An Ensemble Learning Approach for SMS Spam Detection

One of the most accessible ways to communicate via text is through a short message service. In recent years, profit-seeking people have taken advantage of the good features of this service to send large numbers of spam messages to random people for malicious purposes. In this respect, detecting spam messages is an important task. The unbalanced proportion of the spam and ham data and the extraction of efficient features from short messages have been the main challenges in the SMS spam detection problem. So far, various methods have been proposed to filter spam messages, whose accuracy still needs to be improved. In this study, we propose an ensemble learning method based on random forest and logistic regression algorithms to increase the accuracy of SMS spam detection. The proposed approach has been tested on two real datasets. The experimental evaluation based on accuracy and AUC shows the effectiveness of the proposed ensemble learning algorithm.