Identifying Urban Traveling Hotspots Using an Interaction-Based Spatio-Temporal Data Field and Trajectory Data: a Case Study Within the Sixth Ring Road of Beijing

sustainability

Article Identifying Urban Traveling Hotspots Using an Interaction-Based Spatio-Temporal Data Field and Trajectory Data: A Case Study within the Sixth Ring Road of Beijing

Disheng Yi 1,2,3,4, Yusi Liu 1,2,3,4 , Jiahui Qin 1,2,3,4 and Jing Zhang 1,2,3,4,*

1 College of Resources Environment and Tourism, Capital Normal University, Beijing 100048, China; [email protected] (D.Y.); [email protected] (Y.L.); [email protected] (J.Q.) 2 Beijing Laboratory of Water Resources Security, Capital Normal University, Beijing 100048, China 3 3D Information Collection and Application Key Lab of Education Ministry, Capital Normal University, Beijing 100048, China 4 Beijing State Key Laboratory Incubation Base of Urban Environmental Processes and Digital Simulation, Capital Normal University, Beijing 100048, China * Correspondence: [email protected]

Received: 26 September 2020; Accepted: 16 November 2020; Published: 19 November 2020

Abstract: Exploring urban travelling hotspots has become a popular trend in geographic research in recent years. Their identification involved the idea of spatial autocorrelation and spatial clustering based on density in the previous research. However, there are some limitations to them, including the unremarkable results and the determination of various parameters. At the same time, none of them reflect the influences of their neighbors. Therefore, we used the concept of the data field and improved it with the impact of spatial interaction to solve those problems in this study. First of all, an interaction-based spatio-temporal data field identification for urban hotspots has been built. Then, the urban travelling hotspots of Beijing on weekdays and weekends are identified in six different periods. The detected hotspots are passed through qualitative and quantitative evaluations and compared with the other two methods. The results show that our method could discover more accurate hotspots than the other two methods. The spatio-temporal distributions of hotspots fit commuting activities, business activities, and nightlife activities on weekdays, and the hotspots discovered at weekends depict the entertainment activities of residents. Finally, we further discuss the spatial structures of urban hotspots in a particular period (09:00–12:00) as an example. It reflects the strong regularity of human travelling on weekdays, while human activities are more varied on weekends. Overall, this work has a certain theoretical and practical value for urban planning and traffic management.

Keywords: hotspots; spatio-temporal data ﬁeld; spatial interaction; urban travelling; trajectory data

1. Introduction Intra-city human activities have accelerated the urbanization process in recent years. There is a huge urban population due to rural-to-urban migrations, especially. Furthermore, rapid urban expansion causes imbalanced urban development, such as traﬃc congestion, resource shortage, and environmental degradation [1–3]. The emergence of these problems is closely related to human mobility in the city [2]. For example, a large number of commuting activities aggravate traﬃc problems and air pollution in a particular period. These phenomena lead to various issues and discussions about urban structure and urban sustainable development [1,3]. Human activities in the urban area, by its

Sustainability 2020, 12, 9662; doi:10.3390/su12229662 www.mdpi.com/journal/sustainability Sustainability 2020, 12, 9662 2 of 20 very nature, could be categorized as travelling activities. The occurrence of them would be varied alone but would be geographically or temporally regular. Therefore, there is a feasible way to study the characteristics and patterns of these travelling activities. It is critical for planners and managers to understand urban structures and sustainable development for the modern city. At the same time, the acquisition of geo big data became more convenient with the rapid development of location-based service (LBS) and information and communication technology (ICT) [4,5]. Big data such as taxi trajectories, mobile phone records, and social media data include geographic locations and the time when they appeared. Each datum records human activities from individuals, which track people and updates in real-time. These data are widely used to solve the problem of the human–environment relationship [5]. Compared with conventional questionnaires and remote sensing data, they not only could be captured easily but also play a vital role in social sensing as the sensor of individuals [3,5]. The advantages of those geo big data bring new opportunities to observe, quantify, and even predict the dynamic patterns of human activities and the urban environment from an individual to a society level. Most human travelling activities are driven by many reasons. No matter what those reasons are, these urban travels are also known as the spatio-temporal tracks that lead to a series of spatio- temporal phenomena. Exploring inherent patterns of these phenomena has become more and more popular in urban study. Previous research mainly focused on the traces of human travelling, travelling origin–destination (OD) flow, and the patterns of urban areas involved in travels. Specifically, the traces of human travelling included a series of travelling points that are responsible for the human daily mobility patterns at the individual and population levels—for example, travelling motifs [6–8]. There are also several studies uncovering and forecasting the spatio-temporal travelling routines through travelling traces [9–11]. At the same time, travelling OD flows as the expression of spatial interactions between two places also play an essential role in discovering the patterns of human travelling. Yao et al. caught hot spatial interactions with multiple flows using OD flow clustering which could reflect the patterns of human travelling [12]. Evaluating those flows with the mutual constricted relationship is a feasible way as well [13]. Besides, there is research using the volume of travelling flows and snapshots depicting human travelling patterns [14]. Furthermore, analyzing the distribution of human travels and the urban areas in which they appeared is also a research trend from recent years. Some of them explored the spatial distributions of travels for diverse social activities [15] and the geographical characteristics of urban travel demand [16]. Apart from that, Yang et al. evaluated the diverse popularity of places in the urban area by human travelling [17,18]. Other research focused on the spatio-temporal characteristics of urban travelling hotspots [2,19,20]. Generally, most of the studies working on the spatio-temporal patterns of urban travelling could be categorized as discussions about where the popular areas of these travels appeared—in other words, urban travelling hotspots. There are also many hotspots that we will study in the urban area, such as crime hotspots, pollution hotspots, and epidemic hotspots. However, searching for the locations and boundaries of those hotspots accurately is a common goal in the geographical view. One conventional method is spatial autocorrelation, such as local Moran’s I and Getis-Ord Gi*, which means using the quantitative statistic index to identify hot or cold areas at various scales [21,22]. Recently, the idea of the emerging hotspot (EHS) was introduced to detect changes in trends of space–time data [23]. Harris et al. combined EHS analysis to study the condition of forest loss [24]. Another popular method that identifies urban hotspots is density-based clustering. Some research detected urban hotspots of trajectory with density-based spatial clustering of applications with noise (DBSCAN) [25,26]. One discovered the spatio-temporal hotspots of pick-ups and drop-offs and their intersections with an improved DBSCAN method and analyzed their dynamic distributions [27]. Wang et al. improved the traditional DBSCAN algorithm to suit the spatial network space to detect areas with dense events [28]. Furthermore, kernel density estimation (KDE) played an important role in hotspot identification. For example, Zhang et al. explored and clustered the travelling patterns of residents in the urban space using KDE [29]. Yang et al. studied the spatio-temporal characteristics of urban travelling Sustainability 2020, 12, 9662 3 of 20 hotspots by an improved one [2]. Meanwhile, there is an extension in which some studies found Sustainability 2020, 12, x FOR PEER REVIEW 3 of 21 hotspot locations with KDE and its improved methods in the urban road network [30,31]. However, spatial[30,31 autocorrelation]. However, spatial would autocorrelation be bad at narrowingwould be bad the at hot narrowing area down the hot to a area limited down range to a limited with the mostrange events. with the Density-based most events. clusteringDensity-based and cluster kerneling density and kernel estimation density have estimation an obstacle, have an which obstacle is the, determinationwhich is the determination of critical parameters, of criticalincluding parameter thes, including number the of(1) number clusters of ( and1) clusters (2) cluster and ( centers2) cluster and (3)centers radius, and which (3) radius, would which cause would dynamic cause distribution dynamic distribution of hotspots of [19 hotspots]. At the [19 same]. At time,the same all oftime, them wouldall of consider them would the aggregating consider the high-density aggregating center high rather-density than center the influences rather than by surroundingthe influences objects. by surroundingFortunately, objects. the concept of the data field was proposed to solve this problem. It simulates the mutualFortunately, interactions the between concept particles of the data in thefield physical was proposed field to to depict solve this the relationshipproblem. It simulates between the data objectsmutual [19 interactions,32]. There betwe are previousen particles research in the studies physical which field to used depict this the idea relationship in data clustering between [data33,34 ], imageobjects segmentation [19,32]. There [35 are], and previous detecting research urban studies hotspots which [19 used,20,36 this]. Despite idea in thedata increasing clustering success[33,34], in identifyingimage segmentation hotspots with [35], a and data detecting field, the urban spatial hotspots interaction [19,20 in,36 the]. Despite geographical the increasing space would success not in be identifying hotspots with a data field, the spatial interaction in the geographical space would not be considered in mutual interactions, which is an important factor driven by human activities. As a result, considered in mutual interactions, which is an important factor driven by human activities. As a this study improves the spatio-temporal data field with the impact of long-range spatial interaction result, this study improves the spatio-temporal data field with the impact of long-range spatial to discover urban travelling hotspots. A case study employs the proposed method to identify urban interaction to discover urban travelling hotspots. A case study employs the proposed method to hotspots, and then the accuracy of results is evaluated. Finally, we analyze the spatio-temporal identify urban hotspots, and then the accuracy of results is evaluated. Finally, we analyze the spatio- characteristics of the spatial distribution of those hotspots and their interactions. Figure1 illustrates temporal characteristics of the spatial distribution of those hotspots and their interactions. Figure 1 theillustrate overalls framework the overall offramework this study. of this study.

FigureFigure 1. 1.The The overalloverall ﬂowchartflowchart of the the research research structure. structure.

TheThe remainder remainder of of this this article article is is structured structured asas follows.follows. Section Section 2 discussesdiscusses the the basic basic theory theory and and applicationsapplications of of the the data data ﬁeld. field. Section Section3 3introduces introduces thethe proposedproposed methods and and evaluations evaluations used used in in this this study.study. In In Section Section4, we4, we present present a casea case study study within within thethe SixthSixth RingRing Road in in Beijing. Beijing. Next, Next, we we discuss discuss somesome broader broader thinking thinking about about the the interactions interactions betweenbetween hotspots hotspots under under the the urban urban spatial spatial network network in in SectionSection5, before5, before concluding concluding and and pointing pointing out out directions directions forfor futurefuture workwork in Section Section 66..

2. Data2. Data Field Field Theory Theory and and Its Its Classical Classical ExpansionsExpansions ThereThere is is a a mutual mutual interaction between between individual individual particles, particles, described described as “the as concept “the concept of the field” of the field”in physical in physical space. space. Each Eachparticle particle is regarded is regarded as a field as source. a field source.Each field Each source field would source radiate would energy radiate energyand influence and influence others others around around it simultaneously it simultaneously;; its en itsergy energy value valueis considered is considered as field as potential. field potential. Li Li[ 32] developed this this concept concept for for abstract abstract data data objects objects and and their their distributions. distributions. According According to tohis his idea, idea, thethe mutual mutual interaction interaction between between data objects objects is is depicted depicted by by the the aid aid of a of data a data field field [20]. [Meanwhile,20]. Meanwhile, the theco conceptncept of of potential potential energy energy in the in the physical physical system system had hadbeen been brought brought to measure to measure the influence the influence of the of theinteraction interaction between between data data objects objects with with a asuitable suitable potential potential function function [33 [33,35,35]. As]. As a result, a result, the the potential potential valuevalue is anis an exact exact quantitative quantitative expression expression aboutabout how a field field source source (an (an object) object) influences influences its itsneighbors neighbors.. In general, the radiation distance of the field source determines the type of the field [32]. There are long-range fields and short-range fields in the data field. As the position in the long-range field Sustainability 2020, 12, 9662 4 of 20

SustainabilityIn general, 2020, the12, xradiation FOR PEER distanceREVIEW of the field source determines the type of the field [32]. There4 of are 21 long-range fields and short-range fields in the data field. As the position in the long-range field is far awayis far fromawaythe from field the source, field source, the potential the potential value doesvalue not does vanish. not vanish. The short-range The short- field,range however, field, however, can be representedcan be represented the character the character which iswhich the value is the that value dropped that dropped sharply sharply with an with increasing an increasing distance distance in the space.in the The space. potential The potentialfunctions of functions the short-range of the field short fit-range specifically field tofit quantify specifically these to characteristics quantify these of distancecharacteristics decay of with distance the Gaussian decay with kernel the function Gaussian [19 kernel,20,32 ,function33]. [19,20,32,33]. InsteadInstead ofof randomrandom microparticlesmicroparticles inin physicalphysical space,space, location-basedlocation-based datadata points,points, suchsuch asas trajectorytrajectory data,data, wouldwould bebe aggregatedaggregated inin aa particularparticular areaarea becausebecause ofof humanshumans andand thethe environment.environment. Therefore,Therefore, theythey havehave theirtheir ownown datadata fieldfield andand influenceinfluence each each other other due due to to the the nature nature of of them. them. ZhaoZhao etet al.al. followedfollowed thethe ideaidea ofof thethe datadata fieldfield andand proposedproposed aa trajectorytrajectory datadata fieldfield [19[19].]. InIn trajectorytrajectory data,data, eacheach pointpoint isis treatedtreated asas aa fieldfield sourcesource withwith mass.mass. ItsIts potentialpotential valuevalue dependsdepends onon thethe numbernumber ofof fieldfield sourcessources aroundaround itit andand thethe distancedistance betweenbetween them.them. However,However, taxitaxi trajectoriestrajectories areare thethe aidaid ofof distinctivedistinctive spatio-temporalspatio-temporal datadata withwith richrich timetime information.information. ThereThere isis aa temporaltemporal dimensionaldimensional expansionexpansion fromfrom thethe classicalclassical spatialspatial datadata fieldfield [36[36].]. Typically,Typically, the the coordinates coordinates of of two two dimensions, dimensionsx, and푥 andy, describe푦, describe the locationthe location where where the datathe data point point occurred. occurred. In a spatio-temporal In a spatio-temporal space, thespace, energy the radiatingenergy radiating from the fieldfromsource the field is controlled source is bycontrolled the distance by the from distance the source from and the thesource time and after the the time field after source the formedfield source [20,36 formed]. Figure [202 ,illustrates36]. Figure the2 illustrates theoretical the relationship theoretical relationship of two points of in two a spatial points spacein a spatial and in space a spatio-temporal and in a spatio space.-temporal space.

Figure 2. The conceptual expression of points in a spatial space (a) and in a spatio-temporal space (b). Figure 2. The conceptual expression of points in a spatial space (a) and in a spatio-temporal space (b). Point A and point B are represented as A (x1, y1) and B (x2, y2) in the two-dimensional space, while in Point A and point B are represented as A (푥 , 푦 ) and B (푥 , 푦 ) in the two-dimensional space, while the spatio-temporal space, point A and point1 B are1 quantified2 as2 A (x1, y1, t1) and B (x2, y2, t2). in the spatio-temporal space, point A and point B are quantified as A (푥1, 푦1, 푡1) and B (푥2, 푦2, 푡2). Added with the time parameter, the traditional trajectory data field was evolved into the spatio-temporalAdded with data the time field. parameter, Supposing the thattraditional the trajectories trajectory datasetdata fieldD was= evolvedp1, p2, ... into, pn theis spatio in the- spatio-temporaltemporal data field. data field, Supposing each pick-up that the or trajectories drop-off is dataset abstract, 퐷 as=p{i푝1=, 푝2x, …i, y, i푝,푛ti} , is where in thexi , y spatioi, ti is- separatelytemporal represented data field, to each the coordinatespick-up or x dropand -yoffand is the abstract time that, as the 푝푖 event= {푥푖 showed, 푦푖, 푡푖} , up. where The 푥 definition푖, 푦푖, 푡푖 is ofseparately the potential represented value of p toi is the expressed coordinates as the 푥 following and 푦 and Equations the time (1) that and the (2): event showed up. The definition of the potential value of 푝푖 is expressed as the following Equations (1) and (2): k Xn dij ( ) 푘 1 ( ) = 푑σ푖푗 ϕ i mj e−−( ) 1 (1) 푛j=1 × 휎 × ∆tij (1) 휑(푖) = ∑푗=1 푚푗 × 푒 × ∆푡푖푗

∆t∆ij푡푖푗−∆∆푡t푚푖푛min ∆푡 = − ∆tij =푖푗 ∆푡 −∆푡 ((2)2) ∆tmax푚푎푥 ∆t푚푖푛 − min wherewhere m푚j푗represents represents the the massmass ofof thethe datadata objectobject p푝j푗,, normallynormally equaledequaled toto 1;1; d푑ij푖푗 meansmeans thethe distancedistance betweenbetween thethe objectobjectp 푝i 푖and andp j;푝σ푗; is휎 theis the impact impact factor factor that that controls controls the rangethe range of interactions of interaction amongs among the objectsthe objects [33]; [k33is]; the푘 is distance the distance index, index, normally normally equaled equaled to 2; ∆ totij 2is; ∆ defined푡푖푗 is defined as the normalizationas the normalization of the intervalof the interval of two of data two points data point happened;s happened∆tmin; means∆푡푚푖푛 themeans minimal the minimal interval interval in the dataset;in the dataset∆tmax ;means ∆푡푚푎푥 themeans maximal the maximal interval interval in the dataset. in the dataset. TheThe impactimpact factor factor measures measure thes the range range within within which which a field a field source source can influencecan influence others others in the in data the field.data field. When When the impact the impact factor factorσ is greater, 휎 is greater, the interaction the interaction range R rangeis larger. 푅 is According larger. According to the principle to the 3 principle of Gaussian kernel function, the influences of the short-range field vanish at 휎 from the √2 field source. In previous studies, the value of 휎 equaled to around 0.3 [19,36], and entropy was used to evaluate the distributions of data points and select a suitable impact factor for the dataset [32]. Sustainability 2020, 12, 9662 5 of 20 of Gaussian kernel function, the influences of the short-range field vanish at 3 σ from the field source. √2 In previous studies, the value of σ equaled to around 0.3 [19,36], and entropy was used to evaluate the distributions of data points and select a suitable impact factor for the dataset [32].

3. Methodology Data field theory is a data-driven methodology simulating the form and distribution of a dataset that puts data points as the priority. As a result, there are several applications in many study fields, such as discovering urban hotspots [19,20,36] and data clustering [33,34]. In this section, we improve the previous methods for detecting hotspots based on a data field with spatial interaction and introduce evaluations of their accuracy.

3.1. Interaction-Based Spatio-Temporal Data Field (STDF) Identification for Urban Hotspots In this part, the framework of interaction-based spatio-temporal data field (STDF) identification for urban hotspots is introduced. There are three steps, including calculating the potential value for each trajectory point with spatial interaction, calculating the sum of potential value in the urban unit, and detecting urban hotspots by edge detection.

3.1.1. Calculation of Potential Value In geographical space, mutual influences between individual locations are reflected by spatial interaction and temporal activities [20]. However, spatial interaction includes two types, which are touchedSustainability interaction 2020, 12, x FOR and PEER untouched REVIEW interaction. For example, closer objects are more related to others6 of 21 around them and their relationships with distance2 decay are presented by untouched interactions, 푑푖푗 −( ) 1 3 which can also be regarded as short-range푚 interactions.× 푒 휎1 × More, 푑 importantly,≤ 휎 touched interactions caused 푗 ∆푡 푖푗 2 1 휑 (푖) = 푖푗 √ by human-driven activities would푗 also be a vital part1 of spatial1 interactions.3 They can be mentioned(5 in) 푚 × × , 푑 > 휎 long-range interactions since the places seem푗 to be푑 geographically푖푗 푖푗 irrelative.1 Figure3 shows how the 1+ ∆푡푖푗 √2 points from neighbors and couples affect{ one another.휎2 A conventional data field with or without a temporalwhere 푚푗 dimension is defined isas a the mathematical mass of data function point 푗 to and draw the short-range potential value interactions. of data point As is known푗 in the to pair all, 3 3 theof OD long-range flows when interaction 푑 < is휎 one and of the푑 most≥ important휎 , respectively variations; 푑 whenmeans discovering the distance spatio-temporal between data 푖푗 √2 1 푖푗 √2 1 푖푗 patterns in urban space, including detecting urban hotspots. Existing data field methods discovered objects (origin and destination) 푖 and 푗; 휎1 and 휎2 are the impact factors, according to the rule of urban hotspots by the lack of long-range interactions. Therefore, we added the influence of long-range minimal entropy; 휎1 and 휎2 are amounts equal to 0.5 and 0.01, respectively; ∆푡푖푗 is the interactionsnormalization into of thethe datainterval field of in data this point study. 푖 and 푗, calculated by Equation (2).

Figure 3. The sketch sketch-map-map of the data ﬁeldfield with spatial interaction.

FirstPoint ofs A, all, B a, and mathematic C are three function points needed in the tospace. be found The graduated to depict the circles long-range represent interaction the impact in theradiated space. from The data concept points. of Spatial a long-range interaction ﬁeld is with depicted a slow by distancethe graduated decay line can with be referred an arrow. to asThe a long-rangepotential value interaction.s at location Generally,s 1 and 2 the are potentialthe scalar function sum of their of pseudo neighbors gravity A and is oftenB, respectively, used to express while the potential long-range value ﬁeld at [ 32location]. In a 3 spatio-temporal is regarded as the space, overlying a long-range of its neighbor spatial interactionA and coupleF contains C. an

3.1.2. Quantifying the Potential Value of the Study Unit Identifying collective spatio-temporal patterns from geo big data might be a variable challenge caused by the granularity of the analysis unit [37,38]. Trajectory data points with potential values can determine the maximal details as the minimal analysis unit undoubtedly. However, they are not regular enough to extract urban travelling modes. It is more efficient that these data points are projected into a fitting study unit with the increasing size of the study object. As a result, a specific urban unit is introduced as the study unit in this work. The potential value of one study unit is represented as Equation (5), which is quantified to the total potential value of data points within the unit. 푛 휑(푔) = ∑ 휑(푖) (6) 푖=1 where 휑(푔) means the potential value of the unit 푔; 푛 means the number of the points projected in the unit.

3.1.3. Determining the Urban Hotspots by Edge Detection There are various methods to identify hot and cold spots, including clustering based on potential value and edge detection. Thresholding segmentation, introduced by Rosin, is a simple and effective method for finding the boundary of images, which is a bilevel thresholding algorithm based on a histogram plot [39]. It aims to find the location with a maximal distance from a straight line starting at the peak and ending at the first empty bin of the histogram. All values in the histogram plot would be divided by this location. This method is suitable for histograms with only one dominant value (a peak) and a lower end. A histogram ranked by the potential values of urban units gives a long-tailed distribution curve that fits the thresholding segmentation algorithm. Therefore, we employed this method to determine hotspots in the urban area. Sustainability 2020, 12, 9662 6 of 20

origin O = xo, yo, to and a destination D = xd, yd, td ; the potential value of the origin aﬀected by the destination from a long-range interaction can be expressed as follows:

1 1 ϕD(O) = mD (3) × dOD × ∆tOD 1 + σ where mD represents the mass of destination D; dOD means the distance between the origin O and the destination D; σ means the impact factor; and ∆tOD determines the normalization of the interval of the origin O and the destination D that happened as Equation (2). Second of all, the inﬂuence of the long-range interaction can be considered when the distance of a pair of origin and destination fell outside the distance of the short-range interaction R. Nonetheless, there are two common distance measures used for calculations in geographic studies, which are the Euclidean distance and the Manhattan distance. Compared with the Euclidean distance, the Manhattan distance would be suitable to characterize the geographic relationship between taxi trajectories, because their existences rely on the road network. In this study, the Manhattan distance is employed in the distance measurement of two data points. Combined with the spatial interaction, the potential value of the point pi in the spatio-temporal data ﬁeld is calculated as follows: Xn ϕ(i) = ϕj(i) (4) j=1

 d 2  ( ij )  σ 1 3  mj e− 1 , dij σ1  ∆tij √2 ϕj(i) =  × × ≤ (5)  m 1 1 , d > 3 σ  j dij ∆t ij √ 1  × 1+ × ij 2 σ2 where mj is deﬁned as the mass of data point j and the potential value of data point j in the pair of 3 3 OD ﬂows when dij < σ1 and dij σ1, respectively; dij means the distance between data objects √2 ≥ √2 (origin and destination) i and j; σ1 and σ2 are the impact factors, according to the rule of minimal entropy; σ1 and σ2 are amounts equal to 0.5 and 0.01, respectively; ∆tij is the normalization of the interval of data point i and j, calculated by Equation (2). Points A, B, and C are three points in the space. The graduated circles represent the impact radiated from data points. Spatial interaction is depicted by the graduated line with an arrow. The potential values at locations 1 and 2 are the scalar sum of their neighbors A and B, respectively, while the potential value at location 3 is regarded as the overlying of its neighbor A and couple C.

3.1.2. Quantifying the Potential Value of the Study Unit Identifying collective spatio-temporal patterns from geo big data might be a variable challenge caused by the granularity of the analysis unit [37,38]. Trajectory data points with potential values can determine the maximal details as the minimal analysis unit undoubtedly. However, they are not regular enough to extract urban travelling modes. It is more efficient that these data points are projected into a fitting study unit with the increasing size of the study object. As a result, a specific urban unit is introduced as the study unit in this work. The potential value of one study unit is represented as Equation (5), which is quantified to the total potential value of data points within the unit.

Xn ϕ(g) = ϕ(i) (6) i=1 where ϕ(g) means the potential value of the unit g; n means the number of the points projected in the unit.

3.1.3. Determining the Urban Hotspots by Edge Detection There are various methods to identify hot and cold spots, including clustering based on potential value and edge detection. Thresholding segmentation, introduced by Rosin, is a simple and effective method for finding the boundary of images, which is a bilevel thresholding algorithm based on a Sustainability 2020, 12, 9662 7 of 20 histogram plot [39]. It aims to find the location with a maximal distance from a straight line starting at the peak and ending at the first empty bin of the histogram. All values in the histogram plot would be divided by this location. This method is suitable for histograms with only one dominant value (a peak) and a lower end. A histogram ranked by the potential values of urban units gives a long-tailed distribution curve that fits the thresholding segmentation algorithm. Therefore, we employed this method to determine hotspots in the urban area.

3.2. Evaluating the Proposed Methods In this part, there are two aspects involved in the accuracy evaluation of urban travel hotspots, which are basic validation and prediction accuracy index (PAI).

3.2.1. Basic Validation Identifying urban hotspots addresses the discovery of social sensing and the patterns of human activities. The most common validating method is a direct visual comparison. As a result, we ﬁrst compared the results of the proposed identiﬁcation method with two other methods by directly showing results on a map.

3.2.2. Quantifying the Accuracy of Identification To validate the accuracy of detecting hotspots, researchers have begun to consider assessments that can be used to measure how good a method is for identification. There is an assessment named the prediction accuracy index (PAI), frequently used to evaluate the accuracy of crime prediction hotspots. It solved the problem of other assessments which do not take into account the size of the study areas where the crimes are predicted [40,41]. It has become the most popular accuracy measurement of predicting the spatial patterns of crimes [40–43]. The idea of the PAI can also be borrowed for quantifying the accuracy of identifying urban travel hotspots. The popularity of travel hotspots increases with the number of pick-ups and drop-offs that appear in an area in some respects. Counting the proportion of pick-ups and drop-offs that occur in the areas that are identified as hotspots to all of the areas is the method of calculation of the PAI. Equation (7) illustrates the process of the PAI as follows: n N Prediction Accuracy Index = 100% (7) a × A where n means the number of events within the areas identified as hotspots; N means the number of events within the study area; a and A represent the area of the hotspots and the study area, respectively.

4. Case Study

4.1. Study Area and Dataset As the center of politics, economy, technology, and transportation in China, Beijing attracts attention from all over the world. Additionally, it is a suitable area to discover urban travelling problems. Speciﬁcally, the study area in this article is within the Sixth Ring Road in Beijing, including six main administrative districts and also part of the suburbs. This area is the main region of human activities, with a huge population and convenient and complete transportation systems as well as well-developed road networks. Traditionally, 500 m, or a walking time of less than 6 min, is used to deﬁne the walkability of an area [44]. Considering the scale of human activity and spatial distribution of trajectories, we considered 500 m as the research diameter and divided the whole study area into regular grids (500 500 square meters). These regular grids became the urban units in this study. × Figure4 shows the whole study area and the distribution of regular grids. Sustainability 2020, 12, x FOR PEER REVIEW 8 of 21

its taxi’s ID, location, recording time, and status (vacancy or occupancy). Furthermore, taxi OD flows are effective to express a real human trip. The extracted OD flows reflect the spatial interactions between urban areas, especially long-range interactions. Table 1 shows examples of processed taxi data. The study area is denoted in green in Figure 4a. All place names mentioned in Section 4 and SustainabilitySection 5 are2020 corresponded, 12, 9662 to in Figure A1 of Appendix A. 8 of 20

Figure 4. The study area represented (b), which is denoted in green (a). Figure 4. The study area represented (b), which is denoted in green (a). Although the subway and buses are the most popular types of public transportation, taxis played Table 1. Samples of origin–destination (OD) flows with pick-ups and drop-offs. an important role in the urban transportation system in recent years [17]. Trajectory dataare attractive for analyzingTaxi ID intra-cityPick-Up interactions Time Pick and-U humanp Location travelling Drop patterns.-Off Time In this study,Drop- weOff applied Location a taxi dataset1000 including 2016 pick-ups-6-10 10:37:44 and drop-o 116.58874,ffs collected 40.07905 from more2016 than-6-10 15,00011:5:1 taxis116.39498, in a week 39.99156 (10 June to 16 June12179 in 2016) 2016 in Beijing.-6-12 14:33:52 All these 116.39114, data points 39.85552 are integrated 2016 into-6-12 24 14:50:22 h and grouped 116.31633, with 39.89542 4-h intervals for the1970 purpose 2016 of exploring-6-14 20:58:35 the temporal 116.55133, characteristics 40.06325 of2016 them.-6-14 Meanwhile, 21:7:54 116.47403, taxi origin–destination 40.01265 (OD) flows that embody human travelling activities are abstracted as a pair of pick-up and drop-off points.4.2. Spatio After-Temporal data preprocessing,Patterns of Urban each Travelling data point Hotspots contains its taxi’s ID, location, recording time, and status (vacancy or occupancy). Furthermore, taxi OD flows are effective to express a real human Previous studies showed that there is a considerable variation at different hours and different trip. The extracted OD flows reflect the spatial interactions between urban areas, especially long-range days regarding the number of taxi pick-ups and drop-offs [17,18]. Therefore, we selected, separately, interactions. Table1 shows examples of processed taxi data. The study area is denoted in green two groups of OD flows that appeared on weekdays and weekends. In the beginning, we employed in Figure4a. All place names mentioned in Sections4 and5 are corresponded to in Figure A1 of the interaction-based STDF to calculate the potential value of each grid in the study area. There is a AppendixA. 3D view showing the result of the potential value assignment in Figure 5, which was obtained from the taxi OD flows dataset (weekdays 13:00–16:00). The potential value is the measurement of the Table 1. Samples of origin–destination (OD) flows with pick-ups and drop-offs. dimension of height. Furthermore, the red columns and grids with high potential value represent the popularTaxi travelling ID areas Pick-Up in the Time study area. Pick-Up Location Drop-Off Time Drop-Off Location 1000 2016-6-10 10:37:44 116.58874, 40.07905 2016-6-10 11:5:1 116.39498, 39.99156 12179 2016-6-12 14:33:52 116.39114, 39.85552 2016-6-12 14:50:22 116.31633, 39.89542 1970 2016-6-14 20:58:35 116.55133, 40.06325 2016-6-14 21:7:54 116.47403, 40.01265

4.2. Spatio-Temporal Patterns of Urban Travelling Hotspots Previous studies showed that there is a considerable variation at different hours and different days regarding the number of taxi pick-ups and drop-offs [17,18]. Therefore, we selected, separately, two groups of OD flows that appeared on weekdays and weekends. In the beginning, we employed the interaction-based STDF to calculate the potential value of each grid in the study area. There is a 3D view showing the result of the potential value assignment in Figure5, which was obtained from the taxi OD flows dataset (weekdays 13:00–16:00). The potential value is the measurement of the dimension of height. Furthermore, the red columns and grids with high potential value represent the popular travelling areas in the study area. Sustainability 2020, 12, 9662 9 of 20 Sustainability 2020, 12, x FOR PEER REVIEW 9 of 21

Figure 5. An example of the result of calculating potential value in the grid. Figure 5. An example of the result of calculating potential value in the grid. 4.2.1. Dynamic Characteristics of Hotspots on Weekdays 4.2.1. Dynamic Characteristics of Hotspots on Weekdays We divided all taxi trajectories into two groups including pick-ups (O) and drop-offs (D). Normally, the spatio-temporalWe divided all taxi patterns trajectories of them into are two not groups same as including each other. pick As-ups a result,(O) and the drop trend-offs of (D). the Normally,number of the travelling spatio-temporal hotspots inpatterns the study of them area wasare not unwrapped same as ateach the other. beginning. As a result, Figure the6 illustrates trend of thethe number difference of betweentravelling the hotspots number in of the hotspots study identified area was between unwrapped the originsat the beginning. and destinations Figure on 6 illuweekdays.strates the The difference blue line withbetween the circle the numbermark represents of hotspots the number identified of pick-up between hotspots, the origins while and the destinationsyellow line with on weekdays. the square mark The blue constitutes line with the tendency the circle of mark drop-o representsff hotspots. the Although number both of pick of them-up hotspots,have a similar while upwardthe yellow fluctuation, line with thethe quantitysquare mark of destination constitutes hotspots the tend isency greater of drop than-off the hotspots. those of Althoughorigins. The both reason of them for thishave might a similar be that upward passengers fluctuation, would findthe thequantity nearest of placedestination to them hotspots where taxi is greaterdrivers than aggregate the those frequently of origins. when The travel reason started, for this and might they be would that passengers be dropped would off at find the the exact nearest same placedestinations. to themAt where 01:00–04:00, taxi driv theers number aggregate of origin frequently hotspots when was travel the smallest. started, Influencedand they wouldby human be dropactivities,ped off it increasedat the exact to 50same hotspots destinations. as a peak At at01:00 09:00–12:00.–04:00, the After number a slight of origin decrease, hotspots the number was the of smallest.pick-up hotspotsInfluenced rose by to human over 50 activities, again at it 17:00–20:00. increased to Then, 50 hotspots it dropped as a topeak around at 09:00 35– in12:00 the. eveningAfter a slight(21:00–00:00 decrease+, the1). number Similar of to pick the- pick-upup hotspots hotspots, rose to the over number 50 again of at destination 17:00–20:00. hotspots Then, it haddropped been tothrough around a small 35 in increase the evening from (21:00 01:00–04:00–00:00 to+ 05:00–08:00.1). Similar to Then, the the pick maximum-up hotspots, amount the appeared number of at destination17:00–20:00 withhotspots a mild had increase, been through which was a small over inc 60.rease In the from end, 01:00 it fell– back04:00 to to as 05:00 same–08:00. as the numberThen, the of maximumpick-up hotspots. amount appeared at 17:00–20:00 with a mild increase, which was over 60. In the end, it fell back toThe as spatio-temporalsame as the number distributions of pick-up of hotspots. pick-up and drop-off hotspots at six distinct periods are depicted in Figure7. The green grids and blue grids represent the hotspots of origins and destinations, respectively. These hotspots are discovered, almost on the whole, within the Fifth Ring Road, except those from 01:00–04:00. At the same time, there are several areas identified as pick-up and drop-off hotspots simultaneously at each period, such as Sanlitun, Wangjing, Beijing railway station, and Beijing Capital International Airport. This means that human activities are more active to fit the needs of work and business. Furthermore, the travel demands are varied on weekdays. Sustainability 2020, 12, 9662 10 of 20 Sustainability 2020, 12, x FOR PEER REVIEW 10 of 21

Sustainability 2020, 12, x FOR PEER REVIEW 11 of 21 Figure 6. FigureThe di 6.ﬀ Theerent different numbers number ofs originsof origins (O) (O) andand destinations destinations (D) at (D) different at diﬀ timeerents on times weekdays. on weekdays.

The spatio-temporal distributions of pick-up and drop-off hotspots at six distinct periods are depicted in Figure 7. The green grids and blue grids represent the hotspots of origins and destinations, respectively. These hotspots are discovered, almost on the whole, within the Fifth Ring Road, except those from 01:00–04:00. At the same time, there are several areas identified as pick-up and drop-off hotspots simultaneously at each period, such as Sanlitun, Wangjing, Beijing railway station, and Beijing Capital International Airport. This means that human activities are more active to fit the needs of work and business. Furthermore, the travel demands are varied on weekdays. Specifically, there are three places, grouped in Beijing Capital International Airport, Sanlitun, and Nanluoguxiang, where we detected seven pick-up hotspot areas in the early morning. After this period, the urban travelling hotspots were scattered from the center of the city to the Fifth Ring Road for the rest of the periods. However, the time interval 05:00–08:00 is the only period when the hotspots were not detected at Beijing Capital International Airport. There might be two reasons for this phenomenon. One is there are fewer flights than other periods. The other is that other activities such as commuting are more popular and predominant at this time interval. In this period, the hotspots

are identified around the01:00 two–04:00 main railway stations (Beijing railway station05:00–08:00 and Beijing West railway station) and residential communities, such as Xizhimen, Shuangjing, Caoqiao, Shangdi, and Zaoyuan. Except for those places that appeared in the last period, there are more hotspots distributed around places with more entertainment activities, such as Tiantan Park, Qianmen, Xidan, and Nanluoguxiang, or the areas with full of students (Wudaokou, Renmin University, and Datunlu) at 09:00–12:00. During 09:00–16:00, university students, old people, and children could have a chance to travel in the city. As a result, areas including university campuses and local parks would also be identified as hotspots on weekdays. At 13:00–16:00, the quantity and spatial distribution of hotspots both narrowed slightly. The popular pick-up areas that people prefer to choose are around entertainment sites and transportation centers, such as Xidan, Nanluoguxiang, Xizhimen, Sanlitun, Taiyanggong, Wangjing, and Wudaokou as well as Beijing railway station, Beijing West railway station, and Beijing Capital International Airport. After this period, the hotspots were discovered with a larger range than at 13:00–16:00. People tend to depart from some business areas, transportation hubs, and nightlife centers because there are enormous commuting activities and business trips in this period. Nightlife activities began to be popular after work as well. The hotspot areas included Dongzhimen, Sanlitun, Dawanglu, Qingnianlu, Wangjing, Datunlu, Wudaokou,

Zhongguancun, Xizhimen, Xidan, Shuangjing, Zaoyuan, Beijing railway station, Beijing West railway station, Beijing South railway09:00–12:00 station, and Beijing Capital International13:00 Airport.–16:00 In the deep of night, the hotspots are clustered at nightlife centers, housing areas, and transportation centers, such as Sanlitun, Nanluoguxiang, Wudaokou, Wangjing, Dawanglu, Shuangjing, Liujiayao, Zaoyuan, Beijing railway station, and Beijing Capital International Airport.

17:00–20:00 21:00–00:00+1

Figure 7. FigureThe dynamic 7. The dynamic distribution distribution of of pick-up pick-up and drop drop-o-off hotspotsﬀ hotspots on weekdays on weekdays at different at period diﬀerents. periods. Sustainability 2020, 12, 9662 11 of 20

Specifically, there are three places, grouped in Beijing Capital International Airport, Sanlitun, and Nanluoguxiang, where we detected seven pick-up hotspot areas in the early morning. After this period, the urban travelling hotspots were scattered from the center of the city to the Fifth Ring Road for the rest of the periods. However, the time interval 05:00–08:00 is the only period when the hotspots were not detected at Beijing Capital International Airport. There might be two reasons for this phenomenon. One is there are fewer flights than other periods. The other is that other activities such as commuting are more popular and predominant at this time interval. In this period, the hotspots are identified around the two main railway stations (Beijing railway station and Beijing West railway station) and residential communities, such as Xizhimen, Shuangjing, Caoqiao, Shangdi, and Zaoyuan. Except for those places that appeared in the last period, there are more hotspots distributed around places with more entertainment activities, such as Tiantan Park, Qianmen, Xidan, and Nanluoguxiang, or the areas with full of students (Wudaokou, Renmin University, and Datunlu) at 09:00–12:00. During 09:00–16:00, university students, old people, and children could have a chance to travel in the city. As a result, areas including university campuses and local parks would also be identified as hotspots on weekdays. At 13:00–16:00, the quantity and spatial distribution of hotspots both narrowed slightly. The popular pick-up areas that people prefer to choose are around entertainment sites and transportation centers, such as Xidan, Nanluoguxiang, Xizhimen, Sanlitun, Taiyanggong, Wangjing, and Wudaokou as well as Beijing railway station, Beijing West railway station, and Beijing Capital International Airport. After this period, the hotspots were discovered with a larger range than at 13:00–16:00. People tend to depart from some business areas, transportation hubs, and nightlife centers because there are enormous commuting activities and business trips in this period. Nightlife activities began to be popular after work as well. The hotspot areas included Dongzhimen, Sanlitun, Dawanglu, Qingnianlu, Wangjing, Datunlu, Wudaokou, Zhongguancun, Xizhimen, Xidan, Shuangjing, Zaoyuan, Beijing railway station, Beijing West railway station, Beijing South railway station, and Beijing Capital International Airport. In the deep of night, the hotspots are clustered at nightlife centers, housing areas, and transportation centers, such as Sanlitun, Nanluoguxiang, Wudaokou, Wangjing, Dawanglu, Shuangjing, Liujiayao, Zaoyuan, Beijing railway station, and Beijing Capital International Airport. As illustrated in Figure7, the spatio-temporal characteristics of drop-o ff hotspots have differences with pick-up hotspots. First of all, compared with pick-up hotspots, there are some resident hotspots discovered, apart from those at nightlife centers and transportation hubs, at 01:00–04:00. The main reason that passengers went back home is the end of various nightlife activities and business trips. In the next period, there are some hotspots that contain hospitals, such as the Children’s hospital and Jishuitan hospital. The other hotspots are distributed from the northeast of the Fifth Ring Road to the south of the Third Ring Road, which are some areas containing aggregated business activities, such as Dongzhimen, Wangjing, and railway stations. Then, at 09:00–12:00, hotspots are added around some business areas, including Sanlitun, Xizhimen, Caoqiao, Tiantan Park, and Beijing Capital International Airport. After this period, hotspots are identified in some areas with more types of human activities throughout the whole study area, such as Taiyanggong, Wangjing, Sanlitun, Nanluoguxiang, Wudaokou, Xidan, Gongzhufen, Qingnianlu, Jinsong, railway stations, and the airport. In the evening, entertainment activities are more predominant in the form of drop-off hotspots. Therefore, hotspots appeared at Datunlu, Dongzhimen, Xizhimen, Zhongguancun, and Dongdan, and vanished at Wangjing, Gongzhufen, and Jinsong at 17:00–20:00. Similar to the pick-up hotspots, the drop-off hotspots are clustered at nightlife centers, residential communities, and transportation centers, such as Sanlitun, Nanluoguxiang, Wudaokou, Taiyanggong, Shangdi, railway stations, and the airport. Generally, the appearing pick-up and drop-off hotspots reflected the types of human activities in the urban space. The spatio-temporal distributions of travel hotspots fit the routines of daily life on weekdays. There are distinct temporal differences between periods, such as the commuting and business periods and the nightlife period. Only a few nightlife places and the airport were identified in the early morning. Commuting activities dominated the travel hotspots at 05:00–08:00. However, human activities such as student activities during the daytime caused the hotspots to be scattered to Sustainability 2020, 12, 9662 12 of 20 some entertainment sites, universities, and urban parks. At 17:00–20:00, commutes and the closing times of attractions began to influence the hotspot areas. Nightlife activities also led to some travels from entertainment sites. After this period, the hotspots were often around nightlife centers. Apart from that, transportation centers were hotspot areas all day due to frequent business trips on weekdays.

4.2.2. Dynamic Characteristics of Hotspots on Weekends Figure8 illustrates the di fference between the number of hotspots identified between the origins and destinations on weekends. Different from the travelling hotspots on weekdays, the number of them has an opposite trend after 13:00. While the numbers of origin and destination hotspots are similar from 01:00–04:00 to 09:00–12:00, after that, the number of drop-off hotspots is more short-distributed. This might be explained by the fact that more and more people visit some places for recreation from the entire city around the afternoon at weekends. The pick-up hotspots reached more than 50 as a peak at 05:00–08:00. Then, there was a slight drop before 13:00–16:00 and the value fell to just over 40. Influenced by the active nightlife, the number of pick-up hotspots climbed constantly to 50 at 17:00–20:00 and even to 66 in the evening (21:00–00:00+1). As for the drop-off hotspots, there was a sharp fall from the peak of over 50 at 05:00–08:00 to under 20 at 17:00–20:00. The number was consistent into theSustainability next period, 2020, 12, whichx FOR PEER was REVIEW similar to the number from 01:00–04:00. 13 of 21

FigureFigure 8. The 8. The diﬀ differenterent numbers numbers of of origins origins (O)(O) and destinations destinations (D) (D) at atdifferent diﬀerent time timess on weekends. on weekends.

The spatio-temporalThe spatio-temporal patterns patterns of pick-up of pick and-up drop-oand dropff hotspots-off hotspots at six at di sixfferent different periods periods are illustrated are in Figureillustrated9. Di infferent Figure from 9. Different the spatial from distribution the spatial distribution of hotspots of onhotspots weekdays, on weekdays, most of most the hotspotsof the of originshotspots and destinations of origins and appeared destinations at the appeared same gridsat the onsame weekends. grids on weekends. Especially Especially at 09:00–12:00, at 09:00 there– is only12:00, one place there aroundis only one Wangjing place around identified Wangjing as only identified containing as only drop-o containingff hotspots. drop-off Thishotspot phenomenons. This reflectsphenomenon that human reflect activitiess that human are relatively activities homogeneous are relatively homogeneous at weekends. at weekends. TheThe green green grids grids show show the the spatial spatialdistribution distribution of of pick pick-up-up hotspots hotspots on onweekends. weekends. Specifically, Specifically, there are some popular places detected at Beijing Capital International Airport, Sanlitun, Qingnianlu, there are some popular places detected at Beijing Capital International Airport, Sanlitun, Qingnianlu, Shuangjing, and Wudaokou at 01:00–04:00; most of them are distributed in the east of the city. These Shuangjing, and Wudaokou at 01:00–04:00; most of them are distributed in the east of the city. places are popular nightlife centers, except the airport. After this period, the pick-up hotspots are Thesescattered places arethrough popularout the nightlife whole study centers, area exceptfor the rest the of airport. the periods. After At this 05:00 period,–08:00, the hotspot pick-up areas hotspots are scatteredare identified throughout at transportation the whole hubs study and area residence for thes rest in theofthe city, periods. which includeAt 05:00–08:00, Beijing Capital the hotspot areasInternational are identified Airport, at transportation Beijing railway hubs station, and residencesSihui, Wangjing, in the Dongdaqiao, city, which Shuangjing include Beijing-Jinsong, Capital InternationalLiujiayao, Airport,Caoqiao, and Beijing Gongzhufen. railway station,After this Sihui, period, Wangjing, the number Dongdaqiao, of hotspots decreased Shuangjing-Jinsong, gently. Liujiayao,There are Caoqiao, 50 hotspots and that Gongzhufen. appeared regularly After thiswithin period, the Fifth the Ring number Road at of 09:00 hotspots–12:00. decreasedThose places gently. Therecovered are 50 hotspots entertainment that appeared sites, such regularly as Tiantan within Park, the Fifth Xisi, Ring Xizhimen, Road at Wudaokou, 09:00–12:00. Wangjing, Those places coveredDongzhimen, entertainment Dongdan sites,, and such Sanlitun. as Tiantan Further Park,more Xisi,, transportation Xizhimen, Wudaokou, centers are Wangjing, the main Dongzhimen, popular departure sites for people in this period. In the afternoon, the popular pick-up areas transferred Dongdan, and Sanlitun. Furthermore, transportation centers are the main popular departure sites for people to other famous entertainment centers and transportation hubs in the city, such as Xidan, people in this period. In the afternoon, the popular pick-up areas transferred people to other famous Nanluoguxiang, Sanlitun, Taiyanggong, Wangjing, Jinsong, Zhongguancun, Caoqiao, and Zaoyuan, entertainmentas well as Beijingcenters railway and transportation station, Beijing hubs West in railway the city, station, such asand Xidan, Beijing Nanluoguxiang, Capital International Sanlitun, Taiyanggong,Airport. With Wangjing, the development Jinsong, of Zhongguancun, nightlife, more entert Caoqiao,ainment and sites Zaoyuan, in the entire as wellstudy as area Beijing become railway station,popular. Beijing Besides West railwaythose hotspots station, show andn Beijing in the Capital last period, International Dawanglu, Airport. Datunlu, With Wudaokou the development, and of nightlife,Xizhimen more are also entertainment discovered as sites pick in-up the hotspots. entire study In the areadeep becomeof night, popular.the hotspots Besides are clustered those hotspots at nightlife centers and transportation centers from the north Fifth Ring Road to the south Fourth Ring Road, where Sanlitun attracts a huge cluster of hotspots. The blue grids in Figure 9 show the dynamic distribution of drop-off hotspots on weekends. Different from pick-up hotspots at 01:00–04:00, the drop-off hotspots in this period are distributed not only around nightlife centers but also around areas with various university students, such as Sanlitun, Nanluoguxiang, Wudaokou, and Taiyanggong. However, the hotspots of destinations have the same distribution as the hotspots of origins from 05:00 to 12:00. This is because human activities on weekends would not become popular in this period. At 13:00–16:00, there are fewer places discovered as hotspots throughout the entire study area, including Nanluoguxiang, Xidan, Jinsong, Liujiayao, Caoqiao, Qingnianlu, Wangjing, Wudaokou, Haidianhuangzhuang, Xingong, Zaoyuan, Beijing railway station, and Beijing Capital International Airport. In the evening, the spatial distribution of drop-off hotspots is nearly located entirely within the Fourth Ring Road. At 17:00– 20:00, all hotspots detected are scattered at those places highly related to residents’ daily life, such as Xidan, Dongzhimen, Jinsong, Sihui, and Zaoyuan. While Beijing railway station and Beijing Capital International Airport are identified as hotspots in the deep of night, except from that, people prefer to arrive around Sanlitun, Dawanglu, Xizhimen, Taiyanggong, Wangjing, and Liujiayao. Sustainability 2020, 12, 9662 13 of 20 shown in the last period, Dawanglu, Datunlu, Wudaokou, and Xizhimen are also discovered as pick-up hotspots. In the deep of night, the hotspots are clustered at nightlife centers and transportation centers from theSustainability north Fifth 2020, 1 Ring2, x FOR Road PEER REVIEW to the south Fourth Ring Road, where Sanlitun attracts a14 huge of 21 cluster of hotspots.

01:00–04:00 05:00–08:00

09:00–12:00 13:00–16:00

17:00–20:00 21:00–00:00+1

FigureFigure 9. The 9. dynamic The dynamic distribution distribution of of pick-up pick-up and and drop drop-o-off hotspotsﬀ hotspots on weekends on weekends at different at di periodﬀerents. periods.

The blue grids in Figure9 show the dynamic distribution of drop-o ff hotspots on weekends. Different from pick-up hotspots at 01:00–04:00, the drop-off hotspots in this period are distributed Sustainability 2020, 12, 9662 14 of 20 not only around nightlife centers but also around areas with various university students, such as Sanlitun, Nanluoguxiang, Wudaokou, and Taiyanggong. However, the hotspots of destinations have the same distribution as the hotspots of origins from 05:00 to 12:00. This is because human activities on weekends would not become popular in this period. At 13:00–16:00, there are fewer places discovered as hotspots throughout the entire study area, including Nanluoguxiang, Xidan, Jinsong, Liujiayao, Caoqiao, Qingnianlu, Wangjing, Wudaokou, Haidianhuangzhuang, Xingong, Zaoyuan, Beijing railway station, and Beijing Capital International Airport. In the evening, the spatial distribution of drop-off hotspots is nearly located entirely within the Fourth Ring Road. At 17:00–20:00, all hotspots detected are scattered at those places highly related to residents’ daily life, such as Xidan, Dongzhimen, Jinsong, Sihui, and Zaoyuan. While Beijing railway station and Beijing Capital International Airport are identified as hotspots in the deep of night, except from that, people prefer to arrive around Sanlitun, Dawanglu, Xizhimen, Taiyanggong, Wangjing, and Liujiayao. In general, the hotspots detected on weekends are dispersed in the entire urban space, while the types of places that hotspots covered are almost all entertainment sites. The pick-up hotspots distributed around residences and entertainment centers since 13:00 and the drop-off hotspots are concentrated on the busiest entertainment sites in the urban spaces from the afternoon to the deep of night. This is because people tend to be relaxed on weekends. Before the afternoon, the daily routines of residents are stable. As a result, the pick-up hotspots are detected around residence communities in the morning. Besides, transportation hubs, such as railway stations and airports, also are identified as travelling hotspots at weekends. Business trips and tourists are the two important factors that contributed to frequent visits.

4.3. Evaluation of Accuracy We further compared the overlapping results between the Getis-Ord Gi* hotspot analysis, the trajectory data field, and the proposed method to investigate the effectiveness of the proposed method in a particular area. As shown in Figure 10, the yellow grids, the grids with the hashed black line, and the grids with black dots represent the spatial distribution of the Getis-Ord Gi * hotspot analysis, the trajectory data field, and the proposed method at 09:00–12:00 on weekdays, respectively. Furthermore, the orange and the purple ones are the grids overlapped by those three colors. There are 140 complete grids in this particular area. The result of the Getis-Ord Gi * hotspot analysis covered the entire area. It cannot be a piece of evidence to reflect the popular travel areas in people’s daily lives. The spatial distributions of travel hotspots discovered by the other two methods are more identical in comparison. However, the hotspot areas with black lines which appeared in 42 grids crossed a large area from the west of the Second Ring Road to the east of the Third Ring Road. The two main hotspot areas depicted in Figure 10 covered 3.25 km2 and 5.5 km2, which are too big to fit the human cognition of hotspots. The hotspots extracted by the trajectory data field are not precise enough to be representative of the popular area. As for our method, 10 hotspots were detected from all of the grids, which make up a 7.1% share of this area. For instance, the black dot hotspots cover the exact area of a famous entertainment site in Sanlitun, while the black lines cover a large and inaccurate area to reflect this popular place, which surrounds the center of Sanlitun. In general, the method with a greater PAI value has a good performance. Table2 shows the quantitative comparison of different methods that identify urban travel hotspots on weekdays and weekends. It indicates that the pros and cons of the three identifying algorithms are reflected through the trajectory data. The results show that there are distinct differences between Getis-Ord Gi * hotspot analysis and the methods from the perspective of the data field. The PAI values of the trajectory data field and the proposed method are higher than that of the Getis-Ord Gi *. It means that methods from the perspective of a data field could find more pick-ups and drop-offs in a smaller area. Additionally, the proposed identification method performed better than the trajectory data field in most instances, especially those trajectories on weekends. In particular, the mean PAIs of the proposed method of origins on weekdays and weekends are nearly three times higher than the mean values of the Getis-Ord Sustainability 2020, 12, x FOR PEER REVIEW 15 of 21

In general, the hotspots detected on weekends are dispersed in the entire urban space, while the types of places that hotspots covered are almost all entertainment sites. The pick-up hotspots distributed around residences and entertainment centers since 13:00 and the drop-off hotspots are concentrated on the busiest entertainment sites in the urban spaces from the afternoon to the deep of night. This is because people tend to be relaxed on weekends. Before the afternoon, the daily routines of residents are stable. As a result, the pick-up hotspots are detected around residence communities in the morning. Besides, transportation hubs, such as railway stations and airports, also are identified as travelling hotspots at weekends. Business trips and tourists are the two important factors that contributed to frequent visits.

4.3. Evaluation of Accuracy We further compared the overlapping results between the Getis-Ord Gi* hotspot analysis, the trajectory data field, and the proposed method to investigate the effectiveness of the proposed method in a particular area. As shown in Figure 10, the yellow grids, the grids with the hashed black line, and the grids with black dots represent the spatial distribution of the Getis-Ord Gi * hotspot Sustainability 2020, 12, 9662 15 of 20 analysis, the trajectory data field, and the proposed method at 09:00–12:00 on weekdays, respectively. Furthermore, the orange and the purple ones are the grids overlapped by those three colors. There are 140 complete grids in this particular area. The result of the Getis-Ord Gi * hotspot analysis covered Gi * analysis,the entire and area. almost It cannot twice be as a piece great of as evidence the trajectory to reflect the data popular field. travel The area means in valuespeople’sof daily destinations on weekdayslives.and The weekendsspatial distributions have a of similar travel hotspots trend, discoveredalthough by the the proposed other two method methods are discovered more a few more drop-oidenticalffs while in comparison. covering However, a similar the area. hotspot Moreover, areas with the black Wilcoxon lines which signed appeared rank in (WSR) 42 grids test results shown incrossed Figure a 11 large provide area from an the additional west of the insightSecond Ring that Road cannot to the be east gained of the Thi fromrd Ring the Road. specific The and mean two main hotspot areas depicted in Figure 10 covered 3.25 km2 and 5.5 km2, which are too big to fit PAI results.theIn human the casecognition of origins of hotspots. on weekdays,The hotspots theextracted proposed by the methodtrajectory data has field a better are not performance precise than other methodsenough with to be a representative lower significance of the popular(p area.0.05 )As. Therefor our ismethod, a same 10 significanthotspots were result detected for from the proposed ≤ method whenall of the it identified grids, which drop-o make upff hotspots.a 7.1% share Atof this weekends, area. For instance, the proposed the black method dot hotspots outperformed cover the other twothe methods exact area(p of a0.1 famous) for discoveringentertainment site either in Sanlitun, origins while or destinations. the black lines cover a large and inaccurate area ≤to reflect this popular place, which surrounds the center of Sanlitun.

Sustainability 2020, 12, x FOR PEER REVIEW 16 of 21 weekends. It indicates that the pros and cons of the three identifying algorithms are reflected through the trajectory data. The results show that there are distinct differences between Getis-Ord Gi * hotspot analysis and the methods from the perspective of the data field. The PAI values of the trajectory data field and the proposed method are higher than that of the Getis-Ord Gi *. It means that methods from the perspective of a data field could find more pick-ups and drop-offs in a smaller area. Additionally, the proposed identification method performed better than the trajectory data field in most instances, FigureFigure 10. The 10. The comparison comparison ofof different different hotspot hotspot identification identifications.s. especially those trajectories on weekends. In particular, the mean PAIs of the proposed method of originsTable on 2.weekdaysTheIn general, prediction and the weekends accuracymethod with index are a greater nearly (PAI) valuePAI three value of times di hasfferent a highergood methods performance. than on the weekdays meanTable 2 value shows and weekends.s theof the Getis- quantitative comparison of different methods that identify urban travel hotspots on weekdays and Ord GiWeekdays * analysis/Weekends, and almost Method twice as01:00–04:00 great 05:00–08:00as the trajectory 09:00–12:00 13:00–16:00 data field. 17:00–20:00 The mean21:00–00:00 values+1 Mean of destinations on weekdaysGetis-Ord and Gi* weekends 3.680 have a 3.423 similar 3.737trend, although 3.816 the3.815 proposed 3.608 method 3.680 Weekdays Data field 12.270 10.003 9.598 9.174 6.811 6.018 8.979 discovered a few moreInteraction-based drop-offs while STDF covering24.682 a similar8.252 area. 8.377 Moreover, 9.929 the 8.228Wilcoxon 8.857 signed 11.388 rank O (WSR) test results shownGetis-Ord in Figure Gi* 11 provide 3.636 an additional 2.926 3.150 insight 3.305that cannot 3.922 be gained 3.137 from 3.346 the Weekends Data field 7.438 4.802 6.156 6.856 5.108 5.139 5.917 specific and mean PAIInteraction-based results. In STDFthe case10.089 of origins7.787 on weekdays,8.105 the 8.159 proposed 14.935 method 6.498 has a better 9.262 Getis-Ord Gi* 3.669 3.413 3.624 3.951 3.815 3.485 3.660 performanceWeekdays than other methodsData field with a lower 8.454 signi 10.313ficance 8.686 (푝 ≤ 0.05 8.895). There 6.587 is a same 5.846 significant 8.130 Interaction-based STDF 10.565 11.623 8.952 7.983 7.420 6.301 8.807 resultD for the proposed method when it identified drop-off hotspots. At weekends, the proposed Getis-Ord Gi* 3.572 3.057 3.263 3.412 3.895 3.153 3.392 method Weekends outperformed theData fieldother two 10.150 methods 4.703(푝 ≤ 0.1 5.863) for discovering7.088 5.599 either 5.858 origins 6.544 or Interaction-based STDF 6.604 8.422 8.139 8.402 12.761 5.864 8.365 destinations.

Figure 11. TThehe results results of of the the Wilcoxon Wilcoxon signed signed rank (WSR) test, showing the the significancesignificance of any differencesdifferences between the PAI valuesvalues ofof the methods. The green part means the comparison of methods for origins origins on on weekdays, weekdays, the the blue blue part representspart represents destinations destinations on weekdays, on weekdays, the purple the part purple represents part representsorigins on weekends,origins on weekends, and the grey and part the represents grey part destinationsrepresents destinations on weekends. on weekends. All parts follow All parts the followrule that the the rule row that method the row is more method accurate is more than theaccurate column than method. the column When method.the color Whenbecomes the darker, color becomesthe significance darker, is the lower. significance is lower.

Table 2. The prediction accuracy index (PAI) value of different methods on weekdays and weekends.

Weekdays/ 01:00– 05:00– 09:00– 13:00– 17:00– 21:00– Method Mean Weekends 04:00 08:00 12:00 16:00 20:00 00:00+1 Getis-Ord 3.680 3.423 3.737 3.816 3.815 3.608 3.680 Gi* Weekdays Data field 12.270 10.003 9.598 9.174 6.811 6.018 8.979 Interaction- 24.682 8.252 8.377 9.929 8.228 8.857 11.388 based STDF O Getis-Ord 3.636 2.926 3.150 3.305 3.922 3.137 3.346 Gi* Weekends Data field 7.438 4.802 6.156 6.856 5.108 5.139 5.917 Interaction- 10.089 7.787 8.105 8.159 14.935 6.498 9.262 based STDF Getis-Ord 3.669 3.413 3.624 3.951 3.815 3.485 3.660 Gi* Weekdays Data field 8.454 10.313 8.686 8.895 6.587 5.846 8.130 Interaction- 10.565 11.623 8.952 7.983 7.420 6.301 8.807 based STDF D Getis-Ord 3.572 3.057 3.263 3.412 3.895 3.153 3.392 Gi* Weekends Data field 10.150 4.703 5.863 7.088 5.599 5.858 6.544 Interaction- 6.604 8.422 8.139 8.402 12.761 5.864 8.365 based STDF SustainabilitySustainability2020 20, 2012, 1 96622, x FOR PEER REVIEW 16 of17 20of 21

5. Discussion 5. Discussion In this section, we further analyze the spatio-temporal patterns of spatial interaction between In this section, we further analyze the spatio-temporal patterns of spatial interaction between urban travelling hotspots using spatial interaction networks. At the same time, the structures of cities urban travelling hotspots using spatial interaction networks. At the same time, the structures of are closely related to intra-city human travelling patterns [45–48]. The spatial structures of hotspots cities are closely related to intra-city human travelling patterns [45–48]. The spatial structures of will also be discovered in this process. The hotspots detected from 09:00–12:00 are chosen to be an hotspots will also be discovered in this process. The hotspots detected from 09:00–12:00 are chosen example of discovering the characteristics of urban spatial structures. In the beginning, the directed to be an example of discovering the characteristics of urban spatial structures. In the beginning, and weighted spatial-interaction network of hotspots (SINH) has been built. Then, degree centrality the directed and weighted spatial-interaction network of hotspots (SINH) has been built. Then, has been employed to depict the characteristics of spatial interactions between hotspots and urban degree centrality has been employed to depict the characteristics of spatial interactions between structures. hotspots and urban structures. Because of the direction of each node, the in-degree centrality only belongs to the pick-up Because of the direction of each node, the in-degree centrality only belongs to the pick-up hotspots, and the out-degree centrality is suitable for the drop-off hotspots. As illustrated in Figure hotspots, and the out-degree centrality is suitable for the drop-off hotspots. As illustrated in Figure 12, 12, the hotspots discovered on weekends are more popular for both pick-up and drop-off passengers the hotspots discovered on weekends are more popular for both pick-up and drop-off passengers than than on weekdays. It indicates that the departure areas and arriving areas meet the regular demands on weekdays. It indicates that the departure areas and arriving areas meet the regular demands for for residents in the city at 09:00–12:00 on weekdays. However, the pick-up and drop-off hotspots are residents in the city at 09:00–12:00 on weekdays. However, the pick-up and drop-off hotspots are various and contain different types of urban travel on weekdays. There is further evidence to prove various and contain different types of urban travel on weekdays. There is further evidence to prove this this phenomenon from the spatial distribution of urban travelling hotspots. The grids 4358, 4359, phenomenon from the spatial distribution of urban travelling hotspots. The grids 4358, 4359, 4003, 4962, 4003, 4962, 5087, and 5841 contribute significantly to urban travelling on weekdays, which cover 5087, and 5841 contribute significantly to urban travelling on weekdays, which cover Sanlitun, Caoqiao, Sanlitun, Caoqiao, Wufangqiao, Beijing railway station, and Beijing West railway station. These areas Wufangqiao, Beijing railway station, and Beijing West railway station. These areas play a significant play a significant role as bridges in connecting other grids, which are seen as the travelling centers in role as bridges in connecting other grids, which are seen as the travelling centers in the study area. the study area. The grids 3794, 4441, 4856, 5824, and 6037, with higher out-degree centrality than The grids 3794, 4441, 4856, 5824, and 6037, with higher out-degree centrality than others, are the most others, are the most important departure hotspots, which include Xidan, Xizhimen, Jianguomen, important departure hotspots, which include Xidan, Xizhimen, Jianguomen, Tiantan Park, Sanyuan Tiantan Park, Sanyuan Bridge, and Beijing South railway station on weekdays. Meanwhile, Jinsong, Bridge, and Beijing South railway station on weekdays. Meanwhile, Jinsong, Shilihe, Dahongmen, Shilihe, Dahongmen, and Beijing Capital International Airport have a great in-degree centrality to and Beijing Capital International Airport have a great in-degree centrality to attract travel because they attract travel because they are located in grids 4775, 4130, 3692, 8651, and 5303. The nodes with a great are located in grids 4775, 4130, 3692, 8651, and 5303. The nodes with a great in-degree centrality and a in-degree centrality and a great out-degree centrality appeared in different areas, which means the great out-degree centrality appeared in different areas, which means the distributions of urban travels distributions of urban travels with strong regularity are stable on weekdays. with strong regularity are stable on weekdays.

Weekdays Weekends

FigureFigure 12. 12.The The directed directed spatial-interaction spatial-interaction network network of of hotspots hotspots at at09:00–12:00 09:00–12:00 on on weekdays weekdays and and weekends.weekends. The The blue blue nodes, nodes, the the green green nodes, nodes, and and the yellowthe yellow nodes nodes represent represent the grids the grids as both as pick-upboth pick- andup drop-o and dropﬀ hotspots,-off hotspots, only pick-uponly pick hotspots,-up hotspots and, and only only drop-o dropﬀ -hotspots,off hotspots, respectively. respectively. The The more more importantimportant nodes node withs with a greater a greater degree degree centrality centrality are are bigger. bigger.

ComparedCompared with with weekdays, weekdays, the the most most popular popular grids grids in thein the urban urban space space involve involve both both destinations destinations andand places places ofdeparture. of departure The. travelling The travelling hotspots hotspots identiﬁed identified on weekends on weekends which dominate which dominate urban travels urban aretravels aggregated are aggregate at Sanlitun,d at Taiyanggong,Sanlitun, Taiyanggong, Wufangqiao, Wufangqiao, Beijing railway Beijing station, railway and station Beijing, and Capital Beijing InternationalCapital International Airport (grids Airport 4358, (grid 4359,s 6717,4358, 5841, 4359, 5518, 6717, 9289, 5841, and 5518, 5405). 9289 The, and spatial 5405 distributions). The spatial anddistributions the urban functions and the ofurban these functions areas demonstrate of these area thats humandemonstrate activities that ﬁt human various activities residents’ fit needsvarious Sustainability 2020, 12, 9662 17 of 20 more than on weekdays. Indeed, entertainment sites would be the most popular demand for residents’ travel on weekends. Although the spatial distribution of hotspots would be dispersed geographically either on weekdays or on weekends, connections between the two of them exist directly with a short path. This means that the spatial-interaction network of hotspots is similar to a small-world network on weekdays and weekends.

6. Conclusions Identifying urban travelling hotspots has become a popular trend for researchers in recent years. It also plays a vital role in discovering the characteristics of human mobility. At the same time, taxis, an important means of public transportation, reflect urban individual travels. Trajectory data with pick-up points and drop-off points could be seen as an abstract of spatial interactions that connect two different places in the city directly. They also provide a new opportunity for reflecting urban spatial structures, which are helpful to solve urban problems and promote urban sustainable development. Therefore, this study combined the data field theory with spatial interactions and proposed an identification tool named the interaction-based spatio-temporal data field to discover urban travelling hotspots. The proposed method could give us new insights into what is happening to human travels in the urban area. There is a case study to detect urban travelling hotspots using the proposed method within the Sixth Ring Road in Beijing. First of all, trajectory data were aggregated to pick-ups and drop-offs at six periods on weekdays and weekends. Each 4-h period was integrated as a time interval. Then, urban travelling hotspots were identified with the proposed method at each period. Next, we analyzed the spatio-temporal patterns of those hotspots during a day. Finally, qualitative and quantitative evaluations were employed to test the accuracy of the proposed method. The following conclusions were obtained in this study: (1) Urban travelling hotspots are most scattered within Fifth Ring Road at different periods. However, the quantities and geographic distributions of them are quite distinct. In quantity, the number of hotspots was no more than 60 at any period on weekdays. There was a similar trend with two peaks for origins and destinations on weekdays. Compared with the numbers on weekdays, the first peak with around 50 hotspots appeared at 05:00–08:00 for pick-ups and drop-offs at weekends. After 13:00–16:00, the number of pick-up hotspots had an upward trend, reaching over 60, while the number of drop-off hotspots dropped to under 20. The spatial distribution of hotspots fit the needs of regular residents in the city. For instance, job–home areas would be popular during the commute period (including 05:00–08:00 and 17:00–20:00) on weekdays due to a large number of commuting activities. The hotspots were almost all distributed around entertainment sites at weekends. At the same time, transportation hubs and nightlife centers become hotspots at a particular time either on weekdays or weekends. (2) The results from the case study were compared with hotspots identified by Getis-Ord Gi* and identification with data field through overlapping comparison and PAI. In total, the accuracy of the proposed method performed better than the other two methods. Specifically, 99.3%, 30%, and 7.1% of areas were identified as hotspots in the sample area by Getis-Ord Gi*, data field, and interaction-based STDF, respectively. The PAI and mean PAI results show that the hotspots identified by the proposed method could discover more points in smaller areas. These results also pass the WSR test with a lower significance (p 0.1). ≤ (3) We further discuss the centrality of travelling hotspots in urban structures. Workplaces and transportation centers became popular areas on weekdays, while entertainment sites dominated urban travels at weekends. Furthermore, the connections between hotspots reflected the characteristics of the small-world network on weekdays and weekends. This research offers a new perspective for analyzing human travelling and urban structures based on the data field theory, yet there are limitations here. In the beginning, this proposed method is suitable for fixed study units. However, the hotspot area should be irregular with human cognition Sustainability 2020, 12, 9662 18 of 20 Sustainability 2020, 12, x FOR PEER REVIEW 19 of 21 andConsidering experience the inthreshold the urban of urban environment travel time, rather an thanappropriate existing time areas. interval In addition, could help the us di tofference better inexplore time betweenthe characteristics two data of points urban appearing travel. is an important factor in the interaction-based STDF. Considering the threshold of urban travel time, an appropriate time interval could help us to better exploreAuthor Contributions: the characteristics Conceptualization of urban travel., D.Y. and J.Z.; methodology, D.Y.; validation, D.Y.; formal analysis, D.Y.; resources, J.Z.; writing—original draft preparation, D.Y.; writing—review and editing, D.Y., J.Q. and Y.L.; Author Contributions: Conceptualization, D.Y. and J.Z.; methodology, D.Y.; validation, D.Y.; formal analysis, visualization, D.Y.; supervision, J.Z. All authors have read and agreed to the published version of the D.Y.; resources, J.Z.; writing—original draft preparation, D.Y.; writing—review and editing, D.Y., J.Q. and Y.L.; visualization,manuscript. D.Y.; supervision, J.Z. All authors have read and agreed to the published version of the manuscript. Funding:Funding: ThisThis research research was was funded funded by by the the National National Natur Naturee Science Science Foundation Foundation of China of China,, grant grant number number No. No.42071376. 42071376.

Acknowledgments:Acknowledgments: WeWe wouldwould likelike toto acknowledgeacknowledge thethe editorseditors forfor thethe editingediting assistanceassistance andand thethe reviewersreviewers forfor their constructive comments on our paper. We are also grateful for the valuable comments of Jun Xu. their constructive comments on our paper. We are also grateful for the valuable comments of Jun Xu. Conﬂicts of Interest: The authors declare no conﬂict of interest. Conflicts of Interest: The authors declare no conflict of interest. Appendix A Appendix A

FigureFigure A1.A1. The main toponym in human cognition inin thethe studystudy area.area.

ReferencesReferences 1. Li, S.; Ma, Y. Urbanization, economic development and environmental change. Sustainability 2014, 6, 1. Li, S.; Ma, Y. Urbanization, economic development and environmental change. Sustainability 2014, 6, 5143– 5143–5161. [CrossRef] 5161. 2. Yang, X.; Zhao, Z.; Lu, S. Exploring spatio-temporal patterns of urban human mobility hotspots. Sustainability 2. Yang, X.; Zhao, Z.; Lu, S. Exploring spatio-temporal patterns of urban human mobility hotspots. 2016, 8, 674. [CrossRef] Sustainability 2016, 8, 674. 3. Qin, K.; Xu, Y.; Kang, C.; Sobolevsky, S.; Kwan, M.-P. Modeling spatio-temporal evolution of urban crowd 3. Qin, K.; Xu, Y.; Kang, C.; Sobolevsky, S.; Kwan, M.-P. Modeling spatio-temporal evolution of urban crowd ﬂows. ISPRS Int. Geo-Inf. 2019, 8, 570. [CrossRef] flows. ISPRS Int. Geo-Inf. 2019, 8, 570. 4. Crampton, J.W.; Graham, M.; Poorthuis, A.; Shelton, T.; Stephens, M.; Wilson, M.W.; Zook, M. Beyond the 4. Crampton, J.W.; Graham, M.; Poorthuis, A.; Shelton, T.; Stephens, M.; Wilson, M.W.; Zook, M. Beyond the geotag: Situating “Big Data” and leveraging the potential of the geoweb. Cartogr. Geogr. Inf. Sci. 2013, 40, geotag: Situating “Big Data” and leveraging the potential of the geoweb. Cartogr. Geogr. Inf. Sci. 2013, 40, 130–139. [CrossRef] 130–139. 5. Liu, Y.; Liu, X.; Gao, S.; Gong, L.; Kang, C.; Zhi, Y.; Shi, L. Social sensing: A new approach to understanding 5. Liu, Y.; Liu, X.; Gao, S.; Gong, L.; Kang, C.; Zhi, Y.; Shi, L. Social sensing: A new approach to understanding our socioeconomic environments. Ann. Assoc. Am. Geogr. 2015, 105, 512–530. [CrossRef] our socioeconomic environments. Ann. Assoc. Am. Geogr. 2015, 105, 512–530. 6. Schneider, C.M.; Belik, V.; Couronné, T.; Smoreda, Z.; González, M.C. Unravelling daily human mobility 6. Schneider, C.M.; Belik, V.; Couronné, T.; Smoreda, Z.; González, M.C. Unravelling daily human mobility motifs. J. R. Soc. Interface 2013, 10, 20130246. [CrossRef] motifs. J. R. Soc. Interface 2013, 10, 20130246. 7. Pappalardo, L.; Simini, F.; Rinzivillo, S.; Pedreschi, D.; Giannotti, F.; Barabási, A.-L. Returners and explorers dichotomy in human mobility. Nat. Commun. 2015, 6, 8166–8166. Sustainability 2020, 12, 9662 19 of 20

7. Pappalardo, L.; Simini, F.; Rinzivillo, S.; Pedreschi, D.; Giannotti, F.; Barabási, A.-L. Returners and explorers dichotomy in human mobility. Nat. Commun. 2015, 6, 8166. [CrossRef] 8. Yan, X.Y.; Wang, W.X.; Gao, Z.Y.; Lai, Y.-C. Universal model of individual and population mobility on diverse spatial scales. Nat. Commun. 2017, 8, 1639. [CrossRef] 9. Pappalardo, L.; Simini, F. Data-driven generation of spatio-temporal routines in human mobility. Data Min. Knowl. Discov. 2018, 32, 787–829. [CrossRef] 10. Yang, C.; Yan, F.; Ukkusuri, S.V. Unraveling traveler mobility patterns and predicting user behavior in the Shenzhen metro system. Transportmetrica 2018, 14, 576–597. [CrossRef] 11. Chen, P.; Shi, W.; Zhou, X.; Liu, Z.; Fu, X. STLP-GSM: A method to predict future locations of individuals based on geotagged social media data. Int. J. Geogr. Inf. Sci. 2019, 33, 2337–2362. [CrossRef] 12. Yao, X.; Zhu, D.; Gao, Y.; Wu, L.; Zhang, P.; Liu, Y. A stepwise spatio-temporal flow clustering method for discovering mobility trends. IEEE Access 2018, 6, 44666–44675. [CrossRef] 13. Tao, R.; Thill, J.-C. Flow cross K-function: A bivariate flow analytical method. Int. J. Geogr. Inf. Sci. 2019, 33, 2055–2071. [CrossRef] 14. Zhu, D.; Huang, Z.; Shi, L.; Wu, L.; Liu, Y. Inferring spatial interaction patterns from sequential snapshots of spatial distributions. Int. J. Geogr. Inf. Sci. 2018, 32, 783–805. [CrossRef] 15. Xu, Y.; Belyi, A.; Bojic, I.; Ratti, C. Human mobility and socioeconomic status: Analysis of Singapore and Boston. Comput. Environ. Urban Syst. 2018, 72, 51–67. [CrossRef] 16. Tang, J.; Gao, F.; Liu, F.; Zhang, W.; Qi, Y. Understanding spatio-temporal characteristics of urban travel demand based on the combination of GWR and GLM. Sustainability 2019, 11, 5525. [CrossRef] 17. Yang, J.; Yi, D.; Qiao, B.; Zhang, J. Spatio-temporal change characteristics of spatial-interaction networks: Case study within the sixth ring road of Beijing, China. ISPRS Int. Geo-Inf. 2019, 8, 273. [CrossRef] 18. Yang, J.; Yi, D.; Liu, J.; Liu, Y.; Zhang, J. Spatiotemporal change characteristics of nodes’ heterogeneity in the directed and weighted spatial interaction networks: Case study within the sixth ring road of Beijing, China. Sustainability 2019, 11, 6359. [CrossRef] 19. Zhao, P.; Qin, K.; Ye, X.; Wang, Y.; Chen, Y. A trajectory clustering approach based on decision graph and data field for detecting hotspots. Int. J. Geogr. Inf. Sci. 2017, 31, 1101–1127. [CrossRef] 20. Xia, Z.; Li, H.; Chen, Y.; Liao, W. Identify and delimitate urban hotspot areas using a network-based spatiotemporal field clustering method. ISPRS Int. Geo-Inf. 2019, 8, 344. [CrossRef] 21. Anselin, L. Local indicators of spatial association—LISA. Geogr. Anal. 2010, 27, 93–115. [CrossRef] 22. Getis, A. The analysis of spatial association by use of distance statistics. Geogr. Anal. 2010, 24, 189–206. [CrossRef] 23. Reinau, K.H. Emerging hot spot analysis: Estimating the development in the speed of freight in a region using tracking data from logistics companies. In Proceedings of the Association of American Geographers Annual Meeting, Boston, MA, USA, 5–9 April 2017. 24. Harris, N.L.; Goldman, E.; Gabris, C.; Nordling, J.; Minnemeyer, S.; Ansari, S.; Hansen, M. Using spatial statistics to identify emerging hot spots of forest loss. Environ. Res. Lett. 2017, 12, 24012. [CrossRef] 25. Usman, M.; Sitanggang, I.S.; Syaufina, L. Hotspot distribution analyses based on peat characteristics using density-based spatial clustering. Procedia. Environ. Sci. 2015, 24, 132–140. [CrossRef] 26. Zhou, D.; Hong, R.; Xia, J. Identification of taxi pick-up and drop-off hotspots using the density-based spatial clustering method. In Proceedings of the CICTP 2017, Shanghai, China, 7–9 July 2017. 27. Pei, T.; Wang, W.; Zhang, H.; Ma, T.; Du, Y.; Zhou, C. Density-based clustering for data containing two types of points. Int. J. Geogr. Inf. Sci. 2015, 29, 175–193. [CrossRef] 28. Wang, T.; Ren, C.; Luo, Y.; Tian, J. NS-DBSCAN: A density-based clustering algorithm in network space. ISPRS Int. Geo-Inf. 2019, 8, 218. [CrossRef] 29. Zhang, P.; Deng, M.; Shi, Y.; Zhao, L. Detecting hotspots of urban residents’ behaviours based on spatio-temporal clustering techniques. GeoJournal 2017, 82, 923–935. [CrossRef] 30. Zhou, Y.; Fang, Z.; Thill, J.-C.; Li, Q.; Li, Y. Functionally critical locations in an urban transportation network: Identification and space–time analysis using taxi trajectories. Comput. Environ. Urban Syst. 2015, 52, 34–47. [CrossRef] 31. Deng, M.; Yang, X.; Shi, Y.; Gong, J.; Liu, Y.; Liu, H. A density-based approach for detecting network-constrained clusters in spatial point events. Int. J. Geogr. Inf. Sci. 2019, 33, 466–488. [CrossRef] 32. Li, D.; Du, Y. Artificial Intelligent with Uncertainty; Chapman and Hall, CRC: Boca Raton, FL, USA, 2007. Sustainability 2020, 12, 9662 20 of 20

33. Wang, S.; Gan, W.; Li, D.; Li, D. Data field for hierarchical clustering. Int. J. Data. Warehous. 2011, 7, 43–63. [CrossRef] 34. Li, C.; Ding, G.; Wang, D.; Li, Y.; Wang, S. Clustering by fast search and find of density peaks with data field. Chin. J. Electron. 2016, 25, 397–402. 35. Wu, T.; Qin, K. Image data field for homogeneous region based segmentation. Comput. Electr. Eng. 2012, 38, 459–470. [CrossRef] 36. Qin, K.; Zhou, Q.; Wu, T.; Xu, Y.Q. Hotspots Detection from Trajectory Data Based on Spatiotemporal Data Field Clustering; ISPRS—International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences: Wuhan, China, 2017. 37. Chen, L.; Gao, Y.; Zhu, D.; Yuan, Y.; Liu, Y. Quantifying the scale effect in geospatial big data using semi-variograms. PLoS ONE 2019, 14, e0225139. [CrossRef] 38. Openshaw, S. The Modifiable Areal Unit Problem. Concepts and Techniques in Modern Geography; Geo Books: Norwich, UK, 1984. 39. Rosin, P. Unimodal thresholding. Pattern Recognit. 2001, 34, 2083–2096. [CrossRef] 40. Chainey, S.; Tompson, L.; Uhlig, S. The utility of hotspot mapping for predicting spatial patterns of crime. Secur. J. 2008, 21, 4–28. [CrossRef] 41. Adepeju, M.; Rosser, G.; Cheng, T. Novel evaluation metrics for sparse spatio-temporal point process hotspot predictions—A crime case study. Int. J. Geogr. Inf. Sci. 2016, 30, 2133–2154. [CrossRef] 42. Yang, B.; Liu, L.; Lan, M.; Wang, Z.; Zhou, H.; Yu, H. A spatio-temporal method for crime prediction using historical crime data and transitional zones identified from nightlight imagery. Int. J. Geogr. Inf. Sci. 2020, 1740–1764. [CrossRef] 43. Hu, Y.; Wang, F.; Guin, C.; Zhu, H. A spatio-temporal kernel density estimation framework for predictive crime hotspot mapping and evaluation. Appl. Geogr. 2018, 99, 89–97. [CrossRef] 44. Wang, Y.; Gu, Y.; Dou, M.; Qiao, M. Using spatial semantics and interactions to identify urban functional regions. ISPRS Int. Geo-Inf. 2018, 7, 130. [CrossRef] 45. Liu, X.; Gong, L.; Gong, Y.; Liu, Y.Revealing travel patterns and city structure with taxi trip data. J. Transp.Geogr. 2015, 43, 78–90. [CrossRef] 46. Zhong, C.; Arisona, S.M.; Huang, X.; Batty, M.; Schmitt, G. Detecting the dynamics of urban structure through spatial network analysis. Int. J. Geogr. Inf. Sci. 2014, 28, 2178–2199. [CrossRef] 47. Shanmukhappa, T.; Ho, I.W.-H.; Tse, C.K. Spatial analysis of bus transport networks using network theory. Physica A 2018, 502, 295–314. [CrossRef] 48. Wang, Y.; Deng, Y.; Ren, F.; Zhu, R.; Wang, P.; Du, T.; Du, Q. Analysing the spatial configuration of urban bus networks based on the geospatial network analysis method. Cities 2020, 96, 102406. [CrossRef]

Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional aﬃliations.

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).