To search, Click below search items. |
| All Published Papers Search Service |
|
|
|
Title |
Multiple Simultaneous Threat Detection in UNIX Environment |
|
Author |
Zafar Sultan |
| Citation |
Vol. 9 No. 2 pp. 65-75 |
|
Abstract |
Although UNIX is considered a very stable and secure platform, the development of Intrusion Detection Systems (IDS) is essential as current and future generations of hackers are continuously attempting to undermine its integrity. The empirical experiment of multiple simultaneous threat detection system proved that use of hybrid data fusion model of Bayesian, Dempster Shafer and extended Dempster Shafer increased an average 20% threat detection rate. The false positive rate also went down by 51%. The use of Extended Dempster Shafer to combine probability mass of 4 intrusion detection (Multisensor) systems increased precision of threat detection by 36% whilst the initial probability mass of the Dempster Shafer of Multisensor was only 0.03. Set Cover as a middle tier data fusion tool produced incredible results, particularly in data grouping by reducing the population size from 2273 to 429 that amazingly minimise the computational processing cpu and memory overhead cost and time. In order to improve the results of the precision of the multiple simultaneous threat detection system, as a next step of my research is that is an extension to the Bayesian and Dempster Shafer theory. GEP presents a better evidential combination and separate propositions and the decisions. |
|
Keywords |
Multiple Simultaneous Threat Detection, Intrusion Detection Systems, Bayesian Theory, Dempster Shafer, Multisensor Data Fusion, Extended Dempster Shafer, Set Cover, Set Packing, GEP, UNIX |
|
URL |
