The Community for Technology Leaders
Green Image
<p><b>Abstract</b>—We have developed a new indexing strategy that helps overcome the curse of dimensionality for time series data. Our proposed approach, called <it>Skyline Index</it>, adopts new <it>Skyline Bounding Regions (SBR)</it> to approximate and represent a group of time series data according to their collective shape. Skyline bounding regions allow us to define a distance function that tightly lower bounds the distance between a query and a group of time series data. In an extensive performance study, we investigate the impact of different distance functions by various dimensionality reduction and indexing techniques on the performance of similarity search, including index pages accessed, data objects fetched, and overall query processing time. In addition, we show that, for <tmath>k{\hbox{-}}{\rm{nearest}}</tmath> neighbor queries, the proposed Skyline index approach can be coupled with the state of the art dimensionality reduction techniques such as Adaptive Piecewise Constant Approximation (APCA) and improve its performance by up to a factor of 3.</p>
Data approximation, dimensionality reduction, similarity search, skyline bounding region, skyline index, time series data.
Bongki Moon, In?s Fernando Vega L?pez, Quanzhong Li, "Skyline Index for Time Series Data", IEEE Transactions on Knowledge & Data Engineering, vol. 16, no. , pp. 669-684, June 2004, doi:10.1109/TKDE.2004.14
104 ms
(Ver 3.3 (11022016))