4 Hits in 0.71 sec

AWarp: Fast Warping Distance for Sparse Time Series

Abdullah Mueen, Nikan Chavoshi, Noor Abu-El-Rub, Hossein Hamooni, Amanda Minnich
2016 2016 IEEE 16th International Conference on Data Mining (ICDM)  
We derive a new time warping similarity measure (AWarp) for sparse time series that works on the run-length encoded representation of sparse time series.  ...  AWarp is exact for binary-valued time series and a close approximation of the original DTW distance for any-valued series.  ...  This dataset contains the review time series for 3,755 mobile apps. To form review time series, we collect the number of reviews an app receives in a day since the beginning of data availability.  ... 
doi:10.1109/icdm.2016.0046 dblp:conf/icdm/MueenCAHM16 fatcat:smpqztvq2ve5jh5u7g4z4uwtv4

Fast Exact Dynamic Time Warping on Run-Length Encoded Time Series [article]

Vincent Froese and Brijnesh Jain and Maciej Rymar and Mathias Weller
2020 arXiv   pre-print
Dynamic Time Warping (DTW) is a well-known similarity measure for time series.  ...  The worst-case running time is cubic in the encoding length. In experiments we show that our algorithm is indeed fast for time series with short encoding lengths.  ...  Conclusion We developed an asymptotically fast algorithm to compute exact DTW distances between runlength encoded time series. The running time is cubic in the maximum coding lengths of the inputs.  ... 
arXiv:1903.03003v5 fatcat:eqxf7nsewzf3rd4szooahmpzzu

Extracting Optimal Performance from Dynamic Time Warping

Abdullah Mueen, Eamonn Keogh
2016 Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining - KDD '16  
The recursive version of cDTW 10 takes about 0.19 seconds for length 128 (see slide notes) A Mueen, N Chavoshi, N Abu-El-Rub, H Hamooni, A Minnich, Fast Warping Distance for Sparse Time Series, Technical  ...  Pazzani: Iterative Deepening Dynamic Time Warping for Time Series.  ...  What can be made fast? • One-to-One comparison To solve the optimization problem for DTW distance, we need to perform simultaneous alignment of many time series.  ... 
doi:10.1145/2939672.2945383 dblp:conf/kdd/MueenK16 fatcat:o6ak5jxzdrcdvivvsyszr7etee

An approach for pipelining nested collections in scientific workflows

Timothy M. McPhillips, Shawn Bowers
2005 SIGMOD record  
Moreover, he thanks Gianluigi Greco for his recent contribution to the weighted extension, and Alfredo Mazzitelli for his valuable work in designing and implementing the tools for experiments.  ...  When you see these people, please thank them personally for their role in achieving quick reviews of submitted papers.  ...  H-struct is used for fast mining on time-based dataset.  ... 
doi:10.1145/1084805.1084809 fatcat:sgtpcat7vzc3veb4dx2jgskpte