Development of a Web Application for Analyzing Sales of a Telecommunications Company Based on Data Mining Methods

This article describes the structure and functional content of the developed web application for telecommunication company sales research. The research is based on data mining methods. The created application supports the technology (methodology) of sales analysis, including the stages of data visualization (graphical analysis), identification of relationships between sales volume and factors affecting sales (correlation analysis), identification of key factors (principal component analysis) and clustering of company branches. Testing of the web application was conducted on real data about the activities of 70 branches of a telecommunications company. In the course of the analysis, 8 key factors that determine the volume of sales were identified from the original 28 factors. Taking into account the key factors, the branches were divided into three stable clusters and their description was performed. Hierarchical methods, k-means method, and c-means method were used for clustering. The results made it possible to conclude on the effectiveness of the developed methodology and software for sales analysis. The practical significance of the work lies in the possibility of direct use of the developed models and web application in the activities of enterprises of the telecommunications sector.