|
|
 |
| |
|
|
|
New Network Access Control Method Using Intelligence Agent Technology |
|
|
|
PP: 43-48 |
|
|
|
Author(s) |
|
|
|
Hu Ruo,
Guo Jian hua,
|
|
|
|
Abstract |
|
|
| Today, new study has been made to the Abnormal Access Discovery (AAD). While label-based AAD have proven effective
in discovering known abnormal access, label-based AAD can be able to automatically detect previously unrecorded risks. Traditional
AAD cannot meet evolving network workflows in real time. To resolve this limitation, intelligence agent techniques can be used to
create a new type of AAD that can be able to dynamically model a flow of network access. In this paper, we present two methods
for abnormal network access. The two methods include diagram-based probe method and the classifying-based method. Overall, the
diagram-based probe method achieved slightly superior results, but required more parameters than the classifying-based method. As a
result of its fewer parameter requirements, the classifying method can be more easily generalized to different types of network traffic
flows. |
|
|
|
|
|
|
|
