The effective algorithm for constant state detection in time series.

This is a web playground implementation for the article The effective algorithm for constant state detection in time series. The two new versions of the algorithm are presented on this website, the first one utilizes the advanced section addition, the second one uses the output postprocessing also. Article describing the simplest version of the algorithm can be found here (for example) https://figshare.com/authors/Andrei_Keino/5901722

Learn more »

-->

Important:

Note that the version of the algorithm with postprocessing while processing the data with a small Max. section height and a small Min. section length values will find just one constant section. This behavior is defined by the algorithm's design as this version of the algorithm chooses "the best" constant state section from a set of the intersected ones, i. e. algorithm chooses one section with the maximal length and the minimal height from the set of the intersected section. All the presented versions of the algorithm have the computational complexity O(N log(N)) as the original one, where N is the length of the time series. The online versions of the presented algorithm could be developed also.



Index of the sections found by the algorithm:


Index of the regions of the intersected section found by the algorithm:



Web hosting by Somee.com