Today, with the advent of internet, everyone can do information exchange and resource sharing. Even business organization and government agencies are not behind in this move to reach users for their decision making and for business strategies. But at the same time, with ease of use and availability of various software tools, breaching and penetrating into other's network and confidential credential can be done by any individual with little knowledge expertise and hence the internet attacks are rise and are main concerns for all internet users and business organizations for internal as well as external intruders. Even, existing solutions and commercial Intrusion Detection Systems (IDSs) are developed with limited and specific intrusion attack detection capabilities without any prevention capabilities to secure vital resources of the information infrastructure. So, this paper explores the details about the implementation and experimental analysis of Advanced Intrusion Detection System (AIDS) with its prevention capabilities to provide detection of known as well as unknown intrusions in the computer system and also automatic alerts are given to the network administrator for applying prevention capabilities. Further, this system is intended to generate new intrusion signatures from unknown intrusions and store them back into signature database to speed up detection capabilities of this AIDS in next iterative computation. Data mining approach is used to handle the large amount of data captured in the Internet to improve its execution time and to give fast response to the network administrator for prevention of data resource with minimal user intervention. In experimental analysis, this proposed system gives improved and effective intrusion detection rate up to 91% in comparison with existing research IDSs Snort and PHAD with minimization in false positive rate up to 11%.

Attack Data Mining Intrusion Detection Intrusion Prevention