart. The proposed approach extracts the features from the dataset. Based on the features the decision table is constructed. Irrelevant attributes are removed by applying feature selection algorithm. Further, the dependency among the attribute towards identifying the disease is determined by using optimality criterion function. Hence the time taken to predict the heart disease is reduced compared to other algorithms. The dataset is collected from UCI and analyzed using the Optimality Criterion Feature selection algorithm. There are 14 attributes in the dataset, and data such as resting electrocardiography, chest pain type are the three attributes taken into consideration for making decisions.