Title :
Study on encoding and detection for patterns
Author :
Xiong, Jia-Tun ; Zhang, Li ; Yue, Xiu-Qing
Author_Institution :
Dept. of Comput., Radar Acad., Wuhan, China
Abstract :
How to use the patterns mined by KDD algorithms is very important to real-time and effectiveness of IDS. This paper put forward an algorithm of pattern encoding, which can express patterns in digits, thus the intrusion pattern can be detected efficiently. At the same time, it also has proved that the patterns detection methods based on pattern encoding are equal to the methods with similarity theory, as a result, any intrusion pattern detection algorithms based on similarity theory can be realized by pattern encoding.
Keywords :
data mining; encoding; pattern recognition; security of data; data mining; intrusion pattern detection; pattern encoding; similarity theory; Association rules; Data mining; Decision making; Detection algorithms; Encoding; Intrusion detection; Large-scale systems; Pattern matching; Radar detection; Sorting;
Conference_Titel :
Machine Learning and Cybernetics, 2004. Proceedings of 2004 International Conference on
Print_ISBN :
0-7803-8403-2
DOI :
10.1109/ICMLC.2004.1382392