...
  • 主题
高级搜索  >
历史搜索 清空历史
首页> 外文期刊>Advanced Research in Electrical and Electronic Engineering: AREEE >Network Anomaly Detection Model by using PCA and K-Means Clustering Algorithm
【24h】

Network Anomaly Detection Model by using PCA and K-Means Clustering Algorithm

机译：网络异常检测模型使用PCA和K-means聚类算法

获取原文
           
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

The world is now interconnected globally through the internet connectivity which enables to communicate distant corners of globe very easily. As the amount of data increasing day by day the need of communication with secure medium is desirable not only for commercial but also for institutional organizations. Here in this paper, we are implementing one of the popular intrusion Detection Technique (IDS) i.e. Network Anomaly Detection technique. Anomaly is the deviation from the present original data. It is the unwanted form of data which must be removed so that system’s behavior should not get disturbed. All public as well as private organizations wish to have a secured storage of their data, and stored information can be access by the authorized person only. IDS help them to develop a system which can detect novel as well as existing attacks present in the data. IDS is the latest technology after Anti-Virus, Firewall, encryption etc which protects system from known and unknown attacks. IDS has Host Based Intrusion Detection System(HIDS) which examines the log files of individual host and protect the data of host, where as Network Based Intrusion Detection System(NIDS) performs task to defend the whole network of host, by searching for default/ risky packets. IDS can detect existing attacks present in the stored data or it can detect the novel attack occurring in real time. Here in this paper both anomalies are detected from stored as well as real time logs. For this purpose we implementing PCA algorithm, followed by K-Means clustering algorithm. Here we are showing how clustering algorithm can improve the efficiency of the system.
机译：世界现在通过互联网连接全球连接，这使得能够很容易地沟通地球的遥远角落。随着日复一日的数据量，不仅可以用于商业化组织的商业，而且需要与安全介质的沟通的需求。本文在本文中，我们正在实现流行的入侵检测技术（IDS）即网络异常检测技术之一。异常是与本原始数据的偏差。它是必须删除的不需要的数据形式，以便系统的行为不应受到干扰。所有公共和私人组织都希望拥有其数据的安全存储，并且存储信息只能由授权人员访问。 ID帮助他们开发一个系统，可以检测数据以及数据中存在的现有攻击。 IDS是防病毒，防火墙，加密等后的最新技术，可保护系统从已知和未知攻击。 IDS具有基于主机的入侵检测系统（HID），其检查单个主机的日志文件并保护主机的数据，在基于网络的入侵检测系统（NID）执行任务以防御整个主机网络，通过搜索默认/风险数据包。 ID可以检测存储数据中存在的现有攻击，或者它可以检测实时发生的新攻击。本文在本文中检测到两种异常，以及实时日志。为此目的，我们实现了PCA算法，然后实现了K-means聚类算法。在这里，我们展示了聚类算法如何提高系统的效率。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号