Anzeige
Erschienen in:
01.12.2016 | Original Article
An extensive experimental study on segmenting online time series with error bound guarantee
Erschienen in: International Journal of Machine Learning and Cybernetics | Ausgabe 6/2016
EinloggenAktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
Abstract
An improved algorithm, i.e., HFSW, for segmenting online time series with error bound is proposed in our latest paper (IJMLC doi:10.1007/s13042-014-0310-9, 2014). Some researchers engaged in this filed read this paper and point out that there are two another existing strategies named FSW (IEEE TKDE doi:10.1109/TKDE.2008.29, 2008) and DisAlg (VLDBJ doi:10.1007/s00778-014-0355-0, 2014) which can also deal with the segmentation of online time series. And then, they want us to conduct some further experiments to demonstrate the effectiveness of our proposed method through comparing HFSW with FSW and DisAlg. Thus, we conduct such experimental comparison by testing 43 real datasets with the same fixed setting and further give the analysis to main difference among these algorithms.