Mail Detection Model of Campus Network Based on the Adaptive Learning Algorithm
Bin Chen, Yi-zhou Dong, Mingrong Mao · International Journal of Future Computer and Communication · 2017
Concerning the problem that a large number of spam disturbing the user, a method named adaptive learning algorithm was proposed.The passive attack method based on the Simple Mail Transfer Protocol(SMTP) session log initiated by the mail host in the campus during half a year.Analysis on the status of delivery rate and many types of failure message of the host behavior in the session record, and ultimately achieve effective adaptation by detecting spam source host behavior on the recent email classification.The experimental results show that the implementation of a number of rounds of classification strategy adjustment, at last the detection accuracy can reach 94.7%.The design is very useful for network administrators, he can effectively detect spam internal host status, and control the behavior of the spam host in the beginning.