Change Point Detection Algorithm Research Articles

Real-time Change-Point Detection: A deep neural network-based adaptive approach for detecting changes in multivariate time series data

The behavior of a time series may be affected by various factors. Changes in mean, variance, frequency, and auto-correlation are the most common. Change-Point Detection (CPD) aims to track down abrupt statistical characteristic changes in time series that can benefit many applications in different domains. As demonstrated in recently introduced CPD methodologies, deep learning approaches have the potential to identify more subtle changes. However, due to improper handling of data and insufficient training, these methodologies generate more false alarms and are not efficient enough in detecting change-points. In real-time CPD algorithms, preprocessed data plays a vital role in increasing the algorithm’s efficiency and minimizing false alarm rates. Therefore, preprocessing of data should be a part of the algorithm, but in the existing methods, preprocessing of data is done initially, and then the whole dataset is passed to the CPD algorithm. A new three-phase architecture is proposed to address this issue, in which all phases, from preprocessing to CPD, work in an adaptive manner. The phases are integrated into a pipeline, allowing the algorithm to work in real-time. Our proposed strategy performs optimally and consistently based on performance metrics resulting from experiments on real-world datasets and artifacts. This work effectively addresses the issue of non-stationary data normalization using deep learning approaches. To reduce noise and outliers from the data, a recursive version of singular spectrum analysis is introduced. It is demonstrated that the method’s performance has significantly improved by combining adaptive preprocessing with deep learning CPD techniques.

A comparison of online methods for change point detection in ion-mobility spectrometry data

When on-site classification of volatile organic compounds (VOCs) is required, a portable ion mobility spectrometer (IMS) is a suitable choice. However, the IMS readings often show transient phases before they stabilize. Even so the importance of transient phase and features extracted from it has been highlighted in the literature, it has not, to our knowledge, been used for IMS-based classification so far. This paper analyzes whether change point detection algorithms with low computational complexity can separate transient and stable phases in IMS readings. The algorithms were tested on IMS data from different types of mushrooms. All algorithms successfully detected switches from transient to stable phase. The most accurate results were provided by the previously proposed multivariate max-CUSUM algorithm and the matrix form CUSUM algorithm, which is developed in this paper. • A dataset of IMS measurements from 2 types of mushrooms is presented and shared. • A modification of the CUSUM change point detection algorithm is proposed. • The proposed algorithm is compared to existing change point detection algorithms.

Adaptive application of machine learning models on separate segments of a data sample in regression and classification problems

Introduction: Achievement of specified qualitative indicators in machine learning solutions depends not only on the efficiency of algorithms, but also on data properties. One of the lines for the development of classification and regression models is the specification of local properties of data. Purpose: To improve the qualitative predictors when solving classification and regression problems based on the adaptive selection of various machine learning models on separate local segments of data sample. Results: We propose a method that uses a combination of different models and machine learning algorithms on subsamples in regression and classification problems. The method is based on the calculation of qualitative predictors and the selection of the best models on the local segments of data sample. The finding of transformations of data and time series allows to create sample sets, with the data having different properties (for example, variance, sampling fraction, data range, etc.). We consider the data segmentation based on the change point detection algorithm in time series trends and on analytical information. On the example of the real dataset, we show the experimental values of the loss function for the proposed method with different classifiers on separate segments and on the whole sample. Practical relevance: The results can be used in classification and regression problems for the development of machine learning models and methods. The proposed method allows to improve classification and regression qualitative predictors by assigning models that have the best performance on separate segments.

Blinking-Based Multiplexing: A New Approach for Differentiating Spectrally Overlapped Emitters.

Multicolor single-molecule imaging is widely applied to answer questions in biology and materials science. However, most studies rely on spectrally distinct fluorescent probes or time-intensive sequential imaging strategies to multiplex. Here, we introduce blinking-based multiplexing (BBM), a simple approach to differentiate spectrally overlapped emitters based solely on their intrinsic blinking dynamics. The blinking dynamics of hundreds of rhodamine 6G and CdSe/ZnS quantum dots on glass are obtained using the same acquisition settings and analyzed with a change point detection algorithm. Although substantial blinking heterogeneity is observed, the analysis yields a blinking metric with 93.5% classification accuracy. We further show that BBM with up to 96.6% accuracy is achieved by using a deep learning algorithm for classification. This proof-of-concept study demonstrates that a single emitter can be accurately classified based on its intrinsic blinking dynamics and without the need to probe its spectral color.

Bandit Change-Point Detection for Real-Time Monitoring High-Dimensional Data Under Sampling Control

In many real-world problems of real-time monitoring high-dimensional streaming data, one wants to detect an undesired event or change quickly once it occurs, but under the sampling control constraint in the sense that one might be able to only observe or use selected components data for decision-making per time step in the resource-constrained environments. In this article, we propose to incorporate multi-armed bandit approaches into sequential change-point detection to develop an efficient bandit change-point detection algorithm based on the limiting Bayesian approach to incorporate a prior knowledge of potential changes. Our proposed algorithm, termed Thompson-Sampling-Shiryaev-Roberts-Pollak (TSSRP), consists of two policies per time step: the adaptive sampling policy applies the Thompson Sampling algorithm to balance between exploration for acquiring long-term knowledge and exploitation for immediate reward gain, and the statistical decision policy fuses the local Shiryaev–Roberts–Pollak statistics to determine whether to raise a global alarm by sum shrinkage techniques. Extensive numerical simulations and case studies demonstrate the statistical and computational efficiency of our proposed TSSRP algorithm.

Models of Dynamic Belief Updating in Psychosis-A Review Across Different Computational Approaches.

To understand the dysfunctional mechanisms underlying maladaptive reasoning of psychosis, computational models of decision making have widely been applied over the past decade. Thereby, a particular focus has been on the degree to which beliefs are updated based on new evidence, expressed by the learning rate in computational models. Higher order beliefs about the stability of the environment can determine the attribution of meaningfulness to events that deviate from existing beliefs by interpreting these either as noise or as true systematic changes (volatility). Both, the inappropriate downplaying of important changes as noise (belief update too low) as well as the overly flexible adaptation to random events (belief update too high) were theoretically and empirically linked to symptoms of psychosis. Whereas models with fixed learning rates fail to adjust learning in reaction to dynamic changes, increasingly complex learning models have been adopted in samples with clinical and subclinical psychosis lately. These ranged from advanced reinforcement learning models, over fully Bayesian belief updating models to approximations of fully Bayesian models with hierarchical learning or change point detection algorithms. It remains difficult to draw comparisons across findings of learning alterations in psychosis modeled by different approaches e.g., the Hierarchical Gaussian Filter and change point detection. Therefore, this review aims to summarize and compare computational definitions and findings of dynamic belief updating without perceptual ambiguity in (sub)clinical psychosis across these different mathematical approaches. There was strong heterogeneity in tasks and samples. Overall, individuals with schizophrenia and delusion-proneness showed lower behavioral performance linked to failed differentiation between uninformative noise and environmental change. This was indicated by increased belief updating and an overestimation of volatility, which was associated with cognitive deficits. Correlational evidence for computational mechanisms and positive symptoms is still sparse and might diverge from the group finding of instable beliefs. Based on the reviewed studies, we highlight some aspects to be considered to advance the field with regard to task design, modeling approach, and inclusion of participants across the psychosis spectrum. Taken together, our review shows that computational psychiatry offers powerful tools to advance our mechanistic insights into the cognitive anatomy of psychotic experiences.

Quality Change: Norm or Exception? Measurement, Analysis and Detection of Quality Change in Wikipedia

Wikipedia has been turned into an immensely popular crowd-sourced encyclopedia for information dissemination on numerous versatile topics in the form of subscription free content. It allows anyone to contribute so that the articles remain comprehensive and updated. For enrichment of content without compromising standards, the Wikipedia community enumerates a detailed set of guidelines, which should be followed. Based on these, articles are categorized into several quality classes by the Wikipedia editors with increasing adherence to guidelines. This quality assessment task by editors is laborious as well as demands platform expertise. As a first objective, in this paper, we study evolution of a Wikipedia article with respect to such quality scales. Our results show novel non-intuitive patterns emerging from this exploration. As a second objective we attempt to develop an automated data driven approach for the detection of the early signals influencing the quality change of articles. We posit this as a change point detection problem whereby we represent an article as a time series of consecutive revisions and encode every revision by a set of intuitive features. Finally, various change point detection algorithms are used to efficiently and accurately detect the future change points. We also perform various ablation studies to understand which group of features are most effective in identifying the change points. To the best of our knowledge, this is the first work that rigorously explores English Wikipedia article quality life cycle from the perspective of quality indicators and provides a novel unsupervised page level approach to detect quality switch, which can help in automatic content monitoring in Wikipedia thus contributing significantly to the CSCW community.

AdaPool: A Diurnal-Adaptive Fleet Management Framework Using Model-Free Deep Reinforcement Learning and Change Point Detection

This paper introduces an adaptive model-free deep reinforcement approach that can recognize and adapt to the diurnal patterns in the ride-sharing environment with car-pooling. Deep Reinforcement Learning (RL) suffers from catastrophic forgetting due to being agnostic to the timescale of changes in the distribution of experiences. Although RL algorithms are guaranteed to converge to optimal policies in Markov decision processes (MDPs), this only holds in the presence of static environments. However, this assumption is very restrictive. In many real-world problems like ride-sharing, traffic control, etc., we are dealing with highly dynamic environments, where RL methods yield only sub-optimal decisions. To mitigate this problem in highly dynamic environments, we (1) adopt an online Dirichlet change point detection (ODCP) algorithm to detect the changes in the distribution of experiences, (2) develop a Deep Q Network (DQN) agent that is capable of recognizing diurnal patterns and making informed dispatching decisions according to the changes in the underlying environment. Rather than fixing patterns by time of week, the proposed approach automatically detects that the MDP has changed, and uses the results of the new model. In addition to the adaptation logic in dispatching, this paper also proposes a dynamic, demand aware vehicle-passenger matching and route planning framework that dynamically generates optimal routes for each vehicle based on online demand, vehicle capacities, and locations. Evaluation on New York City Taxi public dataset shows the effectiveness of our approach in improving the fleet utilization, where less than 50% of the fleet are utilized to serve the demand of up to 90% of the requests, while maximizing profits and minimizing idle times.

SCONCE: a method for profiling copy number alterations in cancer evolution using single-cell whole genome sequencing.

MotivationCopy number alterations (CNAs) are a significant driver in cancer growth and development, but remain poorly characterized on the single cell level. Although genome evolution in cancer cells is Markovian through evolutionary time, CNAs are not Markovian along the genome. However, existing methods call copy number profiles with Hidden Markov Models or change point detection algorithms based on changes in observed read depth, corrected by genome content and do not account for the stochastic evolutionary process.ResultsWe present a theoretical framework to use tumor evolutionary history to accurately call CNAs in a principled manner. To model the tumor evolutionary process and account for technical noise from low coverage single-cell whole genome sequencing data, we developed SCONCE, a method based on a Hidden Markov Model to analyze read depth data from tumor cells using matched normal cells as negative controls. Using a combination of public data sets and simulations, we show SCONCE accurately decodes copy number profiles, and provides a useful tool for understanding tumor evolution.Availabilityand implementationSCONCE is implemented in C++11 and is freely available from https://github.com/NielsenBerkeleyLab/sconce.Supplementary information Supplementary data are available at Bioinformatics online.

Monitoring changes in physical activity data during strength training of people with myotonic dystrophy type 1

Myotonic dytrophy type 1 (DM1) is an incurable neuromuscular disease and muscle weakness is a prominent symptom. Research has shown that strength training can be an interesting solution to help with this symptom. Therefore an assistive technology aiming at supervising strength training at home for people with DM1 has been developed and tested in the home of 10 patients for 10 weeks. As many change point detection (CPD) techniques have been used for monitoring change in activity data in the past, no one applied these techniques to physical activities of people with DM1 disease. Hence, physical activity data have been collected during the 10-week experiment and state-of-the-art CPD algorithm has been used to analyze changes in physical activity during the strength training program at home. The results prove that many challenges need to be addressed in this context and could act as a guideline for future works.

Online Change Point Detection for Weighted and Directed Random Dot Product Graphs

Given a sequence of random (directed and weighted) graphs, we address the problem of online monitoring and detection of changes in the underlying data distribution. Our idea is to endow sequential change-point detection (CPD) techniques with a graph representation learning substrate based on the versatile Random Dot Product Graph (RDPG) model. We consider efficient, online updates of a judicious monitoring function, which quantifies the discrepancy between the streaming graph observations and the nominal RDPG. This reference distribution is inferred via spectral embeddings of the first few graphs in the sequence. We characterize the distribution of this running statistic to select thresholds that guarantee error-rate control, and under simplifying approximations we offer insights on the algorithm’s detection resolution and delay. The end result is a lightweight online CPD algorithm, that is also explainable by virtue of the well-appreciated interpretability of RDPG embeddings. This is in stark contrast with most existing graph CPD approaches, which either rely on extensive computation, or they store and process the entire observed time series. An apparent limitation of the RDPG model is its suitability for undirected and unweighted graphs only, a gap we aim to close here to broaden the scope of the CPD framework. Unlike previous proposals, our non-parametric RDPG model for weighted graphs does not require a priori specification of the weights’ distribution to perform inference and estimation. This network modeling contribution is of independent interest beyond CPD. We offer an open-source implementation of the novel online CPD algorithm for weighted and direct graphs, whose effectiveness and efficiency are demonstrated via (reproducible) synthetic and real network data experiments.

A new measure between sets of probability distributions with applications to erratic financial behavior

This paper introduces a new framework to quantify distance between finite sets with uncertainty present, where probability distributions determine the locations of individual elements. Combining this with a Bayesian change point detection algorithm, we produce a new measure of similarity between time series with respect to their structural breaks. First, we demonstrate the algorithm’s effectiveness on a collection of piecewise autoregressive processes. Next, we apply this to financial data to study the erratic behavior profiles of 19 countries and 11 sectors over the past 20 years. Our measure provides quantitative evidence that there is greater collective similarity among sectors’ erratic behavior profiles than those of countries, which we observe upon individual inspection of these time series. Our measure could be used as a new framework or complementary tool for investors seeking to make asset allocation decisions for financial portfolios.

Sliding window change point detection based dynamic network model inference framework for airport ground service process

With the rapid development of civil aviation, the number of airport flights is continually increasing and the low efficiency of flight ground services has become one of the main reasons for flight delays. Obtaining a precise description of the flight ground service network is of great importance to the efficient and safe operation of airports. In this study, we constructed a dynamic network model inference framework based on sliding window change point detection. First, a change point detection algorithm was proposed to divide all time points into two parts, namely, a set of possible change points and a set of completely impossible change points, which greatly reduces the true change point search space. Then, two novel change point elimination judgment rules were designed and combined with the structure learning algorithm to achieve iterative optimization within the possible change point sets, obtain the final true change points, and realize precise dynamic network inference. Based on the experimental results, our proposed method has a higher accuracy and computational efficiency than the state-of-the-art methods, and it can be easily adapted to large-scale datasets. Finally, the proposed model was used in an airport flight ground service analysis. Each day was divided into four stages and the potential dependence relationships between different nodes in each stage were accurately determined. Thus, the model has great practical significance for the accurate scheduling and efficient operation of airport ground services.

Origin of Kinetic Dispersion in Eosin-Sensitized TiO2: Insights from Single-Molecule Spectroscopy

By removing the effects of ensemble averaging and molecular aggregation, we untangle the factors that govern the dispersive electron transfer kinetics of eosin-sensitized TiO2, focusing on the impact of environmental heterogeneity versus injection from multiple excited states. The blinking dynamics of single eosin Y chromophores on nanocrystalline TiO2 films are analyzed using a change point detection algorithm for binned data. Robust statistical analysis based on maximum likelihood estimation, Kolmogorov–Smirnov tests, and log likelihood ratio tests is used to determine the functional form that best fits the resulting on- and off-time distributions and to distinguish between mechanisms for dispersive electron transfer. Using this approach, we find that the on- and off-time distributions for eosin Y on TiO2 are best fit to lognormal distributions corresponding to μon = −0.64 ± 0.04, σon = 1.52 ± 0.02, μoff = −0.23 ± 0.04, and σoff = 1.96 ± 0.03, respectively. Monte Carlo simulations based on the Albery model for dispersive electron transfer (i.e., where the median rate constant κ is modified by the exponential of a parameter, x, that is normally distributed, k = κ e–γx) successfully reproduce this behavior using a median rate constant for injection and back electron transfer of ∼1010 and ∼104 s–1, respectively, and a corresponding energetic dispersion, γ, of ∼200–350 meV. To examine how injection from both the singlet and triplet excited states contributes to this dispersion, we studied two rhodamine sensitizers, R123 and 5ROX, that inject only from their singlet excited state. Surprisingly, when access to the T1 state is minimized in going from EY to R123 and 5ROX, kinetic dispersion actually increases. Collectively, these observations support the interpretation that static and dynamic inhomogeneities at the EY–TiO2 interface govern kinetic dispersion, with dynamic fluctuations in binding configuration and/or vibrational motion playing a decisive role.

Online multivariate changepoint detection with type I error control and constant time/memory updates per series

This article presents a simple algorithm for online multivariate changepoint detection of a mean in rare changepoint settings. The algorithm is based on a modified cusum statistic and guarantees control of the type I error on any false discoveries, while featuring O(1) time and O(1) memory updates per series as well as a proven detection delay.

Spatio-temporal clustering of earthquakes based on distribution of magnitudes

It is expected that the pronounced decrease in b-value of the Gutenberg–Richter law for some region during some time interval can be a promising precursor in forecasting earthquakes with large magnitudes, and thus we address the problem of automatically identifying such spatio-temporal change points as several clusters consisting of earthquakes whose b-values are substantially smaller than the total one. For this purpose, we propose a new method consisting of two phases: tree construction and tree separation. In the former phase, we employ one of two different declustering algorithms called single-link and correlation-metric developed in the field of seismology, while in the later phase, we employ a variant of the change-point detection algorithm, developed in the field of data mining. In the later phase, we also employ one of two different types of objective functions, i.e., the average magnitude which is inversely proportional to the b-value, and the likelihood function based on the Gutenberg–Richter law. Here note that since the magnitudes of most earthquakes are relatively small, we formulate our problem so as to produce one relatively large cluster and the other small clusters having substantially larger average magnitudes or smaller b-values. In addition, in order to characterize some properties of our proposed methods, we present a method of analyzing magnitude correlation over an earthquake network. In our empirical evaluation using earthquake catalog data covering the whole of Japan, we show that our proposed method employing the single-link strategy can produce more desirable results for our purpose in terms of the improvement of weighted sums of variances, average logarithmic likelihoods, visualization results, and magnitude correlation analyses.

Bivariate change point detection: Joint detection of changes in expectation and variance

AbstractA method for change point detection is proposed. We consider a univariate sequence of independent random variables with piecewise constant expectation and variance, apart from which the distribution may vary periodically. We aim to detect change points in both expectation and variance. For that, we propose a statistical test for the null hypothesis of no change points and an algorithm for change point detection. Both are based on a bivariate moving sum approach that jointly evaluates the mean and the empirical variance. The joint consideration helps improve inference compared with separate univariate approaches. We infer on the strength and the type of changes with confidence. Nonparametric methodology supports the analysis of diverse data. Additionally, a multiscale approach addresses complex patterns in change points and effects. We demonstrate the performance through theoretical results and simulation studies. A companion R‐package jcp (available on CRAN) is discussed.

SHM deformation monitoring for high-speed rail track slabs and Bayesian change point detection for the measurements

Fiber Bragg grating (FBG) technology has significant advantages in anti-electromagnetic interference and is suitable for structural health monitoring (SHM) in high-speed rail (HSR). In this paper, a deformation monitoring system based on FBG technology is designed for HSR track slab. Regarding to the possible large deformation of the track slab, a Bayesian change point detection (CPD) method is adopted to detect change points of the monitoring data sequence and the posterior probability distribution of change points can also be derived. The Bayesian CPD algorithm can directly sample and simulate from the posterior probability distribution of change point to determine corresponding parameters of regression model, as well as yielding uncertainty evaluation. The designed SHM system was implemented on a segment of HSR track on the Lanzhou-Xinjiang line in China to carry out the measurement of track slab deformation. Results shows that the designed SHM can resist strong electromagnetic interference in the operation environment of high-speed rail and versine deformation monitoring data over two months for the target track segment can be extracted based on SHM measurement. Through posterior distribution information of the Bayesian CPD method, change points which have the highest posterior probability can be identified for the versine deformation of HSR track slabs, which is essential for investigating the cause of the change point and providing maintenance plan in time.

Embedding-based real-time change point detection with application to activity segmentation in smart home time series data

Human activity recognition systems are essential to enable many assistive applications. Those systems can be sensor-based or vision-based. When sensor-based systems are deployed in real environments, they must segment sensor data streams on the fly in order to extract features and recognize the ongoing activities. This segmentation can be done with different approaches. One effective approach is to employ change point detection (CPD) algorithms to detect activity transitions (i.e. determine when activities start and end). In this paper, we present a novel real-time CPD method to perform activity segmentation, where neural embeddings (vectors of continuous numbers) are used to represent sensor events. Through empirical evaluation with 3 publicly available benchmark datasets, we conclude that our method is useful for segmenting sensor data, offering significant better performance than state of the art algorithms in two of them. Besides, we propose the use of retrofitting, a graph-based technique, to adjust the embeddings and introduce expert knowledge in the activity segmentation task, showing empirically that it can improve the performance of our method using three graphs generated from two sources of information. Finally, we discuss the advantages of our approach regarding computational cost, manual effort reduction (no need of hand-crafted features) and cross-environment possibilities (transfer learning) in comparison to others.

Estimating change-point latent factor models for high-dimensional time series

We consider estimating a factor model for high-dimensional time series that contains structural breaks in the factor loading space at unknown time points. We first study the case when there is one change point in factor loadings, and propose a consistent estimator for the structural break location, whose convergence rate is shown to depend on an interplay between the dimension of the observed time series and the strength of the underlying factor structure. Our results reveal that the asymptotic behavior of the proposed estimator can be asymmetric in the sense that a larger estimation error can occur toward the regime with weaker factor strength. Based on the proposed estimator for the structural break location, we also consider the problem of estimating the factor loading spaces before and after the structural break. We show that the proposed estimators for change-point location and loading spaces are consistent when the numbers of factors are correctly estimated or overestimated. The algorithm for multiple change-point detection is also developed in the paper. Compared with existing results on change-point factor analyses of high-dimensional time series, a distinguished feature of the current paper is that the noise process is not necessarily assumed to be idiosyncratic and as a result we allow the noise process with potentially strong cross-sectional dependence. Another advantage for the proposed method is that it is specifically designed for the changes in the factor loading space and the stationarity assumption is not imposed on either the factor or noise process, while most existing methods for change-point detection of high-dimensional time series with/without a factor structure require the data to be stationary or ’close’ to a stationary process between two change points, which is rather restrictive. Numerical experiments including a Monte Carlo simulation and a real data application are presented to illustrate the proposed estimators perform well.