Analysis of eight volume pulse elements based on the BP neural network

摘要

To solve the problem of the low rate of recognition in current complex pulse recognition, this paper puts forward a new approach to it. The paper preprocess and analyze the pulse information by using the neural network and genetic algorithm. The algorithm system includes pulse information collecting, network training, simulant diagnosis, correlation analysis. Pearson's coefficient test shows the system is of high reliability and testing accuracy. This system is more effective to solve the problem of pulse elements recognition. The method that making collection analysis between waveform and finger sense factor will be helpful for further research on the formation mechanism.