Embedded Files
We did several of scalability tests to show that our algorithm is robust to parameters' variations.
Below is an example from The electrical power demand provided here. The time series is of length 20,000 data points:
The pruning rate remained almost the same even when we changed the subsequence length. (w = 16)
The pruning rate remained almost the same even when we changed the warping window length. (subseqlen = 400)
Below is an example from the Original House dataset. The dataset consists of the electrical power demand of a UK house over 18.5 days:
The pruning rate remains almost the same even we changed the time series length. (subseqlen = 400, w= 16)
Page updated
Google Sites
Report abuse