An Enhanced Approach to Infer Potential Host of Coronavirus by Analyzing Its Spike Genes Using Multilayer Artificial Neural Network

Abstract

Numerous coronaviruses are capable of transmitting interspecies. In recent years, transmission of coronavirus created a panic situation in the whole world. Therefore it is very important to infer the potential host of coro- navirus. In this research work nineteen parameters computed from the spike genes of coronavirus has been analysed to infer the potential host of coron- avirus. An enhanced multilayer neural network approach is proposed to analyse the data. The proposed model is compared with the other exiting statistical predictors like decision tree predictor, Support vector machine predictor and PNN predictor. All the model shown the higher accuracy such as 82.051 % by SVM predictor, 85.256% by PNN predictor,94.872% by decision tree predictor, and the highest accuracy 95.% is shown by proposed Multilayer Perceptron Predictor.