Competitive Analysis of different Artificial Neural Network Models
2017
In this paper, three different models for classification of nonlinear problems, i.e., Spiking Neuron Model, Multiplicative Neuron Model and Multilayer Perceptron Model have been taken for analysis. The well known, Ex-OR problem is been taken for the comparative analysis of the above three Artificial Neural Network models. It was found that spiking neuron model can be considered the best model in context of various parameters of Artificial Neural Network like learning rate, execution time, number of iteration, time elapse in training etc. Further, Iris data set (ref: www.lib.uci.edu ) has been taken for the analysis using above three ANN models. Comparing the various parameters learning rate, execution time, number of iteration, time elapse in training etc. it was observed that learning rate, execution time, number of iteration, time elapse in training is minimum in the case of spiking neuron model. On the basis of analysis it can be said that spiking neuron model can be taken as best model for classification. Our study also justifies the earlier studies done by Deepak Mishra and et. al.[13][15].
Keywords:
- Correction
- Source
- Cite
- Save
- Machine Reading By IdeaReader
0
References
0
Citations
NaN
KQI