How does knowledge discovery cooperate with active database techniques in controlling dynam.pdfVIP

How Does Knowledge DiscoveryCooperate with Active Database Techniques inControlling Dynamic Environment? ?Hiroyuki Kawano1, Shojiro Nishio2, Jiawei Han3 and Toshiharu Hasegawa11 Department of Applied Math. and Physics, Kyoto University, Kyoto 606,Japan (fkawano,hasegawag@kuamp.kyoto-u.ac.jp)2 Department of Information Systems Engineering, Osaka University, Osaka 565,Japan (nishio@ise.osaka-u.ac.jp)3 School of Computing Science, Simon Fraser University, Burnaby, BC,Canada V5A 1S6 (han@cs.sfu.ca)Abstract. A dynamic environment, such as a production process, acommunication network, highway trac, etc., may contain a huge amountof information, changing with time, which is a valuable resource for un-derstanding the general behavior of the environment, discovering theregularities and anomalies currently happening in the environment, con-trolling an evolution process, and intelligent modeling or managing theenvironment. Unfortunately, the data generated in a dynamic environ-ment are often expressed in low level primitives and in huge volumes.Because of the dynamic, continuous and rapid changes of the information ow, it is dicult to catch the regularities and anomalies in a dynamicenvironment and react promptly for real-time applications. In this study,a knowledge discovery technique is integrated with data sampling andactive database techniques to discover interesting behaviors of a dynamicenvironment and react intelligently to the environment changes. The dis-covery of the dynamics in a computer communication network and theapplication of the discovered knowledge for network management aretaken as an example in our study. The study shows (1) data samplingis necessary in the collection of information for regularity analysis andanomaly detection; (2) knowledge discovery is important for generaliz-ing low level data to high-level information and detecting interestingpatterns; (3) active database technology is essential for real-time reac-tion to the changes in a dynamic env