Consider the problem of finding the highly correlated pairs of time series over a time window
and then sliding that window to find the highly correlated pairs over successive co-temporous windows such that each successive window starts only a little time after the previous window.
Doing this efficiently and in parallel could help in applications such as
sensor fusion, financial trading, or communications network
monitoring, to name a few.
We have developed a parallel incremental random vector/sketching approach to this problem and compared it with
the state-of-the-art nearest neighbor method iSAX . Whereas iSAX
achieves 100% recall and precision for Euclidean distance, the sketching approach is, empirically, at least 10 times faster and achieves 95% recall and 100% precision on real and simulated data.
For many applications this
is worth the minor
reduction in recall.
Our method scales up to 100 million time series and is linearly scalable in its expensive steps (but quadratic in the less expensive ones).
We carried out our experiments on synthetic, seismic and financial datasets.
|This research has been partially funded by the European Commission under the H2020 programme project #732051