Verification of Classification Model and Dendritic Neuron Model Based on Machine Learning

Artificial neural networks have achieved a great success in simulating the information processing mechanism and process of neuron supervised learning, such as classification. However, traditional artificial neurons still many problems slow difficult training. This paper proposes new dendrite model (DNM), which combines metaheuristic algorithm effectively. Eight learning algorithms including backpropagation, classic evolutionary biogeography-based optimization, particle swarm genetic algorithm, population-based incremental competitive differential evolution, state-of-the-art jSO are used for training dendritic model. The optimal combination user-defined parameters has been systemically investigated, four different datasets involving classification problem investigated using proposed DNM. Compared with common machine methods decision tree, support vector machine, k-nearest neighbor, networks, trained by optimization significant advantages. It characteristics simple structure low cost can be to solve practical high precision.