A fuzzy classification system for prediction of the results of the basketball games

Prediction of the sports game results is an interesting topic that has gained attention lately. Mostly there are used stochastical methods of uncertainty description. In this work it is presented a preliminary approach to build a fuzzy model to basketball game results prediction. Ten fuzzy rule learning algorithms are selected, conducted and compared against standard linear regression with use of the KEEL system. Feature selection algorithms are applied and a majority voting is used in order to select the most representative features.