Optimal Construction of Multi-Dimensional Indexes in Time-Series Databases: A Physical Database Design Approach.

摘要: Similarity search in time-series databases is an operation that finds such data sequences whose changing patterns are similar to of a query sequence. Typically, it hires the multi-dimensional index for its efficient processing. In order alleviate dimensionality curse, problem high-dimensional cases, previous methods similarity apply Discrete Fourier Transform(DFT) sequences, and take only first two or three DFT coefficients selecting organizing attributes index. Other than this ad-hoc approach, there have been no research efforts on devising systematic guideline choosing best among all coefficients. This paper points out problems occurred methods, proposes novel solution construct optimal The proposed method analyzes characteristics target database, then identifies having discrimination power. Finally, determines number by using cost model search. We show effectiveness through series experiments.