A System for Diagnosing Hepatitis Based on Hybrid Soft Computing Techniques Das Tapan Kumar1, Mohapatro Arati2 1Associate Professor, School of Information Technology & Engineering, VIT, Vellore 2Research Scholar, Bharathiar University, Coimbatore Online published on 16 March, 2018. Abstract Analysis of healthcare data is very crucial as it is related to human health, and the automatic classification of medical data reduces the burden of doctors for prediction. In this context, hepatitis data are considered. Classification techniques like decision tree, support vector machine, k-means and artificial neural network are being chosen for the comparison purpose. These algorithms are compared based on the parameters’ accuracy, precision, specificity and sensitivity. In the second embodiment, this paper presents a hybrid system which would detect hepatitis based on the technique of rough set based rule induction and neural network. LEM2 algorithm is used for decision rule induction while multi-layer perceptron is employed for classification. The proposed model performance is compared with standard classifiers, and the accuracy of the proposed system found to be better than other classifiers. Top Keywords Hepatitis, Classification Techniques, Accuracy, Rule Induction. Top |