Database Support for Discovering Patterns in Large Datasets Collected from Multiple Sensors

A large number of applications require mining high rate data streams. An important challenge in dealing with such data intensive applications is fast data change rate that requires improved analysis, new algorithms, tools and techniques. Our purpose for this work is to examine how to integrate collecting data streams from multiple sensors in order to discover new patterns and minimize the number of incorrect alarms. We present a framework for mining patterns from data streams and discuss practical applications.