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：11
页码：16461
DOI：10.15680/IJIRCCE.2017.0511016
出版社：S&S Publications
摘要：In recent era the network dramatically extended, security considered as major issue in networks. Internetattacks are increasing, and there have been various attack methods, consequently. Intrusion Detection System (IDS) isan effective security tool that helps to prevent unauthorized access to network resources by analyzing the networktraffic and classifying the record as either normal or anomalous in this paper proposed method but has significantchallenges in building IDS that are 1) Streaming nature of data and computer networks, 2) Feature selection orreduction because feature may be irrelevant or redundant and may inhabit system performance [1].The proposed method includes online and offline classification on data set. For this Naive Bayes Classifier is used [2],after that active learning enables to solve the problem using subset of labeled data points. Here, we introduced theNetwork Anomaly Detection Using Active Learning (NADAL) online method that allows us three advantages, 1)Overcoming streaming data challenges, 2) Reduce the high cost with instance labeling, 3) improved speed detect 4)Accuracy of detection
关键词：Intrusion Detection; Anomaly Detection; Active Learning; Accurate identification
