Abstract: Using the adaptive boosting (AdaBoost) algorithm, the traditional classifiers, i.e. the BP neural network, the support vector machine (SVM) and the linear discriminant analysis (LDA) classifier, are trained into stronger ones. Based on traditional classifiers, the function weights and data processing are first adjusted according to the mapped output of classifiers. This method could effectively decrease the dependence on data features. These trained classifiers are then used to identify a group of data elicited by motor imagery. Based on the AdaBoost algorithm, the trained classifiers can obtain higher classification accuracy than the traditional ones. These results confirm the potential valuation of AdaBoost algorithm in BCI applications.