ABSTRACT

Expert system are used extensively in many domains. Heart disease diagnosis is a complicated process and requires high level of expertise. This paper describes aiming to develop a expert system for diagnosing of heart disease using support vector machine and feedforward backpropagation technique. Now a days neural network are being used successfully in an increasing number of application areas. This work includes the detailed information about patient and preprocessing was done. The Support Vector Machine (SVM) and feedforward Backpropagation technique have been applied over the data for the expert system. To make the system more authentic and reliable out of 300 patients 250 patients were used for training set and 50 for evaluation process. In conclusion, we have used two neural network techniques but we are getting just 50% to 60% output i.e. not reliable for the patient. This expert system data can also be applied to improve the accuracy the medicine using some other neural network techniques.