...
  • 主题
高级搜索  >
历史搜索 清空历史
首页> 外文期刊>Open Access Library Journal >Implementation of Machine Learning Method for the Detection and Prevention of Attack in Supervised Network
【24h】

Implementation of Machine Learning Method for the Detection and Prevention of Attack in Supervised Network

机译：监督网络攻击检测和预防机器学习方法的实现

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

摘要

The sustainability of a company depends on the permanent availability of its information system. This reality influences the behavior of companies, which are becoming increasingly mature in their investments in information system security, which is an absolutely vital element. The use of a service called “SYSLOG” to centralize the network event logs that are sent by printers, servers, routers, firewalls, IDS and IPS in an SYSLOG server is a perfect example for network optimization. In this work, which consists in setting up a Machine learning algorithm for detection and prevention of attacks, we are interested on one hand with the problems encountered on the SYSLOG service and on the other hand with the problems encountered during the detection and prevention of anomalies in the SYSLOG service. In order to ensure an optimal level of security within the network according to the criteria specified, we will first proceed to an analysis of the log files present in the server; followed by an attack detection based on an automatic machine learning algorithm using the signature and historical behavior of the different attacks. As result, we have the possibility to generate real-time alerts on malfunctions; real-time monitoring of the use of an application (number of users, functions used, etc.); the identification of the origin of incidents occurring in applications.
机译：公司的可持续性取决于其信息系统的永久性可用性。这一现实影响了公司的行为，这在他们对信息系统安全的投资中变得越来越成熟，这是一个绝对重要的元素。使用称为“syslog”的服务将由打印机，服务器，路由器，防火墙，ID和IP中发送的网络事件日志集中在Syslog Server中的一个完美的网络优化示例。在这项工作中，其中包括设置机器学习算法进行检测和预防攻击，我们与Syslog服务中遇到的问题一方面感兴趣，另一方面，在检测和预防异常期间遇到的问题在Syslog服务中。为了根据指定的标准确保网络内的最佳安全程度，我们将首先进行到服务器中存在的日志文件的分析;然后基于使用不同攻击的签名和历史行为的自动机器学习算法进行攻击检测。结果，我们有可能在故障上产生实时警报;实时监控应用程序的使用（用户数，使用的功能等）;识别应用中发生的事故起源。

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