抽象的な

Secured Neuro Genetic Approach for Predicting the Risk of Heart Disease

N.G.Bhuvaneswari Amma, K.Malathi, P.Balasubramanian

Medical diagnosis is done mostly by doctors’ expertise and experience. But in some circumstances, it may lead to wrong diagnosis and treatment. In this paper, a medical diagnosis system is proposed to predict the risk of heart disease using a secured neuro genetic approach. The objective of secured data classification is to build accurate classifiers without disclosing private information in the data being mined. In this paper, the learning capabilities of neural network and the optimization capabilities of genetic algorithms are combined in order to give better classification. To securely compute the activation function ElGamal scheme is used and the data is vertically partitioned. The effectiveness of the classifier is verified by experiments on Cleveland Heart Disease Dataset provided by the University of California, Irvine (UCI) machine learning repository.