Comparative analysis of bagging, stacking and random subspace algorithms

Data mining is a powerful new technology and is an important area of science and engineering. In this paper show that the comparing results using bagging, stacking and random subspace algorithms on forest fire data set in to WEKA data mining suite. We compare better results of these methods and improve classification accuracy. Performance results show that the classifiers built. These classifiers are more accurate than that produced by the classification methods. Finally, we are explaining the combining technique for increasing accuracy on the data set is presented. Experimental results are based on minimum time and minimum error rates.