Efficient k-NN Search of Time Series in Arbitrary Time Intervals
|Termin||Fr 23. März 2018|
|Kurzfassung||The k nearest neighbors (k-NN) of a time series are the k closest sequences within a
dataset regarding a distance measure. Often, not the entire time series, but only specific time intervals are of interest, e.g., to examine phenomena around special events. While numerous indexing techniques support the k-NN search of time series, none of them is designed for an efficient interval-based search. This work presents the novel index structure Time Series Envelopes Index Tree (TSEIT), that significantly speeds up the k-NN search of time series in arbitrary user-defined time intervals.