Instance-Level Subsequence Matching Method based on a Virtual Window
가상 윈도우 기반 인스턴스 레벨 서브시퀀스 매칭 방안

Sun-Young Ihm, Young-Ho Park
2014 KIPS Transactions on Computer and Communication Systems  
A time-series data is the collection of real numbers over the time intervals. One of the main tasks in time-series data is efficiently to find subsequences similar to a given query sequence. In this paper, we propose an efficient subsequence matching method, which is called Instance-Match (I-Match). I-Match constructs a virtual window in order to reduce false alarms. Through the experiment with real data set and query sets, we show that I-Match improves query processing time by up to 2.95 times
more » ... by up to 2.95 times and significantly reduces the number of candidates comparing to Dual Match.
doi:10.3745/ktccs.2014.3.2.43 fatcat:ky7huxgyknbqlfunafgyg35u4u