Spam Classification Based on Machine Learning Algorithm
Yichun Huang · Highlights in Science Engineering and Technology · 2023
As technology continues to advance, email is used in almost every field. However, the dramatic increase in the number of spam emails has led to a growing need for accurate and powerful spam classifiers. Unfortunately, spam in its many forms is constantly being updated as the Internet evolves, and the challenge of fighting spam is enormous. With the emerging of deep learning methods, deep based algorithms are widely applied in most of the real-world scene and tasks. In this paper, we used CNN, RNN, LSTM and naïve Bayes models to implement spam filtering, and compared them based on information such as accuracy, model strengths and weaknesses. We test all the models on the public dataset and measure them by four metrics, including accuracy, precision, recall and F1-score. Finally, we conclude that the naïve bayes model achieves the best performance among all those methods, which can deal with the threat of spam efficiently.