Клієнт-серверне застосування для передбачення результатів спортивних змагань

Abstract

An algorithm for predicting the results of sporting events has been developed, using which it is possible to optimize the process of selecting events, thus increasing the profit from betting. The algorithm is implemented in the form of appropriate software. The client and server software was written in the JavaScript programming language. The Firestore database was selected to store user data. The developed algorithm was implemented on the server part of the project. Based on test data, experimental studies were conducted. The AutoML service was used to create a machine learning model to improve data communication. As a result, the correlation coefficient was increased by 20%, and the percentage of return on investment increased by an average of 15% and became positive.