Development of a Software Module for the Classification of News Texts Based on an Artificial Neural Network

The work is devoted to the development of a software module for classifying news texts based on an artificial neural network. The first chapter provides an overview of existing implementations of text classification methods based on artificial neural networks, as well as a brief description of the system under development, within which a software module is developed. The second chapter discusses vectorization algorithms and describes the most popular architecture models of artificial neural networks. The neural network architecture was developed for the specific task of classifying news texts with optimal hyperparameters. The third chapter demonstrates the development of a module for classifying news texts. The module has an application programming interface implemented in Python using the Flask framework. Key words: artificial neural networks, vectorization, neuron, layer, architecture. The final qualification work consists of 72 pages, including 8 tables, 26 figures, 4 appendices, and refers to 42 sources.