English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Time series shapelets are discriminative subsequences that have been recentlyfound effective for time series clustering (TSC). The shapelets are convenientfor interpreting the clusters. Thus, the main challenge for TSC is to discoverhigh-quality variable-length shapelets to discriminate different clusters. Inthis paper, we propose a novel autoencoder-shapelet approach (AUTOSHAPE), whichis the first study to take the advantage of both autoencoder and shapelet fordetermining shapelets in an unsupervised manner. An autoencoder is speciallydesigned to learn high-quality shapelets. More specifically, for guiding thelatent representation learning, we employ the latest self-supervised loss tolearn the unified embeddings for variable-length shapelet candidates (timeseries subsequences) of different variables, and propose the diversity loss toselect the discriminating embeddings in the unified space. We introduce thereconstruction loss to recover shapelets in the original time series space forclustering. Finally, we adopt Davies Bouldin index (DBI) to inform AUTOSHAPE ofthe clustering performance during learning. We present extensive experiments onAUTOSHAPE. To evaluate the clustering performance on univariate time series(UTS), we compare AUTOSHAPE with 15 representative methods using UCR archivedatasets. To study the performance of multivariate time series (MTS), weevaluate AUTOSHAPE on 30 UEA archive datasets with 5 competitive methods. Theresults validate that AUTOSHAPE is the best among all the methods compared. Weinterpret clusters with shapelets, and can obtain interesting intuitions aboutclusters in two UTS case studies and one MTS case study, respectively.

AUTOSHAPE: An Autoencoder-Shapelet Approach for Time Series Clustering