In this paper, we propose a new algorithm for time series discord discovery which is based on the discrete method called Symbolic Aggregate approXimation (SAX) method using distance measure in SAX space and Euclidean distance associated with the idea of early abandoning. |