Complexiy Of The Algorithm Study And Application Of The Second Type Padé Weight Function Neural Network

Time complexity is an important measure of algorithm. The main purpose of this paper is toresearch the time complexity of the second category of Padéweight function neural network andfind out the factors which affect its time complexity.The research of this paper is the second type Padé weight function neural network algorithmcomplexity. The second type Padéweight function neural network is a new type of neural network.In this paper, with the theory of Newton interpolation Padéapproximation, according to the basicsof the second type spline function neural network, through in-depth research and analysis, thecomplexity of the second type Padéweight function neural network is deduced.Padé weight function neural network training algorithm time complexity is as follows:mnO(N2),Where m is the input dimension, n is the output dimension, N is the number of samples.Simulation results show the accuracy of the second type weight Padéfunction neural networkalgorithm complexity formula theoretical results than comparing the traditional neural network, thesecond spline weight function neural network has a faster training speed.The application on mail classification using the second type padéweight Function achievesgood classification results and also shows that the algorithm can solve the mail classificationproblems.