Predicting Energy Demands Constructed on Ensemble of Classifiers

Analysis of time series data and perfect future prediction are the most stimulating tasks that the data analysts face in countless fields. Forecasting of energy demands is very essential because both excess energy cost and delay lead to a significant reduction storing costs. In order to discover the uniformities in dynamic, non-stationary data and time series prediction needs the use of models to be integrated with multiple forecast models. The Ensemble learning model discovers the dynamic patterns in energy time series data. The performance of two different Ensemble learning techniques is compared against Bagging and stacking in forecasting energy time series data. Stacking technique used in this paper combines different classifiers like Multilayer Perceptron (MLP), Radial Basis Function (RBF), and Support Vector Machine (SVM).