Time Series Segmentation Research Articles

Shapelet selection for time series classification

In recent times, increasing attention has been given to shapelet-based methods for time series classification. However, in the majority of current methods, similar subsequences were often selected as shapelets, thereby reducing the final interpretability of these methods. Aiming to circumvent the selection of similar subsequences as the final shapelets, a novel shapelet selection method (SSM) was proposed in this paper. Firstly, shapelet candidates were generated by SSM through time series segmentation to avoid excessive generation of similar candidates from a single time series. Secondly, all shapelet candidates were evaluated simultaneously to improve evaluation efficiency. Finally, SSM introduced a position-based filter to prevent the selection of similar sequences repeatedly. The results obtained on the UCR TSC archive demonstrated the effectiveness of the proposed method.

An Efficient Algorithm for Exact Segmentation of Large Compositional and Categorical Time Series

ABSTRACTChange‐point detection, also known as signal segmentation, is an essential preprocessing step in many applications, ranging from industrial monitoring to bioinformatics. In short, it consists in finding the temporal boundaries of homogeneous regimes in long and non‐stationary time series. While this area of research is active, most existing methods are designed for Euclidean data. However, in many practical scenarios, the collected time series are compositional, meaning that each observation belongs to the probability simplex (the set of non‐negative vectors whose components sum to one). In this work, we propose an algorithm detecting change‐points in large compositional signals with an underlying piecewise stationary model. We cast the change‐point detection task as a discrete optimization problem, whose solution is shown to converge to the true change‐points. We introduce a new and time‐efficient dynamic programming algorithm that solves exactly this problem. To limit the number of operations, we describe a novel pruning rule that allows us to reduce the set of candidate change‐point indices. Our method is tested on a thorough simulation study, which confirms its efficiency. Additionally, we apply our method to a human activity segmentation task, highlighting the necessity for such novel techniques compared to standard algorithms.

Adaptive feature fusion and disturbance correction for accurate remaining useful life prediction of rolling bearings

As a key component in the transmission system of high-speed trains, bearings need to withstand certain loads while rotating at high speeds. Once a failure occurs, it can directly affect the safety of train operations. Therefore, it is of great significance to establish a reliable remaining useful life model to ensure train operation safety. Addressing the problems of redundant features in the existing multi-feature fusion process, which affects diagnostic performance, and the spurious fluctuations in the fusion features, which cause inaccurate determination of the start fault time and lead to low prediction accuracy of remaining useful life, we propose a method based on adaptive feature fusion and the autoregressive integrated moving average model for rolling bearing prediction. Firstly, features from the time domain, frequency domain, and entropy are extracted. A feature selection mechanism with minimum redundancy is constructed to screen the optimal sensitive feature set. Secondly, based on adaptive feature fusion, the optimal sensitive feature set is dynamically fused, and the spurious fluctuations of the health index are corrected using linear regression and the 3σ principle. Next, a bottom-up time series segmentation method is employed to divide the health status of the Improved Health Indicators. Finally, a remaining useful life prediction model based on the autoregressive integrated moving average model is established. This study demonstrates that the proposed method effectively identifies features that are most sensitive to degradation trends, accurately determines the initial fault moment of bearings, and achieves effective prediction of the remaining useful life of bearings.

Design of Time Series Fuzzy Segmentation Algorithm for Communication Data Classification

In order to solve the problem that the imbalance of communication data sets leads to a significant increase in classification difficulty, a classification algorithm for fuzzy segmentation of time series is proposed. The principal component analysis method is used to obtain the eigenvector with the largest eigenvalue. The time series of data is established according to the interval number theory. The segmentation target of the communication data time series is characterized by the Langmuir distance measurement function between data and categories. The classification result of fuzzy segmentation is obtained based on the judgment relationship between the difference of fuzzy classification matrix and the closing condition. The experimental results show that the number of data in the three classification situations of the algorithm in this paper is always at the corresponding ideal level, with high accuracy and low error and failure.

SIMTSeg: A self-supervised multivariate time series segmentation method with periodic subspace projection and reverse diffusion for industrial process

Subsequences with varied regimes in the industrial multivariate time series (MTS) are closely associated with the dynamic status of the multi-phased industrial process. Time series segmentation (TSS) provides insights into the underlying behavior of industrial systems. However, the complexity of industrial data poses significant challenges to the conventional TSS methods. Motivated by this, a Self-supervised Industrial Multivariate Time-series Segmentation method (SIMTSeg) is presented in this work. An MTS folding module based on Ramanujan periodic subspace projection is first proposed, where the MTS is reshaped into the 3D feature map to realize the compact representation of the intricate data dependencies. Subsequently, a self-supervised module based on the encoder-decoder architecture is adopted to address the problem of deficient and task-specific annotations in industrial data. The folded feature map is denoised step by step following the reverse diffusion process, and finally turns into the segmentation mask without redundant details. The proposed SIMTSeg has been validated by a popular industrial benchmark, the Tennessee Eastman Process, and outperforms the unsupervised data-driven baselines in terms of various performance metrics. SIMTSeg has no prerequisite on the number of segmentation points or regime types, and is capable of giving more meaningful segmentation results that are in line with the high-level semantics.

Deep‐Learning Based Causal Inference: A Feasibility Study Based on Three Years of Tectonic‐Climate Data From Moxa Geodynamic Observatory

AbstractHighly sensitive laser strainmeters at Moxa Geodynamic Observatory (MGO) measure motions of the upper Earth's crust. Since the mountain overburden of the laser strainmeters installed in the gallery of the observatory is relatively low, the recorded time series are strongly influenced by local meteorological phenomena. To estimate the nonlinear effect of the meteorological variables on strain measurements in a non‐stationary environment, advanced methods capable of learning the nonlinearity and discovering causal relationships in the non‐stationary multivariate tectonic‐climate time series are needed. Methods for causal inference generally perform well in identifying linear causal relationships but often struggle to retrieve complex nonlinear causal structures prevalent in real‐world systems. This work presents a novel model invariance‐based causal discovery (CDMI) method that utilizes deep networks to model nonlinearity in a multivariate time series system. We propose to use the theoretically well‐established Knockoffs framework to generate in‐distribution, uncorrelated copies of the original data as interventional variables and test the model invariance for causal discovery. To deal with the non‐stationary behavior of the tectonic‐climate time series recorded at the MGO, we propose a regime identification approach that we apply before causal analysis to generate segments of time series that possess locally consistent statistical properties. First, we evaluate our method on synthetically generated time series by comparing it to other causal analysis methods. We then investigate the hypothesized effect of meteorological variables on strain measurements. Our approach outperforms other causality methods and provides meaningful insights into tectonic‐climate causal interactions.

Self-Supervised Learning across the Spectrum

Satellite image time series (SITS) segmentation is crucial for many applications, like environmental monitoring, land cover mapping, and agricultural crop type classification. However, training models for SITS segmentation remains a challenging task due to the lack of abundant training data, which requires fine-grained annotation. We propose S4, a new self-supervised pretraining approach that significantly reduces the requirement for labeled training data by utilizing two key insights of satellite imagery: (a) Satellites capture images in different parts of the spectrum, such as radio frequencies and visible frequencies. (b) Satellite imagery is geo-registered, allowing for fine-grained spatial alignment. We use these insights to formulate pretraining tasks in S4. To the best of our knowledge, S4 is the first multimodal and temporal approach for SITS segmentation. S4’s novelty stems from leveraging multiple properties required for SITS self-supervision: (1) multiple modalities, (2) temporal information, and (3) pixel-level feature extraction. We also curate m2s2-SITS, a large-scale dataset of unlabeled, spatially aligned, multimodal, and geographic-specific SITS that serves as representative pretraining data for S4. Finally, we evaluate S4 on multiple SITS segmentation datasets and demonstrate its efficacy against competing baselines while using limited labeled data. Through a series of extensive comparisons and ablation studies, we demonstrate S4’s ability as an effective feature extractor for downstream semantic segmentation.

Unsupervised decomposition of natural monkey behavior into a sequence of motion motifs

Nonhuman primates (NHPs) exhibit complex and diverse behavior that typifies advanced cognitive function and social communication, but quantitative and systematical measure of this natural nonverbal processing has been a technical challenge. Specifically, a method is required to automatically segment time series of behavior into elemental motion motifs, much like finding meaningful words in character strings. Here, we propose a solution called SyntacticMotionParser (SMP), a general-purpose unsupervised behavior parsing algorithm using a nonparametric Bayesian model. Using three-dimensional posture-tracking data from NHPs, SMP automatically outputs an optimized sequence of latent motion motifs classified into the most likely number of states. When applied to behavioral datasets from common marmosets and rhesus monkeys, SMP outperformed conventional posture-clustering models and detected a set of behavioral ethograms from publicly available data. SMP also quantified and visualized the behavioral effects of chemogenetic neural manipulations. SMP thus has the potential to dramatically improve our understanding of natural NHP behavior in a variety of contexts.

Multi-scale convolution enhanced transformer for multivariate long-term time series forecasting

In data analysis and forecasting, particularly for multivariate long-term time series, challenges persist. The Transformer model in deep learning methods has shown significant potential in time series forecasting. The Transformer model’s dot-product attention mechanism, however, due to its quadratic computational complexity, impairs training and forecasting efficiency. In addition, the Transformer architecture has limitations in modeling local features and dealing with multivariate cross-dimensional dependency relationship. In this article, a Multi-Scale Convolution Enhanced Transformer model (MSCformer) is proposed for multivariate long-term time series forecasting. As an alternative to modeling the time series in its entirety, a segmentation strategy is designed to convert the input original series into segmented forms with different lengths, then process time series segments using a new constructed multi-Dependency Aggregation module. This multi-Scale segmentation approach reduces the computational complexity of the attention mechanism part in subsequent models, and for each segment of length corresponds to a specific time scale, it also ensures that each segment retains the semantic information of the data sequence level, thereby comprehensively utilizing the multi-scale information of the data while more accurately capturing the real dependency of the time series. The Multi-Dependence Aggregate module captures both cross-temporal and cross-dimensional dependencies of multivariate long-term time series and compensates for local dependencies within the segments thereby captures local series features comprehensively and addressing the issue of insufficient information utilization. MSCformer synthesizes dependency information extracted from various temporal segments at different scales and reconstructs future series using linear layers. MSCformer exhibits higher forecasting accuracy, outperforming existing methods in multiple domains including energy, transportation, weather, electricity, disease and finance.

Optimal segmentation of non-linear and non-stationary time series based on fractal dimension and Poincare section and its application in solving EEG-signal

Non-linearity and non-stationarity characteristics in physiological and biological time series limit the reliability of conventional analysis. The main challenge in pattern recognition problems is determining the Optimal-Window-Duration (OWD) for segmenting these time series. Our study presents an innovative algorithm, based on Individual-Optimal-Segmentation (IOS), which integrates chaos theory using fractal dimension, and Poincare section into the estimation of the OWD for segmenting time series with non-linear and non-stationary characteristics. In the logistic map case study, the IOS algorithm proved to be effective in detecting cycles in a range of numerical examples, from period-2 to chaotic behavior. So that its performance was superior to previous approaches, especially when it comes to detecting chaotic behaviors. Furthermore, the combination of the IOS algorithm with the Welch signal processing method shows promise in identifying abnormal Power-Spectral-Density (PSD) curves related to Beta and Gamma rhythms in depression diagnosis scenarios. The generated depression diagnosis system attained an accuracy (%) of 99.07±0.35 (with 2-channel) using the Monte Carlo simulation method. These findings suggest a potentially valuable tool for addressing segmentation challenges in time series analysis, particularly in contexts involving complex, dynamic data. While, it is suggested that future research investigate the computational efficiency and robustness of the IOS algorithm when handling large-scale or real-time data and increase its application in various fields in data science and signal processing. Furthermore, expanding the application range of the IOS algorithm for other pattern recognition problems with physiological or biological data can strengthen the generalizability and practical application of the algorithm.

A Computational Framework for Intraoperative Pupil Analysis in Cataract Surgery

Pupillary instability is a known risk factor for complications in cataract surgery. This study aims to develop and validate an innovative and reliable computational framework for the automated assessment of pupil morphologic changes during the various phases of cataract surgery. Retrospective surgical video analysis. Two hundred forty complete surgical video recordings, among which 190 surgeries were conducted without the use of pupil expansion devices (PEDs) and 50 were performed with the use of a PED. The proposed framework consists of 3 stages: feature extraction, deep learning (DL)-based anatomy recognition, and obstruction (OB) detection/compensation. In the first stage, surgical video frames undergo noise reduction using a tensor-based wavelet feature extraction method. In the second stage, DL-based segmentation models are trained and employed to segment the pupil, limbus, and palpebral fissure. In the third stage, obstructed visualization of the pupil is detected and compensated for using a DL-based algorithm. A dataset of 5700 intraoperative video frames across 190 cataract surgeries in the BigCat database was collected for validating algorithm performance. The pupil analysis framework was assessed on the basis of segmentation performance for both obstructed and unobstructed pupils. Classification performance of models utilizing the segmented pupil time series to predict surgeon use of a PED was also assessed. An architecture based on the Feature Pyramid Network model with Visual Geometry Group 16 backbone integrated with the adaptive wavelet tensor feature extraction feature extraction method demonstrated the highest performance in anatomy segmentation, with Dice coefficient of 96.52%. Incorporation of an OB compensation algorithm improved performance further (Dice 96.82%). Downstream analysis of framework output enabled the development of a Support Vector Machine-based classifier that could predict surgeon usage of a PED prior to its placement with 96.67% accuracy and area under the curve of 99.44%. The experimental results demonstrate that the proposed framework (1) provides high accuracy in pupil analysis compared with human-annotated ground truth, (2) substantially outperforms isolated use of a DL segmentation model, and (3) can enable downstream analytics with clinically valuable predictive capacity. Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.

Desertification in northern China from 2000 to 2020: The spatial–temporal processes and driving mechanisms

Desertification is one of the most significant environmental and social challenges globally. Monitoring desertification dynamics and quantitatively identifying the contributions of its driving factors are crucial for land restoration and sustainable development. This study develops a standardized methodological framework that combines desertification dynamics with driving mechanisms at the pixel level, applied to northern China from 2000 to 2020. Using multisource data and employing the Time Series Segmentation and Residual Trend analysis (TSS-RESTREND) method alongside geographical detector, we quantitatively assessed desertification reversion, expansion, and abrupt change processes, along with the impacts and interactions of natural and human factors were quantitatively assessed. Over the past two decades, the proportion of desertified land decreased by 5.60%. Notably, 32.88% of the study area experienced significant desertification reversion, while only 5.86% underwent expansion. Abrupt changes in both reversed and expanding areas were observed, primarily in the central and western regions, with these changes concentrated in the periods of 2009–2011 and 2014–2016. The impacts of various factors in different sub-regions exhibited significant spatial heterogeneity. Increased precipitation, temperature, and evapotranspiration contributed to reversion in the western area, while decreased wind speed influenced the eastern area. Additionally, decreased population density and afforestation activities also promoted desertification reversion. In contrast, decreased precipitation and increased temperature contributed to expansion in the western and eastern areas, respectively, with increased population density exacerbating this process. Overall, the interactions between natural and human factors were enhanced. Future desertification control and ecological engineering planning should focus on the coupling effects of different driving factors and relevant abrupt vegetation changes.

2020 CMS prior authorization for hospital outpatient departments: Associated surgical volume impact

BackgroundPrior authorization is common for privately administered Medicare Advantage plans but is rarely used for surgical care when considering publicly administered plans. A 2020 Centers for Medicare and Medicaid services (CMS) policy, CMS-1717-FC, requires prior authorization for Medicare Fee-for-Service beneficiaries undergoing select procedures (blepharoplasty, abdominoplasty, botulinum toxin injection, rhinoplasty, and vein ablation) in hospital outpatient departments. The impact of this policy on surgical volume at hospital outpatient departments and shifts in care to ambulatory surgery centers is unknown. MethodsThis study used a segmented interrupted time series and pre-post logistic regression model. This study was a retrospective cohort study using data from the Healthcare Cost and Utilization Project state ambulatory surgery database and state inpatient database. ResultsFrom 2016 through 2021, a total of 272,879 patients underwent the affected procedures. Pre–CMS-1717-FC, a trend of decreasing hospital outpatient department utilization was found for Medicare Fee-for-Service beneficiaries (–10.82, 95% confidence interval: –18.32 to –3.33, P = .01). In the post-implementation period, no change in the rate of decreasing hospital outpatient department utilization was found for Medicare Fee-for-Service beneficiaries (–3.45, 95% confidence interval: –36.15 to 29.25, P = .83). In the pre-policy period, Medicare Fee-for-Service beneficiaries were 46% less likely to use freestanding ambulatory surgery centers but 27% less likely to use hospital-owned ambulatory surgery centers. ConclusionCMS-1717-FC was not associated with significant changes in hospital outpatient department volume beyond baseline trends. Policy aiming to right-size prior authorization for these procedures and considering site-of-service will balance the need to ensure medical necessity while constraining costs.

Employer-Sponsored Medicare Advantage Plans and the 2018 Therapy Cap Repeal: Reduced Overall Spending Does Not Constrain Out-of-Pocket Costs.

Policy impacting traditional Medicare beneficiaries may have unintended effects for privately insured patients. After the repeal of a longstanding $1500 outpatient therapy cap in 2018, we aimed to evaluate if this policy change was associated with differences in use of cost of postoperative therapy after common hand surgeries, including carpal tunnel release, trigger finger release, ganglion cyst excision, De Quervain tenosynovitis release, carpometacarpal arthroplasty, and distal radius fracture open reduction/internal fixation or percutaneous pinning. The Medicare Supplement and Coordination of Benefits files from Marketscan were used. Frequency of therapy appointments, overall costs, and out-of-pocket costs were obtained. A segmented interrupted time series with Poisson and log-transformed linear regression was performed. No significant monthly change in odds of therapy use was found in the postpolicy period for patients who underwent trigger finger release, carpal tunnel release, Ganglion cyst excision, De Quervain tenosynovitis release, carpometacarpal arthroplasty, or distal radius fracture, pinning, or open reduction/internal fixation. Overall cost decreased in the postpolicy period by 2% for comprehensive plans (95% confidence interval [CI]: -0.03 to -0.01, P < 0.001), by 7% for those with exclusive provider organizations (95% CI: -0.10 to -0.04, P < 0.001), by 1% for HMOs (95% CI: -0.01 to 0.002, P = 0.01), and by 3% for preferred provider organizations (95% CI: -0.03 to -0.02, P < 0.001). In the postpolicy period, no monthly change in out-of-pocket cost was observed for patients with comprehensive, exclusive provider organization, health maintenance organization, preferred provider organization, or point of service with capitation insurance plans. Patients with employer-sponsored Medicare Advantage plans experienced increased out-of-pocket costs for therapy despite lower net costs. These data highlight an urgent need for policy ensuring that patients benefit when overall costs of care decrease.

Involution symmetry quantification using recurrences.

Symmetries are ubiquitous in science, aiding theoretical comprehension by discerning patterns in mathematical models and natural phenomena. This work introduces a method for assessing the extent of symmetry within a time series. We explore both microscopic and macroscopic features extracted from a recurrence plot. By analyzing the statistics of small recurrence matrices, our approach delves into microscale dynamics, facilitating the identification of symmetric time series segments through diagonal macroscale structures on a recurrence plot. We validate our approach by successfully quantifying involution symmetries for three-dimensional dynamical models, specifically, order-2 rotational symmetry in the Lorenz '63 model, and inversion symmetry in the Chua circuit. Our quantifier also detects symmetry breaking in the modified Lorenz model for El Niño phenomenon. The method can be applied in a versatile manner, not only to three-dimensional trajectories but also to univariate time series. Symmetry quantification in time series is promising for enhancing dynamical system modeling and profiling.

Identifying vital nodes for yeast network by dynamic network entropy

BackgroundThe progress of the cell cycle of yeast involves the regulatory relationships between genes and the interactions proteins. However, it is still obscure which type of protein plays a decisive role in regulation and how to identify the vital nodes in the regulatory network. To elucidate the sensitive node or gene in the progression of yeast, here, we select 8 crucial regulatory factors from the yeast cell cycle to decipher a specific network and propose a simple mixed K2 algorithm to identify effectively the sensitive nodes and genes in the evolution of yeast.ResultsConsidering the multivariate of cell cycle data, we first utilize the K2 algorithm limited to the stationary interval for the time series segmentation to measure the scores for refining the specific network. After that, we employ the network entropy to effectively screen the obtained specific network, and simulate the gene expression data by a normal distribution approximation and the screened specific network by the partial least squares method. We can conclude that the robustness of the specific network screened by network entropy is better than that of the specific network with the determined relationship by comparing the obtained specific network with the determined relationship. Finally, we can determine that the node CDH1 has the highest score in the specific network through a sensitivity score calculated by network entropy implying the gene CDH1 is the most sensitive regulatory factor.ConclusionsIt is clearly of great potential value to reconstruct and visualize gene regulatory networks according to gene databases for life activities. Here, we present an available algorithm to achieve the network reconstruction by measuring the network entropy and identifying the vital nodes in the specific nodes. The results indicate that inhibiting or enhancing the expression of CDH1 can maximize the inhibition or enhancement of the yeast cell cycle. Although our algorithm is simple, it is also the first step in deciphering the profound mystery of gene regulation.

Toward Concurrent Identification of Human Activities with a Single Unifying Neural Network Classification: First Step.

The characterization of human behavior in real-world contexts is critical for developing a comprehensive model of human health. Recent technological advancements have enabled wearables and sensors to passively and unobtrusively record and presumably quantify human behavior. Better understanding human activities in unobtrusive and passive ways is an indispensable tool in understanding the relationship between behavioral determinants of health and diseases. Adult individuals (N = 60) emulated the behaviors of smoking, exercising, eating, and medication (pill) taking in a laboratory setting while equipped with smartwatches that captured accelerometer data. The collected data underwent expert annotation and was used to train a deep neural network integrating convolutional and long short-term memory architectures to effectively segment time series into discrete activities. An average macro-F1 score of at least 85.1 resulted from a rigorous leave-one-subject-out cross-validation procedure conducted across participants. The score indicates the method's high performance and potential for real-world applications, such as identifying health behaviors and informing strategies to influence health. Collectively, we demonstrated the potential of AI and its contributing role to healthcare during the early phases of diagnosis, prognosis, and/or intervention. From predictive analytics to personalized treatment plans, AI has the potential to assist healthcare professionals in making informed decisions, leading to more efficient and tailored patient care.

Impact of COVID-19 pandemic on the utilization and quality of antibiotic use in the primary care setting in England, March 2019–March 2023: a segmented interrupted time series analysis of over 53 million individuals

ABSTRACT Background Amid the COVID-19 pandemic, we evaluated the short-term impact of COVID-19 on antibiotic use in primary care in England, focusing on both antibiotic quantity (overuse) and quality (misuse) of use. Research design and methods A population-based segmented interrupted analysis was applied on monthly dispensed antibiotics prescriptions using the Prescription Cost Analysis dataset (March/2019–March/2023). The quantity was assessed using number of items dispensed per 1000 inhabitants (NTI) and defined daily doses per 1000 inhabitants per day (DID), while quality was evaluated using WHO’s Access Watch Reserve (AWaRe) classification, the proportion of ‘4C’ antibiotics and the percentage of broad- to narrow-spectrum antibiotics. Results Findings indicate 8.6 (17.2%) and 0.4 (2.6%) increase in the NTI and DID, respectively, with a statistically significant uptick in trend noted after the second lockdown (β5) for ‘total antibiotics’ for NTI only (β5 = 1.6; 95% CI:0.17, 3.1). Quality assessment showed an increase in ‘Access’ antibiotics from 77% in March/2019 to 86% in March/2023; however, COVID-19 had no significant impact on WHO AWaRe classes. Conclusion COVID-19’s impact on antibiotic use quality and quantity appeared to be minimal, though an increase in utilization post-second lockdown coincided with healthcare system recovery. This suggests a nuanced impact of the pandemic, highlighting the importance of continued antimicrobial stewardship.

Dynamic Real-Time Production Forecasting Model for Complex Subsurface Flow Systems with Variable Length Input Sequences

Summary Production time-series forecasting for newly drilled wells or those with limited flow and pressure historical data poses a significant challenge, and this problem is exacerbated by the complexities and uncertainties encountered in fractured subsurface systems. While many existing models rely on static features for prediction, the production data progressively offer more informative insights as production unfolds. Leveraging ongoing production data can enhance forecasting accuracy over time. However, effectively integrating the production stream data presents significant model training and updating complexities. We propose two innovative methods to address this challenge: masked recurrent alignment (MRA) and masked encoding decoding (MED). These methods enable the model to continually update its predictions based on historical data. In addition, by incorporating sequence padding and masking, our model can handle inputs of varying lengths without trimming, thereby avoiding the potential loss of valuable training samples. We implement these models with gated recurrent unit (GRU) and evaluate their performance in a case study involving 6,154 shale gas wells in the Central Montney Region. The data set encompasses 39 production-related features, including reservoir properties, completion, and wellhead information. Performance evaluation is based on root mean square error (RMSE) to predict 36-month production from 200 wells during testing. Empirical findings highlight the efficacy of the proposed models in handling challenges associated with variable-length input sequences, showcasing their superior performance. Our research emphasizes the value of including shorter time-series segments, often overlooked, to improve predictive accuracy, especially in scenarios with limited training samples.

Temporally aligned segmentation and clustering (TASC) framework for behavior time series analysis

Behavior exhibits a complex spatiotemporal structure consisting of discrete sub-behaviors, or motifs. Continuous behavior data requires segmentation and clustering to reveal these embedded motifs. The popularity of automatic behavior quantification is growing, but existing solutions are often tailored to specific needs and are not designed for the time scale and precision required in many experimental and clinical settings. Here we propose a generalized framework with an iterative approach to refine both segmentation and clustering. Temporally aligned segmentation and clustering (TASC) uses temporal linear alignment to compute distances between and align the recurring behavior motifs in a multidimensional time series, enabling precise segmentation and clustering. We introduce an alternating-step process: evaluation of temporal neighbors against current cluster centroids using linear alignment, alternating with selecting the best non-overlapping segments and their subsequent re-clustering. The framework is evaluated on semi-synthetic and real-world experimental and clinical data, demonstrating enhanced segmentation and clustering, offering a better foundation for consequent research. The framework may be used to extend existing tools in the field of behavior research and may be applied to other domains requiring high precision of time series segmentation.