$GTCH targets AI practical applications evaluations in various domains through its subsidiary companies and joint ventures. GTC is evaluating intelligent solutions, based on the Company’s AI technology, in the field of entertainment, media and sports. GTC hired a development team that are experts in soccer sport statistics and analysis domains for the purpose of developing a software system for GTC, project’s internal code name: smartGOAL.
The smartGOAL program is an intelligent, automatic analytics and prediction system for soccer game’s results. The program analyzes and predicts possible outcomes of soccer games results according to permutations, statistics, historical data, using advanced mathematical methods and machine learning technology. The program downloads ratios and forecasted soccer games results from expert websites including probabilities, estimations and statistics, taking into account heuristic-based assumptions in order to analyze and predict soccer game’s outcomes. smartGOAL program is aimed for soccer game’s results cognitive study, analysis, applying deep learning algorithms, big data models and mathematical techniques. Heuristic algorithms work with betting lines as they represent market price for each result and in a sense provide a “crowd wisdom”, analyzing the data like an expert human would do. The program maximizes results using hybrid strategies of integration, Monte-Carlo simulation and regression-based errors correction. smartGOAL seeks to optimize results for certain level of variation, according to operator choice, maintaining historical data with their analytics. Machine learning-driven predictions data is produced based on historical information, ongoing public league team’s data and advanced mathematical models analysis. GTC plans to further implement GBT’s AI technology into this domain enabling intelligent solutions in the sport sector.
“Considering the popularity of the game in the present world, we believe organizations will be interested in in prediction systems for the better performance of their teams. As interesting as it may seem, prediction of the results of a soccer game is a very hard task and involves a large amount of uncertainty. However, it can be said that the result of football is not a completely random event, and hence, we believe a few hidden patterns in the game can be utilized to potentially predict the outcome. Based on the studies of numerous researchers that is being reviewed in our study as well as of those done in the previous years, one can say that with a sufficient amount of data an accurate prediction system can be built using various machine learning algorithms. While each algorithm has its advantages and disadvantages, a hybrid system that consists of more than one algorithm can be made with the goal of increasing the efficiency of the system as a whole. There also is a need for a comprehensive dataset through which better results can be obtained. Experts can work more toward gathering data related to different leagues and championships across the globe which may help in better understanding of the prediction system. Moreover, the distinctive characteristics of a soccer player, as well as that of the team, can also be taken into consideration while predicting as this may produce a better result as compared to when all the players in a game are treated to be having an equal effect on the game. The more information the system is trained with, we believe the more accurate the predictions and analysis will be. One of our joint venture companies, GTC aimed to evaluate machine learning-driven applications in various fields, among them are entertainment, media and sports. We believe smartGOAL is an intelligent application that has the ability to change the world’s soccer field when it comes to analytics and game score predictions,” provided Danny Rittman, the Company’s CTO.
