In ensemble learning we combine the output of multiple classifiers in order to on better prediction on classification accuracy. The classifcation of resultant classifer is better then the resultant classifier. We need a way to combine the output of multiple classifiers. First we generate a group of base learners. These learners will be using different algorithms.(say learner which uses a Dicision Tree,Neural Network, Support Vector Machine etc..), they could be same algorithms with different hyperparameters, could have different representations or may use different training sets. Muliple learner will give different classifications.