Effective pattern similarity match for multidimensional sequence data sets

Abstract

In this paper we present an effective pattern similarity match algorithm for multidimensional sequence data sets such as video streams and various analog or digital signals. To approximate a sequence of data points we introduce a trend vector that captures the moving trend of the sequence. Using the trend vector, our method is designed to filter out irrelevant sequences from a database and to find similar sequences with respect to a query. Experimental results show that it provides a lower reconstruction error and a higher precision rate compared to existing methods. © Springer-Verlag Berlin Heidelberg 2007.