An approach for heart disease detection by enhancing training phase of neural network using hybrid algorithm
Battula Srinivasa Rao, Kudith Nageswara Rao, S. Pallam Setty · 2014
The disease diagnosis based on artificial intelligence techniques is an effective technique. To enhance the training procedure of the neural network to diagnose the heart disease effectively, we use a hybrid algorithm which is combination of GSO and ABC. Initially, we generate an initial population that has number of members and the members have the weight values which are used to train the neural network. To identify a perfect member to train the neural network, we use the hybrid algorithm operations. We give each member to the neural network and we find the fitness for each member and we categorize the members to perform the hybrid operations i.e. which member has to do which operation. After performing corresponding operations on the categorized members, we get a new set of members and we iterate the process until we get a stable member for producer operation. We choose the weight values of the producer to train the neural network to detect the heart disease.