Adaptive distance measurement for time series databases

Efficient retrieval of time series data has gained recent attention from the research community. In particular, finding meaningful distance measurements for various applications is one of the most important issues in the field, since no single distance measurement works for all applications. In this paper, we propose a different distance measurement for time series applications based on Constraint Continuous Editing Distance (CCED) that adjusts the potential energy of each sequence for optimal similarity. Furthermore, we also propose a lower bounding distance for CCED for efficient indexing and fast retrieval, even though CCED does not satisfy triangle inequality.