Deep Learning on Spam Detection

Abstract:
With the development of the network, people's online social activities become more and more frequent, and the number of spams becomes more and more. Spam on social software such as twitter and Facebook will spread some false information and false news to people. It will even spread fraud information to defraud users' property and even create panic in the city. Therefore, social spam detection has attracted extensive attention in recent years and gradually become a research hotspot. With the continuous development of deep learning in recent years, some researchers began to apply deep neural network to tasks. Although great progress has been made in the research work around social spam detection, but there is very little review on this task, and there is a lack of a comprehensive combing of the development of social spam detection in recent years. Therefore, we give an overview of social spam detection, introduce the methods used by social spam detection to improve task performance in recent years, and analyze the potential problems in the task.