Artificial intelligence

Abstract: The article is devoted to investigation of the problem of informativity dump input data at the stage of artificial neural network synthesis for solution of the task of time series prediction. The new approach to solve this task is offered which is based on neural network sensitivity to input data by analysis of their static characteristics. The new approach to solve actual task of development index which is defined the degree of sensitivity neural network to values of input data. It becomes possible to indicate the sort of input data owing to development such index which is processed the most effective by neural network. The efficiency of the offered index is justified and proved by theoretical calculations and illustrated obviously in the pictures. Perspective of further development of this direction research is indicated.

Keywords:

References: