A Metric Learning-Based Univariate Time Series Classification Method
High-dimensional time series classification is a serious problem. A similarity measure based on distance is one of the methods for time series classification. This paper proposes a metric learning-based univariate time series classification method (ML-UTSC), which uses a Mahalanobis matrix on metric...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-05-01
|
| Series: | Information |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2078-2489/11/6/288 |
| _version_ | 1797566763688263680 |
|---|---|
| author | Kuiyong Song Nianbin Wang Hongbin Wang |
| author_facet | Kuiyong Song Nianbin Wang Hongbin Wang |
| author_sort | Kuiyong Song |
| collection | DOAJ |
| description | High-dimensional time series classification is a serious problem. A similarity measure based on distance is one of the methods for time series classification. This paper proposes a metric learning-based univariate time series classification method (ML-UTSC), which uses a Mahalanobis matrix on metric learning to calculate the local distance between multivariate time series and combines Dynamic Time Warping(DTW) and the nearest neighbor classification to achieve the final classification. In this method, the features of the univariate time series are presented as multivariate time series data with a mean value, variance, and slope. Next, a three-dimensional Mahalanobis matrix is obtained based on metric learning in the data. The time series is divided into segments of equal intervals to enable the Mahalanobis matrix to more accurately describe the features of the time series data. Compared with the most effective measurement method, the related experimental results show that our proposed algorithm has a lower classification error rate in most of the test datasets. |
| first_indexed | 2024-03-10T19:31:49Z |
| format | Article |
| id | doaj.art-dadf540e13ff442abb4e21348547d1cf |
| institution | Directory Open Access Journal |
| issn | 2078-2489 |
| language | English |
| last_indexed | 2024-03-10T19:31:49Z |
| publishDate | 2020-05-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Information |
| spelling | doaj.art-dadf540e13ff442abb4e21348547d1cf2023-11-20T02:05:14ZengMDPI AGInformation2078-24892020-05-0111628810.3390/info11060288A Metric Learning-Based Univariate Time Series Classification MethodKuiyong Song0Nianbin Wang1Hongbin Wang2College of Computer Science and Technology, Harbin Engineering University, Harbin 150000, ChinaCollege of Computer Science and Technology, Harbin Engineering University, Harbin 150000, ChinaCollege of Computer Science and Technology, Harbin Engineering University, Harbin 150000, ChinaHigh-dimensional time series classification is a serious problem. A similarity measure based on distance is one of the methods for time series classification. This paper proposes a metric learning-based univariate time series classification method (ML-UTSC), which uses a Mahalanobis matrix on metric learning to calculate the local distance between multivariate time series and combines Dynamic Time Warping(DTW) and the nearest neighbor classification to achieve the final classification. In this method, the features of the univariate time series are presented as multivariate time series data with a mean value, variance, and slope. Next, a three-dimensional Mahalanobis matrix is obtained based on metric learning in the data. The time series is divided into segments of equal intervals to enable the Mahalanobis matrix to more accurately describe the features of the time series data. Compared with the most effective measurement method, the related experimental results show that our proposed algorithm has a lower classification error rate in most of the test datasets.https://www.mdpi.com/2078-2489/11/6/288Mahalanobismetric learningmultivariabletime seriesunivariate |
| spellingShingle | Kuiyong Song Nianbin Wang Hongbin Wang A Metric Learning-Based Univariate Time Series Classification Method Information Mahalanobis metric learning multivariable time series univariate |
| title | A Metric Learning-Based Univariate Time Series Classification Method |
| title_full | A Metric Learning-Based Univariate Time Series Classification Method |
| title_fullStr | A Metric Learning-Based Univariate Time Series Classification Method |
| title_full_unstemmed | A Metric Learning-Based Univariate Time Series Classification Method |
| title_short | A Metric Learning-Based Univariate Time Series Classification Method |
| title_sort | metric learning based univariate time series classification method |
| topic | Mahalanobis metric learning multivariable time series univariate |
| url | https://www.mdpi.com/2078-2489/11/6/288 |
| work_keys_str_mv | AT kuiyongsong ametriclearningbasedunivariatetimeseriesclassificationmethod AT nianbinwang ametriclearningbasedunivariatetimeseriesclassificationmethod AT hongbinwang ametriclearningbasedunivariatetimeseriesclassificationmethod AT kuiyongsong metriclearningbasedunivariatetimeseriesclassificationmethod AT nianbinwang metriclearningbasedunivariatetimeseriesclassificationmethod AT hongbinwang metriclearningbasedunivariatetimeseriesclassificationmethod |