Classification model for Statlog heart disease prediction through evolutionary feature selection and GMDH neural network

Heart disease prediction is a critical task regarding human health. In order to drop its rate, effective and timely diagnosis of the disease is very essential. Machine Learning methods have been developed to perform impressive predictions and make appropriate decisions. So, simulated annealing search algorithm along with GMDH neural network is introduced to manage the features present in the earlier heart disease classification system. The dimensionality of the features are reduced according to the behavior of simulated annealing search algorithm. The selected features are processed by GMDH neural network classifier. From the obtained results, the proposed model SA-GMDH shows an increase in the classification accuracy by obtaining more than ۸۹.۵۸% when compared to the other feature selection methods

کلیدواژه ها:

Heart disease prediction ، Effective diagnosis ، simulated annealing ، GMDH neural network