Speed Up Similarity Search of Time Series Under Dynamic Time Warping
Similarity search is a foundational task in time series data mining. Although there are many ways to measure the similarity of time series, a lot of evidence indicates that dynamic time warping (DTW) has the best robustness in many applications. Unfortunately, the expensive computational cost limits...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2019-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/8884160/ |
| id |
doaj-e54865e37aeb4f1b939ad84358e82636 |
|---|---|
| record_format |
Article |
| spelling |
doaj-e54865e37aeb4f1b939ad84358e826362021-03-30T00:53:19ZengIEEEIEEE Access2169-35362019-01-01716364416365310.1109/ACCESS.2019.29498388884160Speed Up Similarity Search of Time Series Under Dynamic Time WarpingZhengxin Li0https://orcid.org/0000-0002-9535-261XJiansheng Guo1Hailin Li2Tao Wu3Sheng Mao4https://orcid.org/0000-0002-5122-6224Feiping Nie5College of Equipment Management and UAV Engineering, Air Force Engineering University, Xi’an, ChinaCollege of Equipment Management and UAV Engineering, Air Force Engineering University, Xi’an, ChinaCollege of Business Administration, Huaqiao University, Quanzhou, ChinaCollege of Equipment Management and UAV Engineering, Air Force Engineering University, Xi’an, ChinaCollege of Equipment Management and UAV Engineering, Air Force Engineering University, Xi’an, ChinaSchool of Computer Science, Northwestern Polytechnical University, Xi’an, ChinaSimilarity search is a foundational task in time series data mining. Although there are many ways to measure the similarity of time series, a lot of evidence indicates that dynamic time warping (DTW) has the best robustness in many applications. Unfortunately, the expensive computational cost limits its application in large-scale databases. To speed up similarity search under DTW, we design a framework of two-stage similarity search for time series. In the first stage, we propose an improved lower bounding distance, which can be used to discard plenty of dissimilar series to get a set of candidate sequences. In the second stage, to efficiently get the retrieval result from the set of candidate sequences, we explore early abandoning strategy to avoid the full calculation of DTW. Extensive experiments are conducted on real-world data sets. The experimental results indicate that the proposed method can improve the retrieval efficiency of similarity search under DTW and guarantee no false dismissals.https://ieeexplore.ieee.org/document/8884160/Time seriessimilarity searchdynamic time warpinglower bounding distanceearly abandoning strategy |
| collection |
DOAJ |
| language |
English |
| format |
Article |
| sources |
DOAJ |
| author |
Zhengxin Li Jiansheng Guo Hailin Li Tao Wu Sheng Mao Feiping Nie |
| spellingShingle |
Zhengxin Li Jiansheng Guo Hailin Li Tao Wu Sheng Mao Feiping Nie Speed Up Similarity Search of Time Series Under Dynamic Time Warping IEEE Access Time series similarity search dynamic time warping lower bounding distance early abandoning strategy |
| author_facet |
Zhengxin Li Jiansheng Guo Hailin Li Tao Wu Sheng Mao Feiping Nie |
| author_sort |
Zhengxin Li |
| title |
Speed Up Similarity Search of Time Series Under Dynamic Time Warping |
| title_short |
Speed Up Similarity Search of Time Series Under Dynamic Time Warping |
| title_full |
Speed Up Similarity Search of Time Series Under Dynamic Time Warping |
| title_fullStr |
Speed Up Similarity Search of Time Series Under Dynamic Time Warping |
| title_full_unstemmed |
Speed Up Similarity Search of Time Series Under Dynamic Time Warping |
| title_sort |
speed up similarity search of time series under dynamic time warping |
| publisher |
IEEE |
| series |
IEEE Access |
| issn |
2169-3536 |
| publishDate |
2019-01-01 |
| description |
Similarity search is a foundational task in time series data mining. Although there are many ways to measure the similarity of time series, a lot of evidence indicates that dynamic time warping (DTW) has the best robustness in many applications. Unfortunately, the expensive computational cost limits its application in large-scale databases. To speed up similarity search under DTW, we design a framework of two-stage similarity search for time series. In the first stage, we propose an improved lower bounding distance, which can be used to discard plenty of dissimilar series to get a set of candidate sequences. In the second stage, to efficiently get the retrieval result from the set of candidate sequences, we explore early abandoning strategy to avoid the full calculation of DTW. Extensive experiments are conducted on real-world data sets. The experimental results indicate that the proposed method can improve the retrieval efficiency of similarity search under DTW and guarantee no false dismissals. |
| topic |
Time series similarity search dynamic time warping lower bounding distance early abandoning strategy |
| url |
https://ieeexplore.ieee.org/document/8884160/ |
| work_keys_str_mv |
AT zhengxinli speedupsimilaritysearchoftimeseriesunderdynamictimewarping AT jianshengguo speedupsimilaritysearchoftimeseriesunderdynamictimewarping AT hailinli speedupsimilaritysearchoftimeseriesunderdynamictimewarping AT taowu speedupsimilaritysearchoftimeseriesunderdynamictimewarping AT shengmao speedupsimilaritysearchoftimeseriesunderdynamictimewarping AT feipingnie speedupsimilaritysearchoftimeseriesunderdynamictimewarping |
| _version_ |
1724187754758144000 |