期刊名称：International Journal of Innovative Research in Computer and Communication Engineering
印刷版ISSN：2320-9798
电子版ISSN：2320-9801
出版年度：2017
卷号：5
期号：1
页码：594
DOI：10.15680/IJIRCCE.2017.0501119
出版社：S&S Publications
摘要：The mishandle of talk administrations via robotized programs, known as talk bots, represents a genuinedanger to Internet clients. Talk bots target famous talk systems to circulate spam and malware. In this paper, we firstlead a progression of estimations on a substantial business talk organize. Our estimations catch a sum of 16 distinctsorts of visit bots extending from easy to cutting edge. In addition, we watch that human conduct is more perplexingthan bot conduct. In light of the estimation ponder, we propose a classification framework to precisely recognize visitbots from human clients. The proposed characterization framework comprises of two components: 1) an entropy-basedclassifier; and 2) a Bayesian-based classifier. The two classifiers supplement each other in visit bot detection. Theentropy-based classifier is more exact to recognize unknown visit bots, while the Bayesian-based classifier is quicker toidentify known visit bots. Our trial assessment demonstrates that the proposed characterization framework isexceptionally compelling in differentiating bots from people
