Shape-based time series similarity measure and pattern discovery algorithm

摘要: Pattern discovery from time series is of fundamental importance. Most the algorithms pattern in capture values based on some kinds similarity measures. Affected by scale and baseline, value-based methods bring about problem when objective to shape. Thus, a measure shape, Sh measure, originally proposed, properties this corresponding proofs are given. Then shape algorithm put forward. The proposed terminated finite iteration with given computational storage complexity. Finally experiments synthetic datasets sunspot demonstrate that valid.