Tackle three practical classification problems via Ensemble Learning

News Categorization, Intrusion Detection and Spam Detection are three practical problems¹ in Data Mining and Cybersecurity. Their focus is on string sequences analysis towards application of knowledge discovery techniques for protecting personal computer information by means of detection, prevention, and response to various attacks. These three string sequences analysis problems could be treated as three classification problems. To tackle these three classifications problems, we propose a Ensemble Learning method. The idea of ensemble learning is to employ multiple learners and combine their predictions. These ensemble methods utilize multiple models to obtain better predictive performance than could be obtained from any of the constituent models[13], [14], [16]. In the tasks, we utilize (LDA-, SK-)SVM, (LDA-, SK-)GP and (LDA-, SK-) AdaBoost as the weak classifiers, and the experiments shows that ensemble learning method can improve the classification performance significantly.