A Research of Spam Message Detection based on Ensemble Learning
Tianfu Li, Jiaxing Shao, Haoli Wang · 2023
With the development of the network technology and the increase of active online users, the spam messages transmitted through the Internet are also growing rapidly. Not only do they potentially cause loss to the recipient, but they also waste network resources. Therefore, it is very necessary to develop an efficient automatic classification model. However, traditional machine learning methods have limited capabilities, while the popular neural networks often encounter insufficient data. Therefore, we proposed a three-layer stacking ensemble learning method to deal with medium-magnitude data and achieved awesome results in our spam detection task. More exactly, the accuracy is 97.8% and the F1 score is 0.913. Besides, we believe that the three-layer ensemble learning method we used to deal with this magnitude of data is also one of the novelty of our paper.