Big Data Analytic using Machine Learning Algorithms for Intrusion Detection System: A Survey

Through the increasing amounts of data day by day, big data analytics has appeared as an important trend for many organizations. These collected data can have important information that possibly will be key to solving extensive problems, such as cybersecurity, healthcare, marketing, intrusion detection, and fraud. An intrusion detection system (IDS) work as observers and evaluates the data in order to detect at all intrusion that may occur in the network or system. The huge data that is a high speed, high volume, and variety of data produced through the network has prepared for the process of the data analysis to detect any attacks through traditional techniques that cause big challenges. Big data analytic uses in IDS to increase accuracy and more efficient for detecting the attacks. we present a survey of eight machine learning algorithms use for IDS, intrusion detection techniques, advantages and disadvantages of intrusion detection techniques, all types of an intrusion detection system, advantages and disadvantages for each type of IDS, and five datasets that use on the intrusion detection system. Moreover, the authors attempt to present a strong picture of algorithms and datasets that use for IDS in all aspects through their extensive survey.