Abstract. Similarity of objects is one of the crucial concepts in several applications, including data mining. For complex objects, similarity
is nontrivial to de ne. In this paper we present an intuitive model for
measuring the similarity between two time series. The model takes into
account outliers, di erent scaling functions, and variable sampling rates.
Using methods from computational geometry, we show that this notion
of similarity can be computed in polynomial time. Using statistical approximation techniques, the algorithms can be speeded up considerably.
We give preliminary experimental results that show the naturalness of
the notion.
Описание
Finding Similar Time Series - Das, Gunopulos, Mannila (ResearchIndex)
%0 Conference Paper
%1 DasEtAl1996
%A Das, Gautam
%A Gunopulos, Dimitrios
%A Mannila, Heikki
%B Principles of Data Mining and Knowledge Discovery
%D 1997
%K B_scanpathsimilarity similaritymeasure timeseries
%P 88-100
%T Finding Similar Time Series
%U citeseer.ist.psu.edu/das96finding.html
%X Abstract. Similarity of objects is one of the crucial concepts in several applications, including data mining. For complex objects, similarity
is nontrivial to de ne. In this paper we present an intuitive model for
measuring the similarity between two time series. The model takes into
account outliers, di erent scaling functions, and variable sampling rates.
Using methods from computational geometry, we show that this notion
of similarity can be computed in polynomial time. Using statistical approximation techniques, the algorithms can be speeded up considerably.
We give preliminary experimental results that show the naturalness of
the notion.
@inproceedings{DasEtAl1996,
abstract = {Abstract. Similarity of objects is one of the crucial concepts in several applications, including data mining. For complex objects, similarity
is nontrivial to de ne. In this paper we present an intuitive model for
measuring the similarity between two time series. The model takes into
account outliers, di erent scaling functions, and variable sampling rates.
Using methods from computational geometry, we show that this notion
of similarity can be computed in polynomial time. Using statistical approximation techniques, the algorithms can be speeded up considerably.
We give preliminary experimental results that show the naturalness of
the notion.
},
added-at = {2007-05-23T21:33:38.000+0200},
author = {Das, Gautam and Gunopulos, Dimitrios and Mannila, Heikki},
biburl = {https://www.bibsonomy.org/bibtex/260ce2ed5413cef6481c2adac4782ecb4/tmalsburg},
booktitle = {Principles of Data Mining and Knowledge Discovery},
description = {Finding Similar Time Series - Das, Gunopulos, Mannila (ResearchIndex)},
interhash = {580f5755dbf991e89d91775265080fce},
intrahash = {60ce2ed5413cef6481c2adac4782ecb4},
keywords = {B_scanpathsimilarity similaritymeasure timeseries},
pages = {88-100},
timestamp = {2007-10-04T14:19:55.000+0200},
title = {Finding Similar Time Series},
url = {citeseer.ist.psu.edu/das96finding.html},
year = 1997
}