Introduction

Literature Review

Discussion

Conclusion and Future Works

Acknowledgements

References

Abstract

As the sports betting industry and technology have grown on a large scale, predicting the outcome of a sports match using technologies approach is now crucial. In fact, humans have a certain limitation when processing a large set of information. However, Artificial Intelligence techniques can overcome this issue. Furthermore, sports have a great amount of data to consider, thus, it is a great example of AI problem. A review of some research using different Artificial Intelligence techniques to predict a sport outcome is presented in this article. Different types of sports such as football, soccer, javelin throw, basketball, and horse race were analyzed, and showed distinct approaches to predict results. Finally, a framework to develop a system to predict the outcome of a soccer game based on AI is proposed, considering the present research review.

Introduction

Sports have been part of human lives for millenniums, and the interest to be more than just a simple audience sends us back to more than 2000 years ago, when the Greeks went to the Colosseum to bet on a gladiator of their choosing [1]. Since that time, the interest of trying to know the future in sports matches has increased. For example, commentators on television shows guessing which team will win the Super Bowl or trying to figure out the result of a Premier League match, people on social medias discussing who will win the World Tennis Championship. It is part of the human lives. In addition, sports gambling industry has grown to such an extent that by 2017, the total gross income of the gambling industry totaled $17.8 billion [2]. The interest for knowing the result of sports matches before its conclusion is now clear, and indispensable to the sports gambling industry. For instance, the technology has expanded exponentially and has elevated us to another level of comprehension, and Artificial Intelligence is changing the way sports prediction is seen.

Artificial intelligent algorithms span several branches of computer science. Some examples are pattern recognition, predictive systems, inference and data analytics. The past few years were very important for machine learning technologies, they had an aggressive expansion of its accuracy, and now artificial neural networks can outperform humans in many areas [3]. One area that computers overcame humans is predictions, for instance many people are using this instrument to predict sports outcomes. As several sports have extremely large number of characteristics that are directly related to the result, is difficult for humans to consider all the features and predict with high accuracy a sports match. In these situations, a high-performance technique is needed to deal with all the data, and that is where Artificial Intelligence goes in. The enormous advances in this technology make it possible to process an exorbitant amount of data to draw extremely useful conclusions.