ClaSP: parameter-free time series segmentation
From MaRDI portal
Abstract: The study of natural and human-made processes often results in long sequences of temporally-ordered values, aka time series (TS). Such processes often consist of multiple states, e.g. operating modes of a machine, such that state changes in the observed processes result in changes in the distribution of shape of the measured values. Time series segmentation (TSS) tries to find such changes in TS post-hoc to deduce changes in the data-generating process. TSS is typically approached as an unsupervised learning problem aiming at the identification of segments distinguishable by some statistical property. Current algorithms for TSS require domain-dependent hyper-parameters to be set by the user, make assumptions about the TS value distribution or the types of detectable changes which limits their applicability. Common hyperparameters are the measure of segment homogeneity and the number of change points, which are particularly hard to tune for each data set. We present ClaSP, a novel, highly accurate, hyper-parameter-free and domain-agnostic method for TSS. ClaSP hierarchically splits a TS into two parts. A change point is determined by training a binary TS classifier for each possible split point and selecting the one split that is best at identifying subsequences to be from either of the partitions. ClaSP learns its main two model-parameters from the data using two novel bespoke algorithms. In our experimental evaluation using a benchmark of 107 data sets, we show that ClaSP outperforms the state of the art in terms of accuracy and is fast and scalable. Furthermore, we highlight properties of ClaSP using several real-world case studies.
Recommendations
- Time-series segmentation: A model and a method
- New approach to the segmentation problem for time series of arbitrary nature
- Model-based clustering and segmentation of time series with changes in regime
- Domain agnostic online semantic segmentation for multi-dimensional time series
- scientific article; zbMATH DE number 2166729
Cites work
- A Cluster Analysis Method for Grouping Means in the Analysis of Variance
- A time series forest for classification and feature extraction
- Domain agnostic online semantic segmentation for multi-dimensional time series
- Gaussian process for nonstationary time series prediction
- Graph-based change-point detection
- On tests for detecting change in mean
- Optimal detection of changepoints with a linear computational cost
- ROCKET: exceptionally fast and accurate time series classification using random convolutional kernels
- Sequential change‐point detection based on direct density‐ratio estimation
- Statistical comparisons of classifiers over multiple data sets
- Technical note: Some properties of splitting criteria
- The BOSS is concerned with time series classification in the presence of noise
Cited in
(7)- MultiETSC: automated machine learning for early time series classification
- Logical clustering and learning for time-series data
- Domain agnostic online semantic segmentation for multi-dimensional time series
- SAZED: parameter-free domain-agnostic season length estimation in time series data
- PaSTiLa: scalable parallel algorithm for unsupervised labeling of long time series
- Deep clustering for large-scale interpretable time series segmentation
- Penalty learning for optimal partitioning using multilayer perceptron
This page was built for publication: ClaSP: parameter-free time series segmentation
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6040514)
