Subsequence Time Series Clustering Research Articles

Subsequence Time Series Clustering-Based Unsupervised Approach for Anomaly Detection of Axial Piston Pumps

Axial piston pump is the key component of a hydraulic system. The reliability of the axial piston pump influences the reliability of the fluid power system directly. Discharge pressure signals are easy to obtain and can reflect the dynamic performance of the axial piston pump. Although many studies have been developed for fault diagnosis of the axial piston pump, most of these methods are based on supervised artificial intelligence-based methods that rely on massive labeled data. However, in practice application, the external load of the hydraulic system often varies and it is quite impractical or expensive to obtain massive labeled data. This article proposes a novel Subsequence Time Series(STS) clustering based unsupervised approach for anomaly detection of the axial piston pump using discharge pressure signal. The proposed approach comprises two stages, norm cluster search, and anomaly subsequence clustering. The proposed approach performs multiple STS clustering to search the norm cluster whose center can encode the time series better. The proposed approach comprises of four modules: motif discovery, parameter-free minimum description length(MDL) clustering, subsequence search, and scoring the norm cluster. Subsequence search via dynamic time warping(DTW) enables the approach to discover the subsequences of variable length. In particular, weak fault signal detection is achieved by evaluating the local distribution of subsequence. The effectiveness of the proposed approach is validated through experiments under different external loads.

Earthquake pattern analysis using subsequence time series clustering.

In this paper, a subsequence time-series clustering algorithm is proposed to identify the strongly coupled aftershocks sequences and Poissonian background activity from earthquake catalogs of active regions. The proposed method considers the inter-event time statistics between the successive pair of events for characterizing the nature of temporal sequences and observing their relevance with earthquake epicenters and magnitude information simultaneously. This approach categorizes the long-earthquake time series into the finite meaningful temporal sequences and then applies the clustering mechanism to the selective sequences. The proposed approach is built on two phases: (1) a Gaussian kernel-based density estimation for finding the optimal subsequence of given earthquake time-series, and (2) inter-event time (varDelta t) and distance-based observation of each subsequence for checking the presence of highly correlated aftershock sequences (hot-spots) in it. The existence of aftershocks is determined based on the coefficient of variation (COV). A sliding temporal window on varDelta t with earthquake’s magnitude M is applied on the selective subsequence to filter out the presence of time-correlated events and make the meaningful time stationary Poissonian subsequences. This proposed approach is applied to the regional Sumatra-Andaman (2000–2021) and worldwide ISC-GEM (2000–2016) earthquake catalog. Simulation results indicate that meaningful subsequences (background events) can be modeled by a homogeneous Poisson process after achieving a linear cumulative rate and time-independent lambda in the exponential distribution of varDelta t. The relations COV_{a}(T)>COV_{o}(T)> (COV_{b}(T)approx 1) and COV_{a}(d)>COV_{o}(d)>COV_{b}(d) are achieved for both studied catalogs. Comparative analysis justifies the competitive performance of the proposed approach to the state-of-art approaches and recently introduced methods.

Interpretation Method for Continuous Glucose Monitoring with Subsequence Time-Series Clustering.

We propose mini-batch top-n k-medoids to sequential pattern mining to improve CGM interpretation. Mecical workers can treat specific patient groups better by understanding the time series variation of blood glucose results. For 10 years, continuous glucose monitoring (CGM) has provided time-series data of blood glucose thanks to the invention of devices with low measurement errors. We conducted two experiments. In the first experiment, we evaluated the proposed method with a manually created dataset and confirmed that the method provides more accurate patterns than other clustering methods. In the second experiment, we applied the proposed method to a CGM dataset consisting of real data from 163 patients. We created two labels based on blood glucose (BG) statistics and found patterns that correlated with a specific label in each case.

Unsupervised Feature Learning From Time-Series Data Using Linear Models

In the Internet of Things (IoT), heterogenous sensors generate time-series data with different properties. The problem of unsupervised feature learning from a time-series dataset poses two challenges. First, it is known that centroids obtained by clustering time-series with high overlap do not reflect their patterns, i.e., subsequence time-series clustering is meaningless . In this paper, we show that principal component analysis, sparse coding, and non-negative matrix factorization are also meaningless for the same task, and that the systematic approach to learning meaningful features from time-series is by using the shift-invariant versions of these algorithms. Second, by comparing their shift-invariant versions on different kinds of time-series for reconstruction, prediction and classification, we show that no one algorithm is best suited for all time-series. This comparison leads to a method for automatically selecting the suitable feature learning algorithm for a given time-series dataset based on its structural properties. Generality of the method and significance of the structural properties are examined using statistical tests. The method can be implemented as a simple logic circuit, convenient for embedding in IoT hardware.

A review of subsequence time series clustering.

Clustering of subsequence time series remains an open issue in time series clustering. Subsequence time series clustering is used in different fields, such as e-commerce, outlier detection, speech recognition, biological systems, DNA recognition, and text mining. One of the useful fields in the domain of subsequence time series clustering is pattern recognition. To improve this field, a sequence of time series data is used. This paper reviews some definitions and backgrounds related to subsequence time series clustering. The categorization of the literature reviews is divided into three groups: preproof, interproof, and postproof period. Moreover, various state-of-the-art approaches in performing subsequence time series clustering are discussed under each of the following categories. The strengths and weaknesses of the employed methods are evaluated as potential issues for future studies.

Selective Subsequence Time Series clustering

Subsequence Time Series (STS) Clustering is a time series mining task used to discover clusters of interesting subsequences in time series data. Many research works had used this algorithm as a subroutine in rule discovery, indexing, classification and anomaly detection. Unfortunately, recent work has demonstrated that almost all of the STS clustering algorithms give meaningless results, as their outputs are always produced in sine wave form, and do not associate with actual patterns of the input data. Consequently, algorithms that use the results from the STS clustering as their input will fail to produce its meaningful output. In this work, we propose a new STS clustering framework for time series data called Selective Subsequence Time Series (SSTS) clustering which provides meaningful results by using an idea of data encoding to cluster only essential subsequences. Furthermore, our algorithm also automatically determines an appropriate number of clusters without user’s intervention.

Frequency Analysis of Subsequence Time Series Clustering and Evaluation of Phase Alignment Preprocessing

Although Subsequence Time Series (STS) clustering has been one of the most popular techniques to extract typical subsequence patterns from time-series data, previous studies have gave surprising reports that cluster centers obtained using STS clustering closely resemble ``sine waves'' with little relation to input time-series data. This means that STS clustering cannot be used for its original purpose, extraction of typical subsequences. Despite this serious fact, its mathematical structure has seldom been studied. The main contribution of this paper is that we give a theoretical analysis of STS clustering from a frequency-analysis viewpoint and identify that sine waves are generated due to the superposition of time series subsequences, which have the same spectra but different phases. Another contribution is that we propose a clustering algorithm, which uses a phase alignment preprocessing, to avoid sine-wave patterns.

A TIME SERIES KNOWLEDGE MINING FRAMEWORK EXPLOITING THE SYNERGY BETWEEN SUBSEQUENCE CLUSTERING AND PREDICTIVE MARKOVIAN MODELS

This paper proposes a time series knowledge mining framework, designed to favor the synergy between subsequence time series clustering and predictive tools such as Hidden Markov Models. Many tasks for temporal data mining rely heavily on the choice of the representation scheme and the dissimilarity measure. The first part is concerned with detailed representation taxonomy for numeric and symbolic time series and comprehensive categorization of distance measures. Subsequence time series clustering methods with a sliding window are addressed in the second part and a generalization of Fuzzy C-Means algorithm based on the dynamic time warping distance is proposed as a very effective solution. This involves a shape-based distance tolerant to phase shifts in time or accelerations/decelerations along the time axis. It also allows to determine the degree to which set-defined objects, such as subsequence time series and their cluster centroids (similar in nature) differ from each other. In the third part we discuss the integration of clustering algorithms with probabilistic predictive tools, such as discrete Markov chains or hidden Markov models. We apply these techniques to clustering of non-overlapping sequences extracted from Standard and Poor’s 500 stock index historical data and we suggest different integrations with markovian models to improve the predictive power