Autonomous Network Security Using Unsupervised Detection of Network Attacks

As a use of network increases for critical transaction, so huge damages are caused by intrusion attacks hence there is the need of Computer network security. To protect network against various active and passive attack, various technique have been suggested .Traditional methods depends on specialized signatures of previously seen attacks, or on expensive and difficult to produce labeled traffic datasets for profiling and training. In this paper, network attack are detected in unsupervised way without following the traditional way of signatures or labeled traffic.