This repository provides a Python package for computing a multivariate time series subsequence clustering metric1. The purpose is to have a meaningful metric for comparing time-series clustering algorithms. Motivation To our knowledge no existing clustering metric exists, that takes the time space variat...
Suh WH, Oh S, Ahn CW (2023) Metaheuristic-based time series clustering for anomaly detection in manufacturing industry. Appl Intell 53(19):21723–21742. https://doi.org/10.1007/s10489-023-04594-5 Article Google Scholar Wang H, Lu W, Tang S et al (2022) Predict industrial equipment fa...
DUET, which introduces a DUal clustering on the temporal and channel dimensions to Enhance multivariate Time series forecasting. Specifically, it clusters sub-series into fine-grained distributions with the TCM to better model the heterogeneity of temporal patterns. It also utilizes a Channel-Soft-...
Time series data analysis, especially forecasting, classification, imputation, and anomaly detection, has gained a lot of research attention in recent year
Multivariate chaotic time series prediction has been a challenging problem, especially the coupling relationship between multiple variables need to be carefully considered. We propose a self-attention architecture with an information interaction module, called mixformer, applied to the multivariate chaotic ti...
In brief, representative clusters of the state space \(\varvec{Z}_s\) and \(\varvec{W}_s\) are obtained using clustering methods, such as k-means clustering, where k can be seen as the number of hidden neurons in a GRBF neural network. Let \(c_f\) and \(c_g\) be the ...
However, it is crucial to recognize that the applicability of clustering methods is constrained by specific applications due to their varying complexities. The window-based method divides the sequence into multiple overlapping subsequences and calculates the anomaly score for each window. By calculating ...
upward masking strategy is designed in MHCCL to remove outliers of clusters at each partition to refine prototypes, which helps speed up the hierarchical clustering process and improves the clustering quality. We conduct experimental evaluations on seven widely-used multivariate time series datasets. The...
A Python toolbox/library for machine learning on partially-observed time series with PyTorch, including SOTA models supporting tasks of imputation, classification, clustering, and forecasting on incomplete (irregularly-sampled) multivariate time series with missing values. https://arxiv.org/abs/2305.18811...
as was provided by the DACON’s HAICon2021 competition [17]. The approach showed a good performance, achieving F1 of 0.926 when the provided code was run on the HAI 2.0 dataset [4]. We carried out preliminary evaluations to test if this approach can be applied to the time-series network...