文章基本信息

标题：Implementation of Pattern Classifiers under Attack Using Security Evaluation
本地全文：下载
作者：Rupali B. Navalkar ; Prof. Rajeshri R. Shelke
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：3
页码：5083
DOI：10.15680/IJIRCCE.2017.0503261
出版社：S&S Publications
摘要：Pattern classification is a branch of machine learning that focuses on recognition of patterns andregularities in data. This Pattern classification system are commonly used in adversarial applications, like biometricauthentication, network intrusion detection, and spam filtering, in which data can be purposely manipulated by humansto undermine their operation. As this adversarial scenario is not taken into account by classical design methods, patternclassification systems may exhibit vulnerabilities, whose exploitation may severely affect their performance, andconsequently limit their practical utility. Extending pattern classification theory and design methods to adversarialsettings is thus a novel and very relevant research direction, which has not yet been pursued in a systematic way. In thispaper, we propose a framework for empirical evaluation of classifier security that formalizes and generalizes the mainideas proposed in the literature, and give examples of its use in real applications. Reported results show that securityevaluation can provide a more complete understanding of the classifier’s behavior in adversarial environments, andlead to better design choices. This framework can be applied to different classifiers on one of the application from thespam filtering, biometric authentication and network intrusion detection. Considering Multimodal system, that theproposed methodology to rank score fusion rules is capable of providing correct ranking of score fusion rules underspoof attack. So in this we propose an algorithm for the generation of training and testing sets to be used for securityevaluation.
关键词：Pattern classification; Adversarial classification; performance evaluation; security evaluation.