Application of machine learning models to predict the performance of government contracts

Abstract

Application of machine learning models to predict the performance of government contracts

Korchagin S.A., Rubtsov D.Yu., Serdechny D.V., Bespalova N.V.

Incoming article date: 04.08.2024

The work analyzes existing approaches to forecasting contract execution, including traditional statistical models and modern methods based on machine learning. A comparative analysis of various machine learning algorithms, such as logistic regression, decision trees, random forest and neural networks, was carried out to identify the most effective forecasting models.An extensive database of information on government contracts was used as initial data, including information about contractors, contract terms, deadlines and other significant factors. A prototype of an intelligent forecasting system was developed, testing was carried out on real data, as well as an assessment of the accuracy and reliability of the resulting forecasts. The results of the study show that the use of machine learning methods can significantly improve the quality of forecasting the execution of government contracts compared to traditional approaches

Keywords: intelligent system, mathematical modeling, government procurement, government contracts, software package, forecasting, machine learning