Spatio-temporal Clustering Algorithm Research Articles

An Angle Recognition Algorithm for Tracking Moving Targets Using WiFi Signals with Adaptive Spatiotemporal Clustering.

An angle estimation algorithm for tracking indoor moving targets with WiFi is proposed. First, phase calibration and static path elimination are proposed and performed on the collected channel state information signals from different antennas. Then, the angle of arrival information is obtained with the joint estimation algorithm of the angle of arrival (AOA) and time of flight (TOF). To deal with the multipath effects, we adopt the DBscan spatiotemporal clustering algorithm with adaptive parameters. In addition, the time-continuous angle of arrival information is obtained by interpolating and supplementing points to extract the dynamic signal paths better. Finally, the least-squares method is used for linear fitting to obtain the final angle information of a moving target. Experiments are conducted with the tracking data set presented with Tsinghua’s Widar 2.0. The results show that the average angle estimation error with the proposed algorithm is smaller than Widar2.0. The average angle error is about 7.18° in the classroom environment, 3.62° in the corridor environment, and 12.16° in the office environment; they are smaller than the errors of the existing system.

Roads constrain movement across behavioural processes in a partially migratory ungulate

BackgroundHuman disturbance alters animal movement globally and infrastructure, such as roads, can act as physical barriers that impact behaviour across multiple spatial scales. In ungulates, roads can particularly hamper key ecological processes such as dispersal and migration, which ensure functional connectivity among populations, and may be particularly important for population performance in highly human-dominated landscapes. The impact of roads on some aspects of ungulate behaviour has already been studied. However, potential differences in response to roads during migration, dispersal and home range movements have never been evaluated. Addressing these issues is particularly important to assess the resistance of European landscapes to the range of wildlife movement processes, and to evaluate how animals adjust to anthropogenic constraints.MethodsWe analysed 95 GPS trajectories from 6 populations of European roe deer (Capreolus capreolus) across the Alps and central Europe. We investigated how roe deer movements were affected by landscape characteristics, including roads, and we evaluated potential differences in road avoidance among resident, migratory and dispersing animals (hereafter, movement modes). First, using Net Squared Displacement and a spatio-temporal clustering algorithm, we classified individuals as residents, migrants or dispersers. We then identified the start and end dates of the migration and dispersal trajectories, and retained only the GPS locations that fell between those dates (i.e., during transience). Finally, we used the resulting trajectories to perform an integrated step selection analysis.ResultsWe found that roe deer moved through more forested areas during the day and visited less forested areas at night. They also minimised elevation gains and losses along their movement trajectories. Road crossings were strongly avoided at all times of day, but when they occurred, they were more likely to occur during longer steps and in more forested areas. Road avoidance did not vary among movement modes and, during dispersal and migration, it remained high and consistent with that expressed during home range movements.ConclusionsRoads can represent a major constraint to movement across modes and populations, potentially limiting functional connectivity at multiple ecological scales. In particular, they can affect migrating individuals that track seasonal resources, and dispersing animals searching for novel ranges.

An individual-based spatio-temporal travel demand mining method and its application in improving rebalancing for free-floating bike-sharing system

After the rapid expansion in the early stage, many enterprises have closed down or withdrawn from the Free-floating bike sharing (FFBS) market, the remaining few giants are also generally at a loss at present. One main reason causing these is the serious FFBS imbalance between supply and demand. As there is no fixed docking station, the individual-based identification method such as trip-chain, which is used for identifying travel demand and behaviour in traditional station-based bike sharing (SBBS) cannot be used in FFBS. Therefore, the lack of methods to obtain in-depth demand makes it unable to achieve reasonable rebalancing. This study constructs an individual-based spatio-temporal travel demand mining methodology, which is the first disaggregate travel demand mining model that suitable for FFBS. The proposed methodology consists of three steps. A spatio-temporal trajectory clustering algorithm is first developed to obtain an individual’s frequent trajectory clusters, and then a sequential pattern mining algorithm is applied to users who have multiple spatio-temporal trajectory clusters to extract travel patterns and trajectory sequential relations in patterns. A point clustering method is used finally to identify spatial relationships among different trajectory clusters. Besides, a zone aggregating method is proposed that aggregated granularity could be flexible adjusted for zone demand imbalance analysis. Based on these, how to utilize identified frequent pattern trajectories to improve rebalancing is studied. The proposed methodology is applied to Beijing Mobike dataset, six frequent travel patterns are mined out and analyzed in detail. On this basis, imbalance and rebalancing analysis are carried out with the case study at last. Consequently, this research contributes a powerful tool to achieve accurate FFBS demand analysis and rebalancing.

Vehicle travel path recognition in urban dense road network environments by using mobile phone data

ABSTRACT Vehicle travel paths provide basic information for improving traffic forecasting models, tracking epidemics transmission, and road construction. Nevertheless, the challenge of recognition and verification still exists, especially in urban dense road networks. This paper proposes a vehicle path recognition model combined with mobile phone data. In path fitting module, the spatio-temporal density-based clustering algorithm and Gaussian filter were combined to smooth the position fluctuations of mobile phone data; then non-uniform rational B-splines were used to fit travel paths. In path recognition module, the modified probabilistic map matching algorithm was used to match fitting knots to road networks; then matching results were repaired considering the road network topology and the direction angles. The results were verified from trip lengths, urban environments, and road categories. The recognition accuracy was around 90%, 24.22% higher than that of existing methods. The error rate was around 6%, 30.28% lower than that of existing methods.

DP-GMM clustering-based ensemble learning prediction methodology for dam deformation considering spatiotemporal differentiation

The deformation behavior can effectively reflect the health status of a dam, which can suffer from dynamic and time-varying loading modes and material properties, exhibiting distinct spatiotemporal differentiation. The conventional dam health evaluations, however, focus on the displacement regularity of local monitoring points, ignoring the spatiotemporal diversity in deformation behavior, and thus cannot truly reflect the status of the entire dam. Moreover, most of the existing displacement predictions are conducted on a single-algorithm machine-learning-based model, which focuses on single-task regression and disregards the potential cross-relatedness among the deformation in adjacent areas of the dam, showing less competitiveness in complex input–output relationship learning for various spatiotemporal conditions. For more reliable deformation forecasting, an integrated displacement prediction methodology considering spatiotemporal differentiation is proposed based on the spatiotemporal clustering and machine learning algorithms, in which the minimum-density-entropy-optimized density peaks clustering method is embedded with Gaussian-mixture-model-based clustering to achieve the reliable spatiotemporal differentiation identification of dam behavior and a multioutput ensemble learning framework coupled with the extreme learning machine and support vector machine is designed to capture the complex mapping from environmental factors to deformation, considering the cross-relatedness and spatiotemporal diversity. Moreover, a synchronous optimization strategy based on improved grey wolf optimization is embedded to enhance the performance. Hence, the final deformation behavior is forecasted by the ensemble learning of various data features and mapping rules. Taking the displacement observations of an actual dam as an example, the results show that the proposed methodology achieves excellent performance in dam behavior forecasting.

Basic reproduction number of African swine fever in wild boars (Sus scrofa) and its spatiotemporal heterogeneity in South Korea.

BackgroundAfrican swine fever (ASF) is a hemorrhagic fever occurring in wild boars (Sus scrofa) and domestic pigs. The epidemic situation of ASF in South Korean wild boars has increased the risk of ASF in domestic pig farms. Although basic reproduction number (R0) can be applied for control policies, it is challenging to estimate the R0 for ASF in wild boars due to surveillance bias, lack of wild boar population data, and the effect of ASF-positive wild boar carcass on disease dynamics.ObjectivesThis study was undertaken to estimate the R0 of ASF in wild boars in South Korea, and subsequently analyze the spatiotemporal heterogeneity.MethodsWe detected the local transmission clusters using the spatiotemporal clustering algorithm, which was modified to incorporate the effect of ASF-positive wild boar carcass. With the assumption of exponential growth, R0 was estimated for each cluster. The temporal change of the estimates and its association with the habitat suitability of wild boar were analyzed.ResultsTotally, 22 local transmission clusters were detected, showing seasonal patterns occurring in winter and spring. Mean value of R0 of each cluster was 1.54. The estimates showed a temporal increasing trend and positive association with habitat suitability of wild boar.ConclusionsThe disease dynamics among wild boars seems to have worsened over time. Thus, in areas with a high elevation and suitable for wild boars, practical methods need to be contrived to ratify the control policies for wild boars.

Research on Human Travel Correlation for Urban Transport Planning Based on Multisource Data.

With the rapid development of positioning techniques, a large amount of human travel trajectory data is collected. These datasets have become an effective data resource for obtaining urban traffic patterns. However, many traffic analyses are only based on a single dataset. It is difficult to determine whether a single-dataset-based result can meet the requirement of urban transport planning. In response to this problem, we attempted to obtain traffic patterns and population distributions from the perspective of multisource traffic data using license plate recognition (LPR) data and cellular signaling (CS) data. Based on the two kinds of datasets, identification methods of residents’ travel stay point are proposed. For LPR data, it was identified based on different vehicle speed thresholds at different times. For CS data, a spatiotemporal clustering algorithm based on time allocation was proposed to recognize it. We then used the correlation coefficient r and the significance test p-values to analyze the correlations between the CS and LPR data in terms of the population distribution and traffic patterns. We studied two real-world datasets from five working days of human mobility data and found that they were significantly correlated for the stay and move population distributions. Then, the analysis scale was refined to hour level. We also found that they still maintain a significant correlation. Finally, the origin–destination (OD) matrices between traffic analysis zones (TAZs) were obtained. Except for a few TAZs with poor correlations due to the fewer LPR records, the correlations of the other TAZs remained high. It showed that the population distribution and traffic patterns computed by the two datasets were fairly similar. Our research provides a method to improve the analysis of complex travel patterns and behaviors and provides opportunities for travel demand modeling and urban transport planning. The findings can also help decision-makers understand urban human mobility and can serve as a guide for urban management and transport planning.

Spatiotemporal imaging phenotypes of tau pathology in Alzheimer’s disease

AbstractBackgroundThe Braak staging scheme describes a stereotypical spread of tau pathology in Alzheimer’s disease (AD). However, apparent subtypes described at autopsy and clinical variants of AD both suggest variability in the pattern of tau spreading across the population. To address this discrepancy, we applied a validated spatiotemporal clustering algorithm to a large, multisite dataset of tau‐PET images. We verified these findings both longitudinally and on a separate tau‐PET dataset using a different radiotracer, and we characterized the phenotype and potential biological determinants of each subtype.MethodSubtype and Stage Inference (SuStaIn, Young et al., 2018 ) is a data‐driven, probabilistic algorithm that combines disease progression modeling with clustering to extract distinct spatiotemporal patterns (subtypes) from large cross‐sectional datasets. We applied SuStaIn to a multisite dataset (n=1143) of [18F]‐flortaucipir tau‐PET baseline scans to identify spatiotemporal subtypes of tau spreading. We investigated differences in demographic, cognitive and genetic measures. In a subsample of individuals with multiple tau‐PET scans over time (n=536), we tested longitudinal subtype stability and change in disease stage over time. We verified these subtypes in a second single‐site dataset (n=509) of [18F]‐RO948 tau‐PET scans. Finally we used the Epidemic Spreading Model (ESM, Iturria‐Medina et al., 2014) and imaging or transcriptomic analysis to determine if anatomical brain networks or specific cell‐types, respectively, could explain the differing spatial patterns across subtypes.ResultIn those subjects with significant tau‐PET signal in the brain, SuStaIn identified four spatiotemporal subtypes of tau‐PET spreading: Typical (worse memory, more APOE4 carriers), limbic‐sparing (younger, less APOE4 carriers), posterior (worse visuospatial cognition), and lateral‐temporal (worse language and executive function scores; Fig 1A, Fig 2). These same four subtypes were observed in a separate cohort using a different radiotracer (similarity 0.7–0.9; Fig 1B). 86% of subjects exhibited the same subtype at multiple visits (Fig 1C), and of that group, 83% advanced in disease stage or remained stable within model uncertainty. The ESM suggested different epicenters for each subtype, and each subtype pattern exhibited a different profile of cell‐type enrichment (Fig 3).ConclusionFour stable tau‐spreading phenotypes were observed with distinct cognitive profiles, which may be characterized by vulnerability of different corticolimbic networks involving distinct cell classes.

Detection of Indoor High-Density Crowds via Wi-Fi Tracking Data.

Accurate detection of locations of indoor high-density crowds is crucial for early warning and emergency rescue during indoor safety accidents. The spatial structure of indoor environments is more complicated than outdoor environments. The locations of indoor high-density crowds are more likely to be the sites of security accidents. Existing detection methods for high-density crowd locations mostly focus on outdoor environments, and relatively few detection methods exist for indoor environments. This study proposes a novel detection framework for high-density indoor crowd locations termed IndoorSRC (Simplification–Reconstruction–Cluster). In this paper, a novel indoor spatiotemporal clustering algorithm called Indoor-STAGNES is proposed to detect the indoor trajectory stay points to simplify indoor movement trajectory. Then, we propose use of a Kalman filter algorithm to reconstruct the indoor trajectory and properly align and resample the data. Finally, an indoor spatiotemporal density clustering algorithm called Indoor-STOPTICS is proposed to detect the locations of high-density crowds in the indoor environment from the reconstructed trajectory. Extensive experiments were conducted using indoor Wi-Fi positioning datasets collected from a shopping mall. The results show that the IndoorSRC framework evidently outperforms the existing baseline method in terms of detection performance.

Incremental Spatio Temporal Clustering Application on Hotspot Data

Hotspot is one of indications for forest and land fires. Analysis of hotspot data needs to be done as an early warning activity to prevent the occurrence of forest and land fires. The previous study analyzed hotspot clusters using the incremental spatio-temporal density-based clustering (ST-DBSCAN) algorithm. Hotspots in a cluster are considered as strong indicator for forest and land fires. However, clustering of hotspots is implemented on the command line interface. Through this interface, users are required to execute manually the commands to perform the clustering task on the dataset. The disadvantage of command line interface is that the commands have to be typed precisely and mistypes will cause errors.Therefore, this study aims to build a web-based application for clustering hotspot data using the incremental ST-DBSCAN module. This study uses hotspot dataset in the period 2014 to 2017. The application was developed using R programming language and Shiny framework. The method of Adaptive Software Development (ASD) was adopted in this study. The main functions of the application are incremental clustering on hotspot data and visualization of clustering result. The testing using the black box approach show that all features of the application work properly. Based on the result of usability assessment, the user satisfaction level reach 78.45% meaning that the application is quite easy to be used.

Exploring the homogeneity of theft offenders in spatio-temporal crime hotspots

Offender homogeneity occurs when the same criminal group is composed of offenders with similar attributes (e.g., socio-economic-demographics). Exploring the homogeneity of offenders within spatio-temporal crime hotspots (STCHs) is useful for understanding not only the generational mechanisms of crime hotspots, but also has crime prevention implications. However, the homogeneity of offenders within STCHs has not been explored in criminological studies hitherto. Indeed, current techniques of STCH detection are limited to using statistical clustering methods in existing studies that lack the ability to identify the shape of STCHs or the distribution and variety of offences/offender activity with them. In this study, we utilise a spatio-temporal clustering algorithm called ST-DBSCAN to determine STCHs. We then propose novel entropy-based indices that measure the similarity of offenders (and offences) within STCHs. The method is demonstrated using theft crime records in the central area of Beijing, China. The results show that theft in the city is concentrated in a narrow space and time span (STCHs) and that within these associated offenders with similar social demographics, referred to as homogeneous offender groups are detectable.

Extracting and Predicting Taxi Hotspots in Spatiotemporal Dimensions Using Conditional Generative Adversarial Neural Networks

It is of practical importance to extract and predict taxi hotspots in urban traffic networks. However, the extraction of taxi hotspot is generally influenced by multiple sources of dependence, which has not been well recognized in the existing literature. This study aims to investigate how the integration of clustering models and deep learning approaches can learn and extract the network-wide taxi hotspots in both temporal and spatial dimensions. A density based spatiotemporal clustering algorithm with noise (DBSTCAN) was established to extract the historical taxi hotspots, which changed with time. A conditional generative adversarial network with long short-term memory structure (LSTM-CGAN) model was proposed for taxi hotspot prediction, which is capable of capturing the spatial and temporal variations of taxi hotspots simultaneously. Specifically, the DBSTCAN was applied to process the large-scaled geo-coded taxi pickup data into time-varying historical hotspot information. The proposed LSTM-CGAN model was then trained by the network-wide hotspot data. As illustrated in the numerical tests, it was found that the proposed LSTM-CGAN model provided comparable results with different model layouts and model with 4 LSTM layers in both generator and discriminator performed best. Moreover, four typical prediction approaches were selected to compare with the proposed LSTM-CGAN model. The comparative analyses showed that the proposed LSTM-CGAN model reduced the false positive rate from 5.20% to 0.56% and the false negative rate from 63.50% to 39.93%, and improved the section consistency from 41.78% to 73.43% and the area under the curve from 0.670 to 0.969. The comparison results indicated that the proposed LSTM-CGAN model outperformed all these benchmark methods and demonstrated great potential to enable many shared mobility applications.

Identifying trip ends from raw GPS data with a hybrid spatio-temporal clustering algorithm and random forest model: a case study in Shanghai

ABSTRACTSmartphones have been advocated as the preferred devices for travel behavior studies over conventional surveys. But the primary challenges are candidate stops extraction from GPS data and trip ends distinction from noise. This paper develops a Resident Travel Survey System (RTSS) for GPS data collection and travel diary verification, and then uses a two-step method to identify trip ends. In the first step, a density-based spatio-temporal clustering algorithm is proposed to extract candidate stops from trajectories. In the second step, a random forest model is applied to distinguish trip ends from mode transfer points. Results show that the clustering algorithm achieves a precision of 96.2%, a recall of 99.6%, mean absolute error of time within 3 min, and average offset distance within 30 meters. The comprehensive accuracy of trip ends identification is 99.2%. The two-step method performs well in trip ends identification and promotes the efficiency of travel survey systems.

An integrated feature learning approach using deep learning for travel time prediction

Travel time data is a vital factor for numbers of performance measures in transportation systems. Travel time prediction is both a challenging and interesting problem in ITS, because of the underlying traffic and events’ hidden patterns. In this study, we propose a multi-step deep-learning-based algorithm for predicting travel time. Our algorithm starts with data pre-processing. Then, the data is augmented by incorporating external datasets. Moreover, extensive feature learning and engineering such as spatiotemporal feature analysis, feature extraction, and clustering algorithms is applied to improve the feature space. Furthermore, for representing features we used a deep stacked autoencoder with dropout layer as regularizer. Finally, a deep multi-layer perceptron is trained to predict travel times. For testing our predictive accuracy, we used a 5-fold cross validation to test the generalization of our predictive model. As we observed, the performance of the proposed algorithm is on average 4 min better than applying the deep neural network to the initial feature space. Furthermore, we have noticed that representation learning using stacked autoencoders makes our learner robust to overfitting. Moreover, our algorithm is capable of capturing the general dynamics of the traffic, however further works need to be done for some rare events which impact travel time prediction significantly.

Spatiotemporal clustering: a review

An increase in the size of data repositories of spatiotemporal data has opened up new challenges in the fields of spatiotemporal data analysis and data mining. Foremost among them is “spatiotemporal clustering,” a subfield of data mining that is increasingly becoming popular because of its applications in wide-ranging areas such as engineering, surveillance, transportation, environmental and seismology studies, and mobile data analysis. This review paper presents a comprehensive review of spatiotemporal clustering approaches and their applications as well as a brief tutorial on the taxonomy of data types in the spatiotemporal domain and patterns. Additionally, the data pre-processing techniques, access methods, cluster validation, space–time scan statistics, software tools, and datasets used by various spatiotemporal clustering algorithms are highlighted.

A Framework for Camera-Based Real-Time Lane and Road Surface Marking Detection and Recognition

We propose a novel real-time integrated unsupervised learning framework for lane detection, tracking, and road surface marking detection and recognition using live feed from a camera mounted on the dashboard of a moving vehicle. Accurate and automated lane and road surface marking detection and recognition from cameras play an important role in intelligent vehicles. It helps increase driving safety in advanced driver assistant systems and enhance global localization and path planning in autonomous driving vehicles. Map building and updation can also benefit from automated detection of lanes and road surface markings. In our framework, we propose a spatio-temporal incremental clustering algorithm coupled with curve-fitting for simultaneously detecting lanes and road surface markings. The clusters that do not represent the lanes are further processed using Grassmann manifold learning to accurately recognize the road surface markings. Our framework is independent of the number and type of lanes, be it continuous, non-continuous, straight, or curved. Moreover, the framework works for all types, sizes, and orientations of road surface markings. It is only limited by the type of road surface markings present in the training data. We have measured the performance of our framework on state-of-the-art datasets.

CorClustST—Correlation-based clustering of big spatio-temporal datasets

Increasing amounts of high-velocity spatio-temporal data reinforce the need for clustering algorithms which are effective for big data processing and data reduction. As currently applied spatio-temporal clustering algorithms have certain drawbacks regarding the comparability of the results, we propose an alternative spatio-temporal clustering technique which is based on empirical spatial correlations over time. As a key feature, CorClustST makes it easily possible to compare and interpret clustering results for different scenarios such as multiple underlying variables or varying time frames. In a test case, we show that the clustering strategy successfully identifies increasing spatial correlations of wind power forecast errors in Europe for longer forecast horizons. An extension of the clustering algorithm is finally presented which allows for a large-scale parallel implementation and helps to circumvent memory limitations. The proposed method will especially be helpful for researchers who aim to preprocess big spatio-temporal datasets and who intend to compare clustering results and spatial dependencies for different scenarios.

Spatio-Temporal Association Query Algorithm for Massive Video Surveillance Data in Smart Campus

When using traditional methods to query massive video surveillance data in intelligent campus, there are some problems such as unstable query and inefficient query. Therefore, a query algorithm for intelligent campus video surveillance data based on spatio-temporal correlation is proposed. In this paper, a spatio-temporal association query algorithm for massive video surveillance data in smart campus was proposed. First, the spatio-temporal data clustering algorithm was introduced to cluster the massive data of surveillance video in smart campus. Then, HBase was used as the overall query structure of spatio-temporal association query algorithm based on the clustering results. Through combining the spatio-temporal features and attributed characteristics, a hierarchical record table was generated to construct the spatio-temporal attribute index of queries. According to the index of attribute columns, we can query massive data in many cases. The query condition was determined by Z curve, and the spatio-temporal association query of massive video surveillance data in smart campus was realized. Experimental results showed that when the number of data node was 5, the execution time of the algorithm of this paper was only 1200 s, which was much shorter than the other traditional algorithms. It was proved that the algorithm can maintain spatio-temporal index, improve query efficiency, and enhance query stability.

Mixed Fuzzy Clustering for Misaligned Time Series

Data mining in medical databases often involves the comparison of time series, which represent the evolution of a physiological variable. Temporal misalignment of physiological variables can conceal the discovery of patterns and trends shared between different patients. To address this problem, this paper proposes the mixed fuzzy clustering (MFC) algorithm with the dynamic time-warping (DTW) distance. We developed the MFC algorithm by 1) incorporating the DTW distance into the standard fuzzy c-means to handle misaligned time series; 2) introducing a new dimension into the spatiotemporal clustering algorithm to handle P time-variant features; and 3) incorporating unsupervised learning of cluster-dependent attribute weights. The algorithm is designed to simultaneously cluster time-variant and time-invariant data. We demonstrate the advantages of the proposed algorithm in four synthetic datasets and in two real-world applications in intensive care units. The first application is the classification of patients who will need the administration of vasopressors, and the second is the classification of patients with a high risk of mortality. Time-variant features consist of physiological variables collected with different sampling rates at different points in time. Time-invariant features consist of patients’ demographics and score records. The performance is evaluated using cluster validity measures, showing that the proposed algorithm outperforms fuzzy c-means.

Network-based study of Lagrangian transport and mixing

Abstract. Transport and mixing processes in fluid flows are crucially influenced by coherent structures and the characterization of these Lagrangian objects is a topic of intense current research. While established mathematical approaches such as variational methods or transfer-operator-based schemes require full knowledge of the flow field or at least high-resolution trajectory data, this information may not be available in applications. Recently, different computational methods have been proposed to identify coherent behavior in flows directly from Lagrangian trajectory data, that is, numerical or measured time series of particle positions in a fluid flow. In this context, spatio-temporal clustering algorithms have been proven to be very effective for the extraction of coherent sets from sparse and possibly incomplete trajectory data. Inspired by these recent approaches, we consider an unweighted, undirected network, where Lagrangian particle trajectories serve as network nodes. A link is established between two nodes if the respective trajectories come close to each other at least once in the course of time. Classical graph concepts are then employed to analyze the resulting network. In particular, local network measures such as the node degree, the average degree of neighboring nodes, and the clustering coefficient serve as indicators of highly mixing regions, whereas spectral graph partitioning schemes allow us to extract coherent sets. The proposed methodology is very fast to run and we demonstrate its applicability in two geophysical flows – the Bickley jet as well as the Antarctic stratospheric polar vortex.