HIGHLIGHTS
- who: Responsible editor Johannes Fu00fcrnkranz. and colleagues from the Humboldt-Universitu00e4t zu Berlin, Berlin, Germany have published the research work: ClaSP: parameter-free time series segmentation, in the Journal: (JOURNAL)
- what: In the experimental evaluation using a benchmark of 107 sets the authors show that ClaSP outperforms the state of the art in terms of accuracy and is fast and scalable. The authors show that ClaSP, when using a k-NN classifier, achieves both highly accurate results and also is very fast, because most of the work can be factored out into a preprocessing phase . . .
If you want to have access to all the content you need to log in!
Thanks :)
If you don't have an account, you can create one here.