Algorithms for Segmenting Time Series

Abstract

As with most computer science problems, representation of the data is the key to e cient and e ective solutions. Piecewise linear representation has been used for the representation of the data. This representation has been used by various researchers to support clustering, classi cation, indexing and association rule mining of time series data. A variety of algorithms have been proposed to obtain this representation, with several algorithms having been independently rediscovered several times. In this paper, we examine the techniques and then introduce the best-known algorithm.