Digital forecast of the efficiency of the enterprise based on the machine learning model Random forest

Abstract

Theoretical foundations of the economic development of the real sector of the economy, the issues of forecasting the efficiency of an individual enterprise have been studied. The scientific novelty lies in the fact that, based on the dynamics of external and internal factors, a machine learning model "Random Forest" ("Random Forest Regressor") was developed, with the help of which the predictive value of the value of the eff ective attribute "Net profi t" was formed. The relevance lies in the fact that in order to form a net profi t forecast for JSC Caustic, an artificial intelligence system was used, namely the ML "Random Forest Regressor" machine learning model, which made it possible to successfully solve a complex problem, due to the action of many factorial features on the parameter under study. The use of the proposed approach is especially relevant in the light of the May Decrees of the President of the Russian Federation, which indicated the main direction — the digitalization of the economy, as a vector for further movement in the "Strategy for Scientific and Technological Development of the Russian Federation". As you know, digital technologies will serve as the foundation for innovative transformations of the state. With the help of artificial intelligence "ML-model", the forecast value of the net profi t of the enterprise for 2022 was obtained, the value of which was: 6773451.99 thousand rubles.