BIG DATA, NEURAL NETWORK AND PREDICTIVE ANALYTICS: APPLICATION IN THE FIELD OF SPORT

The potential of big data and neural network to improve the prediction of sport outcomes is tremendous. In this study, the Recurrent Neural Network (RNN), like a class of artificial neural network, has been investigated for predicting the outcomes of football matches. The aim of this paper is to focus on the application of the neural network for predictive analysis purposes in the field of sport. Classification is a task that is often encountered in our life. A classification process involves assigning objects into predefined groups or classes based on a number of observed attributes related to those objects. Although there are some more traditional tools for classification, such as certain statistical procedures, neural networks have shown to be an effective solution for this type of problems. Neural networks classify objects rather simply - they take data as input, derive rules based on those data, and make decisions. The methodology of this study is based on data from the English Premier League for the period 1993-2017. The analyzed database includes: Full Time Home Team Goals, Full Time Away Team Goals, Half Time Home Team Goals, Half Time Away Team Goals, Home Team Shots on Target and Away Team Shots on Target. In this paper we have studied several different ways of forming up input data sequences, as well as different architectures of RNNs that may lead to effective prediction. The test results have shown that neural networks may be used for successfully predicting the outcomes of football matches. For further increasing the performance of the prediction, prior information about each team would be desirable.