A Deep Learning Approach for Intrusion Detection Using Recurrent Neural Networks

Project Discription

As the network uses are increases day by day which also increase the vulnerability of the system Network Security is always important issue for the network managers and it is constantly increasing because of the growing application areas and requirements of flexible, fast and efficient networking however making the network more flexible, fast and efficient involves loop holes in security aspects of the communication protocol which are used by the attackers for performing various kind of attack on network. The significance of the safety efforts becomes greater as the assurance of information for instance, information of organization and the information of research and the safeguarding against the computer related noxious code or intrusion attack like worm, infection, trojan stallion, and so on, are presently happening as often as possible for the nearly everybody working with digital systems. Today it is improbable to find out an institution, a company, or a private home computer that is not protected against the steadily emerging risks from the networks for example, Internet. Events of intrusion can cost to the companies a large volume of time and money. It can also detect that the existing data could be got end up in the illegal hands, and it leads to the waste of effort and time. There are several proposals to how to enhance security.

The details of proposed work & main properties of our framework are given in Chapter This chapter describes problem identifications based on literature survey and also covers dissertation objective and proposed solution for IDS.

Project Sample Image