A novel nature inspired firefly algorithm with higher order neural network: Performance analysis

摘要: Abstract The applications of both Feed Forward Neural network and Multilayer perceptron are very diverse saturated. But the linear threshold unit feed forward networks causes fast learning with limited capabilities, while due to multilayering, back propagation errors exhibits slow training speed in MLP. So, a higher order can be constructed by correlating between input variables perform nonlinear mapping using single layer units for overcoming above drawbacks. In this paper, Firefly based neural has been proposed data classification maintaining avoids exponential increase processing units. A vast literature survey conducted review state art previous developed models. performance method tested various benchmark datasets from UCI machine repository compared other established Experimental results imply that is fast, steady, reliable provides better accuracy than others.