期刊名称：International Journal of Innovative Research in Science, Engineering and Technology
印刷版ISSN：2347-6710
电子版ISSN：2319-8753
出版年度：2017
卷号：6
期号：12
页码：22478
DOI：10.15680/IJIRSET.2017.0612047
出版社：S&S Publications
摘要：Nowadays, email has the efficient become communication medium for exchange of informationbetween people in organizations and the society throughout the world. With the increased email spam attacks theefficiency of communication and security is at stake. Spam emails are adaptive in nature as the spammers constantlyupdate the spam contents that could bypass the spam filters. Negative Selection Algorithm is a one class classifieralgorithm that is employed for email classification and here the classification is between Spam and Non-Spam which isone class classification. This research introduces an email spam detection system that shows an improvement in theperformance of Negative Selection Algorithm. The performance can be improved by hybridizing this algorithm withadditional two concepts which addresses the adaptive nature of unsolicited email spam. Hybridization is done byemploying Firefly Algorithm in the Random detector generation step where Local Outlier Factor is used as the fitnessfunction. The hybridization of Firefly Algorithm helps in selecting subset of features that will be used for themonitoring phase. The result shows higher accuracy than any other optimization algorithms for email spam detectionsystems.
