DETECTION OF DISTRIBUTED DENIAL OF SERVICE ATTACK USING DLMN ALGORITHM IN HADOOP
Y.S. Kalai Vani, P. Ranjana · Journal of Critical Reviews · 2020
Inside the cyber world the main issue is to protect the systems from Intrusions in various cyber attacks. Important issue in today’s world is protecting systems from various network attacks. Cyber attack results in system intrusions which are connected in the network is Denial of Service (DoS). This kind of attack which makes the network resources unavailable for the cyber users connected through network so as a result of unavailable resources there is huge loss of data, resources and money. Significance of detection and prevention in computer network proven the well security. Intrusion Detection Systems (IDSs) is a tool which is utilized to distinguish the various kinds of digital attacks in the system, in such detection to detect the unusual pattern in the network data mining techniques were utilized. Data mining approach takes a major role to detect the intrusions in network and it is used to develop the IDS in an effective manner. Using the classification technique in data mining in which it classifies normal data and affected data. This research paper which follows the strategy in which the improvised Deep Learning Modified Network (DLMN) applies the strategy to detect the intrusions in the form of normal and affected data. To execute the proposed DLMN algorithm NSL-KDD informational index has been used, the data comprises of various characteristic which is utilized to recognize network intrusion. The Non attacked data’s are secured using MCS-ECC algorithm. ECC is the one type of existing cryptography algorithm. In our proposed work Modified Crow Search algorithm is hybrid with existing ECC algorithm. To optimize the encryption modified crow search algorithm is mainly used and ECC algorithm is used decrypt the data. In Modified Crow Search Algorithm, the levy flight function will be replaced for a random search position. Using the Big data Analytics in solving the problem with Hadoop and Map-Reduced Platform which takes the data packets in huge size and detects the normal and attacked packets.