DETECTION OF CYBER ATTACK IN NETWORK USING MACHINE LEARNING

Contrasted with the past, improvements in PC and correspondence innovations have given broad and propelled changes. The use of new innovations give incredible advantages to people, organizations, and governments, be that as it may, messes some up against them. For instance, the protection of significant data, security of put away information stages, accessibility of information and so forth. Contingent upon these issues, digital fear based oppression is one of the most significant issues in this day and age. Digital fear, which made a great deal of issues people and establishments, has arrived at a level that could undermine open and nation security by different gatherings, for example, criminal association, proficient people and digital activists. Along these lines, Intrusion Detection Systems (IDS) has been created to maintain a strategic distance from digital assaults. Right now, learning the bolster support vector machine (SVM) calculations were utilized to recognize port sweep endeavors dependent on the new CICIDS2017 dataset with 97.80%, 69.79% precision rates were accomplished individually.

相较于传统时代，个人计算机（Personal Computer, PC）与通信技术的革新带来了广泛且深刻的变革。新兴技术的应用在为个人、企业与政府带来诸多利好的同时，也衍生出诸多棘手的安全问题，例如敏感数据保护、存储数据平台的安全性、数据可用性等。在此类问题的影响下，网络恐怖主义已成为当今时代最为严峻的挑战之一。此类由犯罪组织、专业从业者以及数字活动者等多方群体发起的网络威胁，已达到足以危害公共安全与国家安全的程度。为此，人们研发了入侵检测系统（Intrusion Detection Systems, IDS）以抵御各类网络攻击。本研究基于全新的CICIDS2017数据集，采用增强型支持向量机（Support Vector Machine, SVM）算法识别端口扫描行为，最终分别实现了97.80%与69.79%的识别精度。