Feature selection using C4.5 algorithm for electricity price prediction

The electricity price forecasting is important in our daily life. It does not only benefit to the customers but also the providers since the pressure of the load station in the rush hours can be reduced. As there are a lot of history information can be adopted, one of the problems for the electricity price forecasting is how to select the useful features in order to increase the accuracy of the forecasting and also reduce the time complexity. This paper we apply the decision tree c4.5 to select the relevant features for electricity price forecasting. We show the performance of C4.5 is better than the ID3 in terms of accuracy experientially.