Search

The development of visual tools for the timely identification of spatio-temporal clusters will assist in implementing control measures to prevent further damage. From January 2015 to June 2020, a total number of 1463 avian influenza outbreak farms were detected in Taiwan and further confirmed to be affected by highly pathogenic avian influenza subtype H5Nx. In this study, we adopted two common concepts of spatio-temporal clustering methods, the Knox test and scan statistics, with visual tools to explore the dynamic changes of clustering patterns. Since most (68.6%) of the outbreak farms were detected in 2015, only the data from 2015 was used in this study. The first two-stage algorithm performs the Knox test, which established a threshold of 7 days and identified 11 major clusters in the six counties of southwestern Taiwan, followed by the standard deviational ellipse (SDE) method implemented on each cluster to reveal the transmission direction. The second algorithm applies scan likelihood ratio statistics followed by AGC index to visualize the dynamic changes of the local aggregation pattern of disease clusters at the regional level. Compared to the one-stage aggregation approach, Knox-based and AGC mapping were more sensitive in small-scale spatio-temporal clustering.

Dengue fever is a well-known vector-borne disease caused by Aedes aegypti mosquito. It has become a major burden to economy and society of affected country. In Malaysia, dengue incidence in Selangor has been worsening and alarming. The aim of this study is two-fold; to examine the trend and behaviour of dengue incidence across time and districts in Selangor and to cluster the endemic areas in Selangor using Wards hierarchical clustering method. The spatial and temporal analysis found that the dengue incidence is worsening in the early and middle of the year. The Wards minimum variance method was able to cluster Selangor's endemic area into high endemic areas (Gombak, Hulu Langat, Klang and Petaling), medium endemic area (Sepang) and low endemic areas (Hulu Selangor, Kuala Langat, Kuala Selangor, SabakBernam). The findings of the study are significant to respective local authorities in providing information for monitoring and planning the early dengue warning systems. This is important to reduce the dengue incidence in hot spot areas and to safeguard the community from dengue outbreak.

International audience; ovine theileriosis (caused by Theileria parva) is the most important tick-borne transboundary animal disease endemic to Zimbabwe, yet its distribution dynamics data in the country remain scant and outdated. A retrospective study was conducted to determine high-risk areas of bovine theileriosis and associated risk factors in Zimbabwe. Records on bovine theileriosis spanning 23 years (January 1995 to December 2018) were obtained from the Epidemiological Unit of the Division of Field Veterinary Services of Zimbabwe (DVSZ). Data were analysed using Studio R® version 11.0 for regression analysis and SatScan® version 9.4.6 for spatio-temporal clustering. Communal farmers (72%), adult cattle (29%), the year 2018 (60%) and the hot wet season (42%) had the highest proportion (p

Sheep and goat pox is highly contagious viral infection of sheep and goats caused by the genus Capripox virus. Clinically, the disease is characterized by fever, macules developing into papules, and necrotic lesions in the skin and nodular lesions in internal organs. In Ethiopia, there are seroprevalence epidemiological studies on the disease. However, the spatiotemporal clustering of sheep and goat pox incidence has not been investigated. A retrospective study design using the outbreak reported data from Kombolcha Regional Laboratory for the years from September 2013 to December 2019 was performed to determine the temporal and spatial distribution of sheep and goat pox outbreaks. A total of 663 sheep and goat pox disease outbreaks were reported in all major parts of Eastern Amhara region between 2013 and 2019. In this period, sheep and goat pox was reported in all administrative zones of Eastern Amhara region (n = 5). The average incidence of sheep and goat pox outbreaks at the district level was 8.61 per 7 years. The incidence differed between areas, being the lowest in hot dry month and highest in warm and cold moist months. Sheep and goat pox outbreaks generally have a peak in November followed by August and a low in May. There is a significant difference in the occurrence of sheep and goat pox disease outbreaks between months and years ( p < 0.001 ). The forecast for the period 2020–2026 revealed that a high number of sheep and goat pox disease outbreaks will occur than the previous one. Therefore, all stakeholders should work cooperatively to combat this disease occurrence, and there should be capacity development for participatory disease search, risk analysis, laboratory diagnosis, and information management in order to respond properly to outbreak of sheep and goat pox disease; thereby, it enhances the prevention and control the disease.

The development of smart factories has put forward more flexible logistics needs for automobile assembly system, and efficient scheduling strategies to meet these requirements still demand prompt solution. Thus, this paper focuses on the problem of materials distribution with automated guided vehicles (AGVs) in automobile assembly lines. The mathematical model is established in the light of actual situation with mixed time windows and an improved genetic algorithm (GA) is developed. Considering the demand characteristics both in time and space, material demand points are clustered based on their spatiotemporal distance to generate the initial population. Then, selection, crossover and mutation operators of GA are also ameliorated as necessary to minimize the total travel cost. Finally, practical examples are carried out to demonstrate the effectiveness of this methodology.

Extended from the spatially defined nearest neighbor index, the nearest neighbor index measures the levels of spatiotemporal clustering of a set of points, using only their spatial locations and th...

Understanding the spatial and temporal dynamics of active wildfires is critical for supporting policies and legislation related to wildfire management. The study tested the extent to which active w...

Background To investigate the spatial distribution of congenital syphilis (CS) and its association to social vulnerability indexes in northeast Brazil. Methods This was an ecological study referring to all cases of CS and CS deaths recorded in the northeast region of Brazil from 2008 to 2015. Data were obtained from three Brazilian information systems. We examined statistical correlations between CS indicators by state and municipality and their socioeconomic and social vulnerability characteristics. We used Bayesian empirical local models to identify fluctuations of the indicators. Spatial statistical tests were used to identify spatial clusters and the municipalities at high risk of CS. Results The incidence of CS ranged from 2.1 cases/1000 live births (LB) in 2008 to 6.9/1000 LB in 2015, with an annual increase of 19.9% (p Conclusions CS incidence is associated with social vulnerability. CS control programmes should target spatial clusters and populations with high levels of social vulnerability.

Traffic accidents are a common problem in any transportation network. Road traffic accidents are predicted to be the seventh leading cause of deaths by the year 2030. Recently research in the integration of geographical information systems (GIS) for analyzing accidents, road design and safety management has increased considerably. The perpetual use of GIS tools, lead this study to propose the identification of accident hotspots by exploiting GIS technology coupled with kernel density estimation (KDE). This paper proposes the use of KDE technique and GIS technology to automatically identify the accident hotspots using UK as the study area. Analysis shows that most of the accidents occur when there is a 30 mph speed limit, a weekend, in the evening time, during the months of October and November, on the single carriageway, where there is ‘T’ or staggered junction and on ‘A’ road class. Moreover, this study also proposed techniques to classify the accident severity that is classified as either fatal, serious or slight. The driver behavior and environmental features achieved an accuracy up to 85% on the severity classification with Bagging technique. Further, the shortcomings, limitations and recommendations for future work are also identified.

The objective of this paper is to identify hot spots of bus bunching events at the network level, both in time and space, using Automatic Vehicle Location (AVL) data from the Athens (Greece) Public...

Objective:To determine whether Clostridioides difficile infection (CDI) exhibits spatiotemporal interaction and clustering.Design:Retrospective observational study.Setting:The University of Iowa Hospitals and Clinics.Patients:This study included 1,963 CDI cases, January 2005 through December 2011.Methods:We extracted location and time information for each case and ran the Knox, Mantel, and mean and maximum component size tests for time thresholds (T = 7, 14, and 21 days) and distance thresholds (D = 2, 3, 4, and 5 units; 1 unit = 5–6 m). All tests were implemented using Monte Carlo simulations, and random CDI cases were constructed by randomly permuting times of CDI cases 20,000 times. As a counterfactual, we repeated all tests on 790 aspiration pneumonia cases because aspiration pneumonia is a complication without environmental factors.Results:Results from the Knox test and mean component size test rejected the null hypothesis of no spatiotemporal interaction (P < .0001), for all values of T and D. Results from the Mantel test also rejected the hypothesis of no spatiotemporal interaction (P < .0003). The same tests showed no such effects for aspiration pneumonia. Our results from the maximum component size tests showed similar trends, but they were not consistently significant, possibly because CDI outbreaks attributable to the environment were relatively small.Conclusion:Our results clearly show spatiotemporal interaction and clustering among CDI cases and none whatsoever for aspiration pneumonia cases. These results strongly suggest that environmental factors play a role in the onset of some CDI cases. However, our results are not inconsistent with the possibility that many genetically unrelated CDI cases occurred during the study period.

The Maunder butterfly pattern is the most complete spatial-temporal representation of observed changes in solar activity in the 11-year cycle over a period of 12–24 cycle. The well-known empirical relation is used to transform Greenwich sunspot areas into magnetic flux, and the distances between the butterfly wings are then calculated with the Fisher–Rao metric. We found that the similarities or differences in the patterns of the individual butterfly wings in this metric are approximately the same for each hemisphere. The wings were the closest for a sequence of strong cycles, while there is a tendency for a series of weak cycles to form cycles in pairs with the implementation of an analog of the Gnevyshev–Olya rule.

At least five surface rupturing earthquakes that occurred during a

Many fluorescence super-resolution techniques, such as (d)STORM, PALM, and DNA-PAINT, generate datasets wherein multiple localizations across many camera frames may arise from a single blinking event of an emitter. These repeated localizations not only hinder interpretation and analysis of such datasets, but also represent an incomplete use of the fluorescence photons. Such localizations are typically combined into a single localization either by clustering with hard distance and time thresholds, or by classical hypothesis testing assuming Gaussian localization errors. In this work, we describe a method for clustering that accounts for localization precision, local emitter density estimates, and a kinetic model for blinking which is used to optimize connections within a group of spatiotemporally colocated localizations.

Background African 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. Objectives This study was undertaken to estimate the R0 of ASF in wild boars in South Korea, and subsequently analyze the spatiotemporal heterogeneity. Methods We 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. Results Totally, 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. Conclusions The 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.

Smartphones 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...

Using an open-access spatiotemporal analytics program, we mapped spatiotemporal heterogeneity loci in tuberculosis (TB) cases (clusters) and dynamic changes, and characterized the drug-resistant TB clustering risk using routine microbiological data from KwaZulu-Natal, South Africa. The data may provide insight into transmission dynamics and support efficient deployment of public health resources.

The widespread acceptance of context-sensing applications is generating voluminous movement data on high speed, which has fueled research studies in mining of spatio-temporal trajectories. The anal...

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.

Considering the critical role of trajectory data in Big Data era for dynamic geographical processes, human behaviour analysis and meteorological prediction, trajectory clustering has attracted growing attention. Many literatures have discussed the spatiot emporal clustering method of simple trajectories (i.e., has no branches, e.g. vehicle trajectories), yet there are few researches for clustering complex trajectories (i.e., has at least one split and/or merger and/or split -merger branch, e.g. ocean eddy trajectories, rainstorm trajectories). For addressing this issue, we propose a Process-Oriented Spatiotemporal Clustering Method (POSCM) for clustering complex trajectory data. The POSCM includes three parts: the first uses the semantic of process-sequence-state to represent the complex trajectories; the second proposes a Hierarchical Similarity Measurement Method (HSMM) to get the similarity between any two complex trajectories; in the last step, the complex trajectories clustering pattern is extracted through density-based clustering algorithm. Experiments on simulated trajectories are used to evaluate the POSCM and demonstrate the advantage by comparing against that of the VF2 algorithm. The POSCM is applied to the sea surface temperature abnormal variations trajectories from January 1950 to December 2017 in the Pacific Ocean. As shown in this case study, some new mined spatiotemporal patterns can provide new references for understanding the behaviours of marine abnormal variations under the background of the global change.

Smart card datasets in the public transit network provide opportunities to analyse the behaviour of passengers as individuals or as groups. Studying passenger behaviour in both spatial and temporal space is important because it helps to find the pattern of mobility in the network. Also, clustering passengers based on their trips regarding both spatial and temporal similarity measures can improve group-based transit services such as Demand-Responsive Transit (DRT). Clustering passengers based on their trips can be carried out by different methods, which are investigated in this paper. This paper sheds light on differences between sequential and combined spatial and temporal clustering alternatives in the public transit network. Firstly, the spatial and temporal similarity measures between passengers are defined. Secondly, the passengers are clustered using a hierarchical agglomerative algorithm by three different methods including sequential two-step spatial-temporal (S-T), sequential two-step temporal-spatial (T-S), and combined one-step spatiotemporal (ST) clustering. Thirdly, the characteristics of the resultant clusters are described and compared using maps, numerical and statistical values, cross correlation techniques, and temporal density plots. Furthermore, some passengers are selected to show how differently the three methods put the passengers in groups. Four days of smart card data comprising 80,000 passengers in Brisbane, Australia, are selected to compare these methods. The analyses show that while the sequential methods (S-T and T-S) discover more diverse spatial and temporal patterns in the network, the ST method entails more robust groups (higher spatial and temporal similarity values inside the groups).

Nowadays, Twitter has become one of the fastest-growing microblogging services; consequently, analyzing this rich and continuously user-generated content can reveal unprecedentedly valuable knowledge. In this paper, we propose a novel two-stage system to detect and track events from tweets by integrating a Latent Dirichlet Allocation (LDA)-based approach and an efficient density–contour-based spatio-temporal clustering approach. In the proposed system, we first divide the geotagged tweet stream into temporal time windows; next, events are identified as topics in tweets using an LDA-based topic discovery step; then, each tweet is assigned an event label; next, a density–contour-based spatio-temporal clustering approach is employed to identify spatio-temporal event clusters. In our approach, topic continuity is established by calculating KL-divergences between topics and spatio-temporal continuity is established by a family of newly formulated spatial cluster distance functions. Moreover, the proposed density–contour clustering approach considers two types of densities: “absolute” density and “relative” density to identify event clusters where either there is a high density of event tweets or there is a high percentage of event tweets. We evaluate our approach using real-world data collected from Twitter, and the experimental results show that the proposed system can not only detect and track events effectively but also discover interesting patterns from geotagged tweets.

Using fishery-dependent observer data from National Marine Fisheries (NMFS) provides insight into the location and intensity of bottom-trawl fishing effort, and allows those areas most exposed to fishing pressure to be identified. In this study, the spatial and temporal extent of Alaskan bottom-trawl fishing effort in the Bering Sea, Aleutian Islands and Gulf of Alaska between 1993 and 2015 is explored within a space-time cube in ArcGIS Pro. The variables analysed were number of hauls per area and total catch per area. Statistical techniques were used to examine spatiotemporal clustering within the data. Results indicate that fishing was significantly clustered over space and time. A three-dimensional hotspot analysis shows which areas were most intensely fished and illustrates the trends over the relatively long study period. The data were then compared with sea ice concentration to determine the effect of changing climate on fishing activity. Sea ice had a limited effect on the spatial patterns of fishing effort, but certain areas in the Bering Sea exhibited increased fishing effort in years with less sea ice.

The categorization of multidimensional data into clusters is a common task in statistics. Many applications of clustering, including the majority of tasks in ecology, use data that is inherently spatial and is often also temporal. However, spatiotemporal dependence is typically ignored when clustering multivariate data. We present a finite mixture model for spatial and spatiotemporal clustering that incorporates spatial and spatiotemporal autocorrelation by including appropriate Gaussian processes (GP) into a model for the mixing proportions. We also allow for flexible and semiparametric dependence on environmental covariates, once again using GPs. We propose to use Bayesian inference through three tiers of approximate methods: a Laplace approximation that allows efficient analysis of large datasets, and both partial and full Markov chain Monte Carlo (MCMC) approaches that improve accuracy at the cost of increased computational time. Comparison of the methods shows that the Laplace approximation is a useful alternative to the MCMC methods. A decadal analysis of 253 species of teleost fish from 854 samples collected along the biodiverse northwestern continental shelf of Australia between 1986 and 1997 shows the added clarity provided by accounting for spatial autocorrelation. For these data, the temporal dependence is comparatively small, which is an important finding given the changing human pressures over this time.

Natural hazards are inherently spatio-temporal processes. Spatio-temporal clustering methodologies applied to natural hazard data can help distinguish clustering patterns that would not only identify point-event dense regions and time periods, but also provide insight into the hazardous process. Here we review spatio-temporal clustering methodologies applicable to point event datasets representative of natural hazards and we evaluate their performance using both synthetic and real life data. We first present a systematic overview of major spatio-temporal clustering methodologies used in the literature, which include clustering procedures that are (i) global (providing a single quantitative measure of the degree of clustering in the dataset) and (ii) local (i.e. assigning individual point events to a cluster). A total of seven methodologies from these two groups of clustering procedures are applied to real-world (lightning) and synthetic datasets. For (i) global procedures, we explore Knox, Mantel, Jacquez k-NN tests and spatio-temporal K-functions and for (ii) local procedures we consider spatio-temporal scan statistic, kernel density estimation and density-based clustering method OPTICS. The dataset of 7021 lightning strikes is from 1 and 2 July 2015 over the UK, when a severe three-storm system crossed the region with different convective modes producing each of the storms. The synthetic datasets are representative of various topologies of a point-event natural hazard data with a moving source. We introduce a two-source model with input parameters related to the physical properties of the source. Each source has a set number of points events, initiation point in space and time, movement speed, direction, inter-event time distribution and spatial spread distribution. In addition to a base model of two identical moving sources with a set temporal separation, we produce four different topologies of the data by incrementally varying the speed parameter of the source, spatial spread parameters, direction and initiation points, and angle of two sources. With these five synthetic datasets representative of various two-source models, we evaluate the performance of the methodologies. The performance is assessed based on the ability of each methodology to separate the point events produced by the two sources and the sensitivity of these results to changes in the model input parameters. We further discuss the benefits of combining global and local clustering procedures in the analyses as we gain an initial understanding of the spatial and temporal scales over which clustering is present in the data by using global clustering procedures. This information then helps to inform and limit the choice of input parameters for the local clustering procedures.

Drought is considered one of the costliest climate extremes that have wide impacts on humans and ecosystems. Understanding different drought stages, for example, onset, propagation, and its recovery, especially for tropical (the vulnerable region in Earth’s climate system) catchments are crucial for ecosystem sustainability and food security. Utilizing gauge-based quality-controlled daily streamflow records from 98 catchments of rain-fed Peninsular River Basins (PRB) in India, here we investigate different phases of hydrological droughts in a multi-stage framework. While several studies so far have investigated the propagation of hydrological droughts at a monthly resolution, a credible understanding of drought dynamics requires analyzing low-flow series at a higher spatial and temporal resolution, ensuring the issuance of timely alerts related to regional water scarcity. Owing to high seasonality in the daily streamflow records, a variable threshold approach is adopted to delineate streamflow-based drought events. To assess the temporal evolution of droughts, the events are categorized into various inter-related phases, i.e., growth, persistence, and recovery stage over the study period 1965 – 2018. For most of the gauges, the mean timing of drought onset mostly lies between August and September revealing failure of monsoon as the primary causal factor for drought development in peninsular catchments. Furthering this, we identify four distinct hydrological drought regimes, which includes, Regime 1: persistent droughts with longer duration and moderate deficit volume with average termination during mid-monsoon (in September). These gauges are mostly situated in Central India and typically show a longer recovery time coincided with shorter return times (i.e., the time between two consecutive drought events), making it one of the most vulnerable regions in PRB; Regime 2: droughts with a shorter duration, least deficit volume with average termination in October, the post-monsoon period. These gauges are located in the western part of the country; Regime 3: droughts with the highest variability in drought deficit volume with the largest subsurface contribution from groundwater recharge. These sites are primarily located in eastern India and do not show any specific trend in the termination period; Regime 4: droughts with least regularity in drought termination with the average termination month clustered around November. These gauges are mostly concentrated in the southwestern part of the country. Our findings add value to the systematic understanding of hydrological drought propagations in rain-fed catchments, which serves as a basis for exploring future changes in droughts under concurrent shifts in rainfall and temperature extremes in a warming climate.

Various spatio-temporal clustering methods have been proposed to detect groups of jointly moving objects in space and time. However, such spatio-temporal clustering methods are rarely compared against each other to evaluate their performance in discovering moving clusters. Hence, in this work, we present a spatio-temporal clustering benchmark for the field of collective animal behavior. Our reproducible benchmark proposes synthetic datasets with ground truth and scalable implementations of spatio-temporal clustering methods. The benchmark reveals that temporal extensions of standard clustering algorithms are inherently useful for the scalable detection of moving clusters in collective animal behavior. published

In this research, a new uncertainty clustering method has been developed and applied to the spatial time series with seasonality. The new unsupervised grouping method is based on Neighbourhoods and Median Seasonal Entropy. This classification method aims to discover similar behaviours for a time series group and find a dissimilarity measure concerning a reference series r. The Neighbourhood’s Internal Verification Coefficient criterion makes it possible to measure intra-group similarity. This clustering criterion is flexible for spatial information. Our empirical approach allows us to measure accommodation decisions for tourists who visit Spain and decide to stay either in hotels or in tourist apartments. The results show the existence of dynamic seasonal patterns of behaviour. These insights support the decisions of economic agents. This research is associated with the group of Faculty of Economic and Business Sciences at the University of Malaga: “Social Indicators-SEJ157”. The research group has funded the professional editing service in English. Research Funders: “Funding for open access charge: Universidad de Málaga/CBUA”.

In this paper, we address the problem of automatically extracting several clusters consisting of spatio-temporally similar earthquakes whose average magnitudes are substantially different from the total average. 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 our empirical evaluation using earthquake catalog data covering the whole of Japan, we show that the proposed method employing the single-link algorithm can produce more desirable results for our purpose in terms of the improvement of weighted sums of variances and visualization results.

Visual analytics combines advanced visualisation methods with intelligent analysis techniques in order to explore large data sets whose complexity, underlying structure and inherent dynamics are beyond what traditional visualisation techniques can handle. The ultimate goal is to expose relevant patterns and relationships from the data, since not everything can be exposed easily through intelligent analysis techniques. On the contrary, the human eye can outperform algorithms in grasping and interpreting subtle patterns, provided it is supported by intelligent visualisations. In this paper, we propose three novel visual analytics techniques for analysing spatio-temporal data. First, we present a fingerprinting technique for discovering and rapidly interpreting temporal and recurring patterns by use of circular heat maps. Next, we present a technique supporting comparisons in time or space by use of circular heat map subtraction. Finally, we propose a technique enabling to characterise and get insights of thetemporal behaviour of the phenomenon under study by use of label maps. The potential of the proposed approach to reveal interesting patterns is demonstrated in a case study using traffic data, originating from multiple inductive loops in the Brussels-Capital Region, Belgium.

Summary Many behaviors that are critical for survival and reproduction are expressed over extended time periods. The ability to inexpensively record and store large volumes of video data creates new opportunities to understand the biological basis of these behaviors and simultaneously creates a need for tools that can automatically quantify behaviors from large video datasets. Here, we demonstrate that 3D Residual Networks can be used to classify an array of complex behaviors in Lake Malawi cichlid fishes. We first apply pixel-based hidden Markov modeling combined with density-based spatiotemporal clustering to identify sand disturbance events. After this, a 3D ResNet, trained on 11,000 manually annotated video clips, accurately (>76%) classifies the sand disturbance events into 10 fish behavior categories, distinguishing between spitting, scooping, fin swipes, and spawning. Furthermore, animal intent can be determined from these clips, as spits and scoops performed during bower construction are classified independently from those during feeding., Graphical Abstract, Highlights • A dataset of more than 14,000 annotated animal behavior videos was created • 3D residual networks can be used to classify animal behavior • Different intents of similar behavioral actions can be distinguished • A working solution to study long-term behaviors was established, Piscine Behavior; Zoology; Computer Science; Artificial Intelligence

With the rise of Dockless Shared Bicycles in cities, the rational division of docks for Shared bikes has become an important issue in urban road planning. In the initial stage of Dockless Shared Bicycles, randomly parked bicycles bring great pressure to urban road traffic because there is no fixed parking area. The appearance of an electronic fence has improved the user's shared bicycles specification to a certain extent. However, due to the variability of urban road planning, it is very necessary to optimize the parking area of bicycles regularly. In this paper, the cycling data of Shared bicycles are studied by using the Spatio-temporal clustering algorithm, the flow and parking rules of dockless Shared bicycles are explored to seek the optimal division of bicycles areas.

Forest fire is a condition where an area where there are many trees and plants undergo a change in shape caused by massive burning. Forest fires occur in Indonesia especially on the island of Sulawesi. Sulawesi Island itself is an island that still has quite extensive forests. The cause of forest fires is caused by two factors, namely natural and human. Indicators of forest fires can be known from the appearance of hotspots or hot spots. A hotspot is an indicator of an area that has a relatively higher surface temperature than the surrounding surface temperature that has been detected by Terra / Aqua satellites with a MODIS sensor. Furthermore, the hotspots are not only randomly distributed but also clustered. The research related to the identification process of hotspot distribution patterns in an area in Sulawesi is still limited and has its novelty. Here we showed the implementation of Spatio-temporal clustering method with ST-DBSCAN algorithm, aided with Waterfall system development. The result showed that there are several types of distribution patterns generated by the system (Stationary, Regular Reaping, Irregular, Occasional, and Track) from the data used from 2016 to 2018 over Sulawesi island. In this research, we also found that between 2016 and 2018, the emergence of high tendency hotspots occurred in July, August, September, and October.

One of the document analyzer's essential functions is dividing input documents into stroke groups that belong to the particular type of handwriting. In this paper, we analyze existing approaches for clustering input strokes of the handwritten document and present a new one that establishes a novel state of the art result. It is the first approach where a grouping of input strokes is considered as a clustering problem and standard metrics for it are provided. This research could be used as a baseline for further ones in the clustering of handwritten documents.

Introduction: In Ecuador, the first COVID-19 case, the disease caused by the SARS-CoV-2 virus, was officially reported on February 29, 2020. As of April 2, the officially confirmed numbers of COVID-19 cases and deaths from it were 3 163 and 120, respectively, that is, a mortality rate of 3.8%. Objective: To identify spatial and spatiotemporal clusters of COVID-19 cases officially confirmed in Ecuador. Materials and methods: Case series study. An analysis of all COVID-19 cases officially confirmed in Ecuador from March 13, 2020 to April 2, 2020 was performed. Relative Risk (RR) of COVID-19 contagion was determined using the discrete Poisson distribution model in the SaTScan software. Clusters were generated using purely spatial and spatiotemporal scan statistics. Significance of each cluster was obtained through 999 iterations using the Monte Carlo simulation, obtaining the most probable random model. Results: As of April 2, spatiotemporal clustering allowed identifying two clusters in Ecuador, a main cluster in the Guayas province (area: 15 430 km2; population: 3.6 million inhabitants; RR: 7.08; p

Universal vaccination coverage is still far from desired targets in many global regions including Pakistan, despite the success stories and its scientifically proven benefits. EPI Pakistan vaccination coverage data 2012–2016, at district level was collected from Federal EPI Pakistan. District-wise population data were collected from Pakistan Bureau of Statistics. Descriptive statistics and sequence plots were performed in SPSS 13.0. Purely spatial scanning analysis was done in SaTScan 9.4.4 using discrete Poisson model for detection of low vaccination coverage clusters. Geographical information system (GIS) was used to display spatial patterns and clusters of low vaccination coverage districts in Pakistan. Average annual EPI vaccination coverage in each study year were; 70.98 in 2012, 69.39% in 2013, 66.74% in 2014, 61.47% in 2015, and 67.01% in 2016, respectively. Cumulative average national vaccination rate (2012–2016) for all types of EPI vaccines was 60.60%. Average national vaccination rate for BCG, OPV3, pentavalent3 and measles1 was 67.12%, 58.53%, 58.47%, and 58.29%, respectively. Spatial cluster analysis demonstrated that most of low coverage districts for BCG, OPV3 and pentavalent3 were from FATA and KPK; while measles1 low coverage districts belonged to Balochistan. Future research should probe factors involved in low vaccination coverage in high risk districts.

Breast cancer is the first cause of cancer-related death in Portuguese women. This study aimed to characterize female breast cancer mortality in Portugal in the period between 2002 and 2013, with a special focus on spatiotemporal patterns.The breast cancer mortality rate was studied using descriptive analysis (unadjusted and age-adjusted), and spatiotemporal clustering analyses.In 2002 - 2013 the breast cancer mortality rate was 28.47/100 000 inhabitants and the age-adjusted mortality rate was 19.46/100 000 inhabitants. In this period the Lisbon region (urban), Alentejo and Algarve (rural) presented higher breast cancer mortality rate, but Madeira (urban), Lisbon and Algarve had higher age-adjusted mortality rate. In the spatiotemporal analysis, the overall mortality rate showed an increasing trend of 1.218%/year, without spatial variations. Also, different patterns were detected in the50, 50 - 64 and ≥ 65 age-groups (+ 0.725%, - 1.781% and + 0.896%, respectively). One temporal (2004 - 2006) and one spatiotemporal cluster (North coast) presented significantly lower mortality rate than expected for the period and/or area (26.2 and 16.1/100 000 inhabitants, respectively). Conversely, two spatiotemporal clusters, located in the city of Lisbon (2002 - 2007) and in the Centre region (2008 - 2013), presented significantly higher breast cancer mortality rate than expected (48.6 and 34.9/100 000 inhabitants, relative risk: 1.74 and 1.26, respectively).The annual female crude and adjusted breast cancer mortality rate matched previous publications. However the annual increase detected in the unadjusted rate clashes with the published literature. Overall, the presence of spatiotemporal clusters supports the uneven distribution of the breast cancer mortality reported previously in the different Portuguese regions.This study identified areas and trends of the female breast cancer mortality rate, showing high spatiotemporal variations that must support further detailed studies/interventions.Introdução: O cancro da mama é a primeira causa de morte relacionada com cancro em mulheres portuguesas. Este estudo pretende caracterizar a mortalidade feminina por cancro da mama em Portugal no período de 2002 a 2013, com enfoque nos padrões espácio-temporais. Material e Métodos: A taxa de mortalidade por cancro da mama foi estudada com recurso a análise descritiva (bruta e ajustada para a idade), e análise de clustering espácio-temporal. Resultados: Em 2002 – 2013 a taxa de mortalidade por cancro da mama foi 28,47/100 000 habitantes e a taxa de mortalidade ajustada pela idade foi de 19,46/100 000 habitantes. Neste período a região de Lisboa (urbana), Alentejo e Algarve (rural) apresentaram taxas de mortalidade mais elevadas, mas após ajustamento pela idade a Madeira (urbana), Lisboa e Algarve demonstraram taxas de mortalidade superiores. Na análise espácio-temporal, a taxa de mortalidade geral apresentou um crescimento de 1,218%/ano, sem variações espaciais. Adicionalmente, padrões diferentes foram detetados nos grupos de mulheres com50, 50 - 64 e ≥ 65 anos (+ 0,725%, - 1,781% e + 0,896%, respetivamente). Um cluster temporal (2004 – 2006) e um espácio-temporal (costa Norte) apresentaram taxa de mortalidade significativamente mais baixas que o esperado para o período e/ou área (26,2 e16,1/100 000 habitantes, respetivamente). Por outro lado, dois clusters espácio-temporais, localizados na cidade de Lisboa (2002 – 2007) e na zona Centro (2008 – 2013), apresentaram taxas de mortalidade por cancro da mama superiores às expectáveis (48,6 e 34,9/100 000 habitantes, risco relativo: 1,74 e 1,26, respetivamente). Discussão: A taxa anual bruta e ajustada para a idade de mortalidade por cancro da mama aproximam-se das anteriormente publicadas. No entanto, o aumento anual nas taxas brutas contrasta com a literatura neste tópico. A presença de clusters espácio-temporais suporta a distribuição variável da taxa de mortalidade por cancro da mama nas diferentes regiões do país. Conclusão: Este estudo identificou áreas e tendências na taxa de mortalidade feminina por cancro da mama, demonstrando variações espácio-temporais nesta taxa que suportam estudos e intervenções mais detalhadas nesta área.

Spatio-temporal clustering of seismicity features is an interesting phenomenon that is relevant for earthquake generation process and operational earthquake forecasting. We analyze successive earthquakes that closely occur in space and time in order to clarify how large earthquakes successively occur. We use the Global Centroid Moment Tensor catalog for the period from 1976 to 2016. Shallow earthquakes with a moment magnitude, Mw , of larger than or equal to 5.0 are analyzed. We first sort all of the earthquakes in time to select a master event from the beginning. Then, we group the earthquakes that occur within a horizontal distance ( D ) and a lapse time ( Ta ) from the master event into a cluster. Next master event is selected from the catalog in order, and the same procedure is repeated. We count the number of the clusters, which represent the successive earthquakes, for different D and Ta To examine whether or not successive earthquakes randomly occur, we compare the results with simulations in which earthquakes are set to occur randomly in time but at the locations same with the estimated centroid. The results show that the cumulative numbers of clusters for the simulation more rapidly increase with the horizontal distance than those for real data at short distance ranges, and the formers approach to the latter at long distance range. The triggering distance, at which the cumulative numbers of real and simulation data merge, increases with increasing the magnitude of master event. The triggering distance becomes smaller as the lapse time increases, which implies that the seismic activity turns to become the normal condition in which the occurrence time intervals of large earthquakes obey a Poisson distribution. The triggering distance increases with being almost proportional to the 1/3 of the seismic moment of master earthquake, and the number of earthquakes occurring in the region with positive Coulomb stress change (ΔCFF) are more than 60-80% of the total number of the successive earthquakes. These results suggest that static stress change introduced by a master event is one of the triggering mechanism of successive earthquakes.

Identifying an ozone pollution zone during the pollution processes is significant for ozone pollution management and environmental health risk assessment. However, few studies have focused on ozone pollution zone identification during pollution processes. A spatial-temporal clustering framework for identifying pollution zones during ozone pollution processes was initially proposed in this study, and an ozone pollution process in China in May 2017 was selected as a case. The results showed that the framework can help selecting one more accurate method to identify the pollution zone according to the pollution characteristics of air pollution process. In addition, different ozone pollution zone identification methods work well in different scenarios: The self-organizing map (SOM) method was suitable for identifying the zone with the duration of pollution between 24 and 48 h, the image fusion based on wavelet transform (IFbWT) method for the zone with the duration of pollution over 48 h and the Apriori method for the zone with obvious boundaries between high-value and low-value ozone concentrations. The proposed procedure can also be applied to identify the pollution zone of the pollution process of other pollutants.

&ldquo, Strike Hard&rdquo, is an enhanced law-enforcement strategy in China that aims to suppress crime, but measurement of the crime-reducing effect and potential changes in the spatiotemporal concentration of crime associated with &ldquo, remain unknown. This paper seeks to examine the impact, if any, of &ldquo, on the spatiotemporal clustering of burglary incidents. Two and half years of residential burglary incidents from Chaoyang, Beijing are used to examine repeat and near-repeat burglary incidents before, during, and after the &ldquo, intervention and a new technique that enables the comparison of repeat and near repeat patterns across different temporal periods is introduced to achieve this. The results demonstrate the intervention disrupted the repeat pattern during the &ldquo, period reducing the observed ratio of single-day repeat burglaries by 155%, however, these same single-day repeat burglary events increased by 41% after the cessation of the intervention. Findings with respect to near repeats are less remarkable with nominal evidence to support that the intervention produced a significant decrease, but coupled with other results, suggest that spatiotemporal displacement may have been an undesired by-product of &ldquo, This study from a non-Western setting provides further evidence of the generalizability of findings related to repeat and near repeat patterns of burglary and further highlights the limited preventative effect that the &ldquo, enhanced law enforcement campaign had on burglary.

Low-frequency and regular earthquakes have similar clustering properties; SSEs and earthquakes follow similar scaling relations., Strain accumulated on the deep extension of some faults is episodically released during transient slow-slip events, which can subsequently load the shallow seismogenic region. At the San Andreas fault, the characteristics of slow-slip events are difficult to constrain geodetically due to their small deformation signal. Slow-slip events (SSEs) are often accompanied by coincident tremor bursts composed of many low-frequency earthquakes. Here, we probabilistically estimate the spatiotemporal clustering properties of low-frequency earthquakes detected along the central San Andreas fault. We find that tremor bursts follow a power-law spatial and temporal decay similar to earthquake aftershock sequences. The low-frequency earthquake clusters reveal that the underlying slow-slip events have two modes of rupture velocity. Compared to regular earthquakes, these slow-slip events have smaller stress drop and rupture velocity but follow similar magnitude-frequency, moment-area, and moment-duration scaling. Our results connect a broad spectrum of transient fault slip that spans several orders of magnitude in rupture velocity.

Classical Swine Fever (CSF) is an endemic disease in Cuba, and an eradication strategy by zones is planned by the Official Veterinary Service. The aim of this study was to identify high-risk areas of CSF and the risk factors associated with the disease occurrence in the Pinar del Rio province, one of the prioritized areas in the eradication strategy. The outbreak occurrence at district level was analyzed through a 7-year period (2009 to 2015). A high-risk cluster (RR = 5.13, 95% CI 3.49-7.56) was detected during the last 2 years of the study period in the eastern half of the province, with 38 out of 97 districts included. The rate of CSF-affected holdings had a significant increase during 2014-2015 and seems to have occurred mainly in the high-risk cluster area. Swine population density by district (heads/km(2)) and road length (km) by district were associated with the disease outbreak occurrence. These results provide new insights into the knowledge of the epidemiology of the disease in Cuban endemic conditions and can contribute to improving the control and the eradication strategy in this situation.

Public buses are an important part of the urban transportation mix. However, a considerable disadvantage of buses is their slow speed, which is in part due to frequent stops, but also due to the lack of segregation from other vehicles in traffic. As such, assessing bus routes and the respective sections that are prone to congestion is an important aspect of route planning, scheduling, and the creation of dedicated bus lanes. In this work we use bus tracking data from the Washington Metropolitan Area Transit Authority, to discover speed patterns of specific bus routes in relation to the road network throughout the day. Specifically, we focus on using these patterns to identify free flow segments, bus stop locations, traffic light locations and road segments prone to congestion.

The timely analysis of deformation monitoring data and reasonable diagnosis of the structural health are key tasks in dam health monitoring studies. This article presents a spatio-temporal clustering and health diagnosis method for super-high concrete arch dams that uses deformation monitoring data obtained from plumb meters. The spatio-temporal expression of the deformation monitoring data is proposed first by upgrading a punctuated time series to a curved panel time series, including cross-sectional, dam axial, and temporal changing directions. Second, a comprehensive similarity indicator on three aspects, namely, the absolute distance, incremental distance, and growth rate distance, is constructed after a deep discussion on deformation similarity characteristics both temporally and spatially. Next, the temporal clustering method is proposed by keeping the key features, namely, extreme points and turning points, while eliminating extraneous details, namely, noise points. Finally, the optimal spatio-temporal clustering of dam deformation is achieved by designing a multi-scale fuzzy C-means method of data mining and its iterative algorithm. The proposed method is applied to the Jinping-I hydraulic structure, which is the highest concrete arch dam in the world. The clustering results is quite sensitive in different weight coefficients of the comprehensive similarity indicator and clustering numbers of fuzzy C-means method. The dam deformation behaviors on high-water-level, water-falling, and low-water-level periods are analyzed and diagnosed. The advanced version of proposed methods is verified by comparative analysis on dam health diagnosis results obtained from ordinary deformation distribution figures and the spatio-temporal clustering figures. The proposed method will facilitate the recognition of abnormal deformation areas and associated safety diagnoses.

This paper proposes a novel method for dynamically extracting and monitoring the entrances of areas of interest (AOIs). Most AOIs in China, such as buildings and communities, are enclosed by walls and are only accessible via one or more entrances. The entrances are not marked on most maps for route planning and navigation in an accurate way. In this work, the extraction scheme of the entrances is based on taxi trajectory data with a 30 s sampling time interval. After fine-grained data cleaning, the position accuracy of the drop-off points extracted from taxi trajectory data is guaranteed. Next, the location of the entrances is extracted, combining the density-based spatial clustering of applications with noise (DBSCAN) with the boundary of the AOI under the constraint of the road network. Based on the above processing, the dynamic update scheme of the entrance is designed. First, a time series analysis is conducted using the clusters of drop-off points within the adjacent AOI, and then, a relative heat index ( R H I ) is applied to detect the recent access status (closed or open) of the entrances. The results show the average accuracy of the current extraction algorithm is improved by 24.3% over the K-means algorithm, and the R H I can reduce the limitation of map symbols in describing the access status. The proposed scheme can, therefore, help optimize the dynamic visualization of the entry symbols in mobile navigation maps, and facilitate human travel behavior and way-finding, which is of great help to sustainable urban development.