Convolutional neural network as a method for choosing a specialty

Makhnev Sergei Alexandrovich, Shedikov Andrey Alexandrovich

This work is devoted to the use of convolutional neural networks for solving problems of finding dependence and the subsequent prediction of the results of text belonging to independent words. The purpose of this work is to develop a convolutional neural network and subsequent verification of the semantic-graphic method as input parameters. The task of the research is to consider the search for weights of each element of a text set using the product of the methods of semantic analysis TF and RIDF. The result is an analysis of the proposed method for further processing of input data, based on graphs of functions that have huge potential.

Keywords:convolutional neural, semantic analysis, TF and RIDF, function graphs, neural network input data.