An enhanced bat algorithm for intrusion detection of DOS and DDOS attack in cloud

Abstract

Cloud computing is a technology that is emerging which allows customers to get computing services along with resources like applications, storage, service and network. The cloud computing technology is going through a few security issues. The Intrusion Detection System (IDS) is defined as a defense mechanism that detects the network and its antagonistic activities. An IDS may be host-based or a network based on the cloud environment. In the Denial of Service (DoS) attack there is an overload of intruders that target the system which has the service requests but cannot newlinerespond to other requests and so the resources are available to the users. Distributed Denial of Service (DDoS) attack is defined based on the hosts attacking to degrade the services for completing its removal from the internet. The attacks deprive all the legitimate users to make use of the target services by means of depleting the resources of the victims. Recently, the activities of Denial of Service (DoS) attacks newlineagainst cloud applications and online services to extort, disable or impair the newlinecompetition have increased.It becomes challenging to identify attacks on the infrastructures of the cloud owing to its complexity and the distributed nature. Besides, providing access from different smart devices to the cloud infrastructures it also increases the challenge and the complexity of the attacks. In data mining, decision tree is a classification approach. To create a scheme from a reclassified data set, this approach should be used. In this work, the Classification and Regression Tree (CART), C4.5 and Random Tree (RT) classifiers are proposed. The CART algorithm that is used has a tendency to grow more trees on nodes, until no tree can be grown deciding upon whether a node is terminal or not. newline newline