Local morphological patterns for time series classification

The key problem of time series classification is the similarity measure between time series. In recent years, efficient and accurate similarity measurement methods of time series have attracted extensive attention from researchers. According to the different similarity measure strategies, the existing time series classification methods can be roughly divided into shape-based (original value) methods and structure-based (symbol transformation) methods. Shape-based methods usually use Euclidean distance (ED), dynamic time warping (DTW), or other methods to measure the global similarity between sequences. The disadvantage of these methods is that their measurement process does not necessarily achieve local sensible matchings of time series, which leads to a decrease in their accuracy and interpretability. To better capture the local information of the sequence, the structure-based methods discretize or symbolize the local value of the time sequence, which leads to the loss of the original information of the sequence. To address these problems, this paper proposes a novel similarity measurement method named dynamic time warping based on the local morphological pattern (MPDTW), which first decomposes the local subsequences of time series using discrete wavelet transforms for extracting the local structure information. Then, the decomposed subsequence will be encoded by the morphological pattern. Finally, the ED between points and their local structure difference based on morphological pattern will be weighted and applied to the DTW algorithm to measure the similarity between sequences. Experiments have been carried out on the classification tasks of the UCR datasets and the results show that our method outperforms the existing baselines.

Automated summary

The key problem in time series classification is the measurement of similarity between time series. In recent years, researchers have paid extensive attention to efficient and accurate methods for measuring the similarity of time series. Existing methods for time series classification can be broadly categorized into shape-based (original value) methods and structure-based (symbol transformation) methods, depending on the similarity measurement strategies they employ. Shape-based methods typically use Euclidean distance (ED), dynamic time warping (DTW), or other methods to measure the overall similarity between sequences. These methods often fail to capture local sensible matchings of time series, resulting in decreased accuracy and interpretability. To address this issue, structure-based methods discretize or symbolize the local values of time sequences, leading to a loss of original information. This paper proposes a novel similarity measurement method called dynamic time warping based on the local morphological pattern (MPDTW), which decomposes local subsequences using discrete wavelet transforms to extract local structure information, encodes the decomposed subsequence using morphological pattern, and applies the weighted ED between points and their local structure difference based on morphological pattern to the DTW algorithm for similarity measurement. Experimental results on the UCR datasets demonstrate that our method outperforms existing baselines.