DocumentCode :
481745
Title :
Scalable and Accurate Application Signature Discovery
Author :
Zhang, Ming-wei ; Liu, Dai-ping
Author_Institution :
Comput. Sch., Wuhan Univ., Wuhan
Volume :
1
fYear :
2008
fDate :
19-20 Dec. 2008
Firstpage :
482
Lastpage :
487
Abstract :
Newly emerged applications are producing a large amount of traffic and connection in the Internets. And they are becoming increasingly difficult to detect. Signature based method are currently the approaches for discovering and detecting the patterns of application. However, these methods may confront their difficulty in validating the efficiency and quality of signatures for unknown applications. Therefore, how to generate the more accurate and representative patterns and validate the quality of signatures is a critical issue.In this paper, a new method has been proposed with a new structure to generate high quality signatures. Different from traditional methods, this one employs a signature learning mechanism that is designed to refine the signatures by merging the similar patterns to improve the signature quality. The experiment indicates that this method is efficient to generate accurate and robust signatures. And the quality of signatures is improved by signature learning.
Keywords :
data mining; learning (artificial intelligence); Internets; signature based method; signature discovery; signature learning mechanism; signature quality; Application software; Computational intelligence; Computer industry; Computer worms; Conferences; Internet; Learning systems; Merging; Payloads; Robustness; clustering; signature generation; string alignment;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Computational Intelligence and Industrial Application, 2008. PACIIA '08. Pacific-Asia Workshop on
Conference_Location :
Wuhan
Print_ISBN :
978-0-7695-3490-9
Type :
conf
DOI :
10.1109/PACIIA.2008.104
Filename :
4756606
Link To Document :
بازگشت