Temporal Data Mining

Temporal Data Similarity Computation    ◾    31

at least one order of magnitude faster than the best existing method and

sometimes up to 222 times faster.

2.2.3 The Longest Common Subsequence

In contrast to the Euclidean distance, the longest common subsequence

(LCSS) is a measure that is tolerant to gaps in the two time series. A dis-

cussion of LCSS can be found in [Das97]. e LCSS assumes the same

baseline and scale for the two time series. However, it is superior to DTW

in the following areas [Vla05], [Vla03]:

LCSS• handles noisy data and outliers better.

e • DTW can distort the actual distance between points in the time

series by overtt ...