自组织特征映射网络的人口分类

he human brain to the external input of the feelings of neurons in response to some kind of order in accordance with, and reflect the excitement felt by some external physical characteristics. Although the human nerve system is the large number of nerve cells, but in different regions of space on the nerve cells have different division of labor. Self-organizing feature map network, also known as Kohonen networks, it is Teuvo Kohonen scholars from Finland in 1981. The network is connected to a wide array of neurons composed of teachers without self-organization, self-learning network. Kohonen that, in the space of neurons in different regions have different division of labor, when a neural network to accept outside input mode, will be divided into different regions, the regional input to the model with different effects, and this process Neural network is automatically completed. This article give a simple example of the classification of the population, according to the data sample to study and adjust its own weight in order to achieve the purpose of training to illustrate the characteristics of the Kohonen network.