Integration of machine learning into web platforms for automating forecasting calculations

Abstract

In the context of digitalization of the economy and active implementation of intelligent technologies in various business areas, there is a need to create flexible tools that can effectively analyze data and provide accurate forecasts. Today, the integration of machine learning into the web environment opens up new opportunities for increasing productivity and automation. The article is devoted to the study of the combination of modern machine learning methods and web technologies for automating forecasting processes using the example of determining the cost of diamonds. The article provides an analysis of publications on the contribution of machine learning to various areas of activity. The authors also studied the experience of using machine learning to estimate the prices of jewelry, which showed the feasibility of the selected task. The paper proposes a mathematical model based on regression methods that takes into account key parameters for determining the cost of diamonds. The proposed mathematical model was implemented in a web application, the implementation of which was carried out using modern technologies, in particular Node.js, which allows for scalability, accessibility, and ease of use. The article presents a general and detailed scheme of the developed web application for determining the cost of a diamond. The authors conducted a study, the results of which confirmed the compliance of the web application with the theoretical calculations carried out in Anaconda. A comparison of the obtained results showed minor fluctuations in the cost of diamonds, which is explained by the degree of accuracy of training and some data anomalies. Therefore, in the future, it is planned to integrate and compare different machine learning models to increase accuracy and expand the functionality of the platform. In general, the obtained result confirmed that the implementation of machine learning models to web platforms will allow to obtain an effective, convenient and affordable tool and will contribute to increasing the level of automation of business processes