Neural network modelling of services and goods sales analysis

摘要

This article explores the benefits of using neural networks in order to process a large amount of data for different scientific research. The study of sales of goods and services of the regions of the Russian Federation for 2018 is presented as an example confirming the feasibility of using specified method. A brief review of used software, “Deductor”, and data mining tool, self-organizing Kohonen maps (TSO) is presented, and the methodology of neural network modelling of trading activity is also described. The research is limited by the reviewed specter of goods in sales, based on the data obtained from the state statistics website. As a result, the main directions of sales industries in different regions are established, and the main sales centers in Russia are identified. Successful analysis of the dynamics of sales of goods and services will help to establish the direction of development of different regions and to increase the volume of services for consumption.