CN108280550B - Visual analysis method for comparing community division of public bicycle stations - Google Patents
Visual analysis method for comparing community division of public bicycle stations Download PDFInfo
- Publication number
- CN108280550B CN108280550B CN201810092381.9A CN201810092381A CN108280550B CN 108280550 B CN108280550 B CN 108280550B CN 201810092381 A CN201810092381 A CN 201810092381A CN 108280550 B CN108280550 B CN 108280550B
- Authority
- CN
- China
- Prior art keywords
- community
- station
- color
- sites
- division
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 230000000007 visual effect Effects 0.000 title claims abstract description 23
- 238000004458 analytical method Methods 0.000 title claims abstract description 22
- 238000009826 distribution Methods 0.000 claims abstract description 32
- 238000010586 diagram Methods 0.000 claims description 26
- 239000003086 colorant Substances 0.000 claims description 15
- 239000011159 matrix material Substances 0.000 claims description 6
- 230000004931 aggregating effect Effects 0.000 claims description 4
- 238000007781 pre-processing Methods 0.000 claims description 2
- 238000000034 method Methods 0.000 abstract description 21
- 238000005192 partition Methods 0.000 description 5
- 230000002776 aggregation Effects 0.000 description 3
- 238000004220 aggregation Methods 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 239000004576 sand Substances 0.000 description 2
- 238000000638 solvent extraction Methods 0.000 description 2
- 101100460704 Aspergillus sp. (strain MF297-2) notI gene Proteins 0.000 description 1
- 238000005054 agglomeration Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 238000006116 polymerization reaction Methods 0.000 description 1
- 239000004575 stone Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/40—Business processes related to the transportation industry
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0137—Measuring and analyzing of parameters relative to traffic conditions for specific applications
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Economics (AREA)
- General Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Theoretical Computer Science (AREA)
- Tourism & Hospitality (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Development Economics (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Instructional Devices (AREA)
Abstract
The invention discloses a novel visual analysis method for comparing community division of public bicycle stations, which is characterized in that a plurality of visual views are designed to show geographical distribution and borrowing and returning correlation among areas of the stations after the community division, and the commonalities and differences of different community division algorithm results are supported and compared in a visual mode; in order to show the geographical distribution condition of the divided sites more clearly, a color assignment strategy is provided, and the sites located in similar geographical areas are kept consistent in color as much as possible for community division results obtained by different algorithms; a community division comparison graph based on circular inclusion is designed, and is helpful for comparing which communities the sites are divided into under different methods. The method can visually display the result difference of different community division algorithms acting on the public bicycle network, is helpful for understanding the internal division mechanism of the algorithms, helps traffic managers to master the operation condition of the public bicycle system, and provides an auxiliary decision for vehicle scheduling and system management.
Description
Technical Field
The invention belongs to the technical field of traffic information, and particularly relates to a visual analysis method for comparing community division of public bicycle stations.
Background
The urban public bicycle system provides shared bicycle rental service, has the advantages of no pollution and environmental friendliness, and can effectively solve the problem of 'last kilometer' and relieve traffic pressure. The user can borrow and return the bicycle at any bicycle station in the city. The system records the user car borrowing and returning data which contains rich time and space information.
In order to find the operation rule of the system, the public bicycle system can be regarded as a network, the stations are nodes in the network, and the borrowing and returning amount of the users among the stations is the network edge. The existing method adopts a community discovery algorithm to divide a bicycle network, divides sites with close association in the network into the same community, and divides sites with sparse connection in the network into different communities. Community discovery algorithms can be broadly divided into several categories: (1) based on an optimization method, such as a luvain algorithm, the modularity is used as a standard for evaluating community partition quality to obtain network partition with the maximum modularity; (2) the method based on hierarchical clustering can be subdivided into a top-down split algorithm and a bottom-up agglomeration algorithm; (3) dynamics-based methods, which expose the structural properties of a network by analyzing the dynamics of the network, such as the infomap algorithm. The above classification is not strict, for example, the louvain algorithm belongs to a method based on hierarchical clustering and a method based on optimization.
The existing research only adopts a certain specific method to divide the bicycle network into communities, and cannot analyze how a certain station is divided into a certain community or compare the commonness and difference of the division results of different methods. Therefore, a visual analysis method needs to be designed, which not only can intuitively display the community division result and discover the association of the community, but also can compare the result difference of different methods and explore the reason why the site is divided into a certain community, so that the division result can better help a manager analyze the operation rule of the system and the renting characteristics of the site.
Disclosure of Invention
The invention aims to provide a visual analysis method for comparing community division results of public bicycle stations facing a public bicycle data set, and the visual analysis method is characterized in that a plurality of visual views are designed to show the geographical distribution, borrowing and returning correlation among areas and borrowing and returning quantity distribution of a single station of the stations after the community division, so that the visual comparison of the result difference of the stations divided into different communities by different methods is supported, and a manager can analyze the operation rule of a system and the renting characteristics of the stations. The specific technical scheme is as follows:
a visual analysis method for comparing community division of public bicycle stations comprises the following steps:
step 1: collecting public bicycle data and preprocessing the data;
step 2: constructing a public bicycle network, and representing the flow relation among public bicycle stations;
and step 3: based on the constructed public bicycle network, dividing the stations in the network into communities by adopting a community division algorithm;
and 4, step 4: designing a clustering scatter diagram, and visually displaying the geographical position distribution of the station community division by adopting a new color assignment strategy;
and 5: designing a clustering correlation diagram to visually display the bicycle flow correlation of the social interval;
step 6: designing a community division comparison graph based on circular inclusion, and comparing differences of dividing the sites into different communities by adopting different algorithms; designing a community division comparison graph comprises visualizing the result difference of community division by adopting a plurality of circular inclusion graphs, wherein in the circular inclusion graphs, a small round point represents a site, the small round point belonging to the same community is surrounded by a larger circle at the outer circle, and when a certain small round point is clicked, site information corresponding to the small round point is displayed.
Further, the method also comprises the following steps: when the analyst clicks a certain station in step 4, a car borrowing/returning amount distribution map of the station is displayed, and people who borrow cars from the most frequent car returning area and which area of the car borrowing area of the station return the cars to the central station, so that the analyst can understand the community division result.
Further, the step 1 comprises:
step 1.1: acquiring a public bike data set comprising:
the bicycle lease data table stores the lease information of all users, and each lease record journeyRec is expressed as follows:
journeyRec＝[userID,bikeID,cardNo,startStation,startTime,returnStation,returnTime]
wherein userID is user ID, bikeID is vehicle ID, cardNO is user card number, startStation is station for borrowing vehicle, startTime is time for borrowing vehicle, return station is station for returning vehicle, and return time is time for returning vehicle;
the station information table stores information of bicycle stations, each station record stationRec being represented as follows:
stationRec＝[stationID,stationName,stationAddr,longitude,latitude]
wherein, stationID is site ID, stationName is site name, stationAddr is site address, longitude is longitude and latitude is latitude;
step 1.2: aggregating the lease records, aggregating data of each site based on journeyRec, counting the car borrowing amount of a certain site in unit time by taking the hour as a unit, and storing, wherein each aggregated record is represented as journeyAggreRec:
journeyAggrRec＝[startDate,startHour,startStation,endStation,bikeNum]
wherein, startDate represents the date of borrowing the car, startHour represents the hour of borrowing the car, startState and endStation represent the station ID of borrowing and returning the car, bikeNum represents the number of cars which are borrowed and returned to endStation from the startState by the user in a certain hour (startHour) of a certain day (startDate);
step 1.3: and calculating the distance and the angle between every two stations, expressing the distance by using the linear distance between the two stations, and storing the result.
Further, the step 2 comprises:
building a public bicycle network GτThe network nodes are stations, the borrowing and returning relations of the vehicles among the stations are directed edges of the network, and tau represents a certain analysis time period. Gτ＝{N,Eτ}. N is a site set, Nie.N (i is more than or equal to 1 and less than or equal to N) represents a site, and N is the total number of sites; eτ＝{eijI is more than or equal to 1 and less than or equal to n, j is more than or equal to 1 and less than or equal to n is a directed adjacency matrix which represents an edge set under a tau time period, and each matrix element eijIs the value of from station n within the time period tauiBorrowing a car to a stop njThe number of cars returned.
Further, the step 3 comprises:
public bicycle network G adopting various community division algorithmsτDividing communities, dividing a series of sites with close borrowing and returning association into the same community, dividing the sites into k communities by the community division algorithm, and using C as a division resultτ＝{ciI is less than or equal to 1 and less than or equal to k. c. CiRepresenting a community, wherein the community comprises a plurality of sites, the clustering center position of the community is the average value of the longitude and latitude of all the sites in the community, and the division result is obtained
And
further, the step 4 comprises:
step 4.1: to pair
Randomly distributing a color value to each community in the system, and caching the community and the corresponding color relation;
step 4.2: for the
The cluster center of each community in
Finding a clustering center with the closest longitude and latitude, and taking the color value of the clustering center if the color value is the same as the color value of the clustering center
Is greater than the number of communities
The number of communities in the list is used for randomly giving color values to the rest communities;
step 4.3: designing a clustering scatter diagram, and displaying the geographical position distribution of the station community division; drawing a circular point on a map to represent the site based on the longitude and latitude of the site; based on the color assignment results of the step 4.1 and the step 4.2, sites belonging to the same community are drawn in the same color, so that areas with similar geographic positions keep consistent in color;
further, the step 5 comprises:
the sites belonging to the same community in the cluster association graph are abstracted to be a cluster center and drawn on the map by a round point. The size of the dots represents the number of the sites in the community; connecting the clustering centers by using arcs, wherein the colors and the thicknesses of the arcs simultaneously code the flow sizes of the social intervals; the thicker the arc, the larger the flow rate in the representative section. (ii) a Colors are coded by gradient colors, wherein darker colors indicate a larger flow rate correlation in the social section, and lighter colors indicate a smaller flow rate correlation in the social section.
Further, the step 6 comprises:
step 6.1: based on the result of the division
Generating a first circular containing graph, wherein the color of the small dots in the graph is consistent with the color assignment of the sites in the clustering scatter diagram;
step 6.2: based on the result of the division
Generating another circle containing diagram; the color assignment of the dots in the graph is consistent with that of the first circle containing graph, and the division of the dots in the graph is based on
Is divided.
Further, the step 7 comprises:
the borrowing/returning vehicle amount distribution diagram adopts radial layout, wherein the circle center represents a station to be analyzed, other stations related to a central station aggregate according to distance and angle, the aggregated information is visually encoded in the mode of an annular diagram, the angle encoding direction and the radius encoding distance of a ring are increased progressively from the center of the ring to the edge of the ring according to the step length of 1 km; each sector within the ring represents the number of bicycle borrowings/returns in that direction; the color of the sector codes the number of vehicles, and the darker the color is, the larger the number of the vehicles borrowed/returned in the current distance and angle range is; the sector of the borrowing distribution map is coded from light color to dark color, and the sector of the returning distribution map is coded from light color to dark color; when a certain sector area is clicked, a new map is displayed; in the map, a central station is represented by a five-pointed star icon, the station points belonging to the sector are displayed on the map, and the color of the station also encodes the flow size corresponding to the station.
The method is characterized by comprising the steps that a novel visual analysis method for comparing community division of public bicycle stations is provided, and the commonalities and differences of results of different community division algorithms are visually compared by designing a plurality of visual views to show the geographical distribution of the stations after the community division and the borrowing and returning relations among areas; in order to show the geographical distribution condition of the divided sites more clearly, a color assignment strategy is provided, and the sites located in similar geographical areas are kept consistent in color as much as possible for community division results obtained by different algorithms; a community division comparison graph based on circular inclusion is designed, and is helpful for comparing which communities the sites are divided into under different methods. The method can visually display the result difference of different community division algorithms acting on the public bicycle network, is helpful for understanding the internal division mechanism of the algorithms, helps traffic managers to master the operation condition of the public bicycle system, and provides an auxiliary decision for vehicle scheduling and system management.
Drawings
The invention will be further explained with reference to the drawings.
FIG. 1 is a flow chart of a visual analysis method according to the present invention.
Fig. 2a-2c are comparisons of results of community division of public bike sites using the method of the present invention, wherein fig. 2a uses an infomap algorithm, fig. 2b uses a combo algorithm, and fig. 2c uses a louvain algorithm.
Fig. 3a-3b are diagrams illustrating the analysis of the operation of different stations by means of the distribution diagram of the amount of vehicles borrowed/returned.
Detailed Description
The invention will be further explained with reference to the drawings.
As shown in fig. 1, the visual analysis method for comparing community division of public bicycle stations of the present invention comprises the following steps:
step 1: public bicycle data is collected and pre-processed.
Step 2: and constructing a public bicycle network, and representing the flow relation among public bicycle stations.
And step 3: based on the constructed public bicycle network, the stations in the network are divided into communities by adopting a community division algorithm.
And 4, step 4: and designing a clustering scatter diagram, and visually displaying the geographical position distribution of the station community division by adopting a new color assignment strategy.
And 5: and designing a clustering association graph to visually display the bicycle flow association of the social intervals.
Step 6: and designing a community division comparison graph based on circular inclusion, and comparing the difference of dividing the sites into different communities by adopting different algorithms.
When the analyst clicks a certain station in step 4, a car borrowing/returning amount distribution map of the station is displayed, and people who borrow cars from the most frequent car returning area and which area of the car borrowing area of the station return the cars to the central station, so that the analyst can understand the community division result.
The step 1 comprises the following steps:
step 1.1: a common bicycle data set is acquired. The bicycle rental data table stores the borrowing and returning information of all users. A lease record journeyRec is expressed as follows:
journeyRec＝[userID,bikeID,cardNo,startStation,startTime,returnStation,returnTime]
the userID is a user ID, the bikeID is a vehicle ID, the cardNO is a user card number, the startStation is a vehicle borrowing station, the startTime is a vehicle borrowing time, the return station is a vehicle returning station, and the return time is a vehicle returning time.
The station information table stores information of bicycle stations. One site record stationRec is expressed as follows:
stationRec＝[stationID,stationName,stationAddr,longitude,latitude]
wherein, stationID is site ID, stationName is site name, stationAddr is site address, longitude is longitude and latitude is latitude.
Step 1.2: the lease records are aggregated. If the original rent record journeyRec is directly processed, the data volume is huge, and the related information cannot be directly represented and visualized analysis can not be carried out. And performing data aggregation on each station based on journeyRec, counting the borrowing amount of a certain station in unit time by taking the hour as a unit, and storing the borrowing amount and the borrowing amount in the unit time to accelerate subsequent calculation. One record after polymerization is denoted as journeyaaggrec:
journeyAggrRec＝[startDate,startHour,startStation,endStation,bikeNum]
wherein, startDate represents the date of borrowing the car, startHour represents the hour of borrowing the car, startState and endStation represent the station IDs of borrowing and returning the car, bikeNum represents the number of cars which are borrowed and returned to endStation from the startState by the user in a certain hour (startHour) on a certain day (startDate).
Step 1.3: and calculating the distance and the angle between every two stations, expressing the distance by using the linear distance between the two stations, and storing the result for accelerating the generation of a subsequent visual view.
The step 2 comprises the following steps:
building a public bicycle network GτThe network nodes are stations, the borrowing and returning relations of the vehicles among the stations are directed edges of the network, and tau represents a certain analysis time period. Gτ＝{N,Eτ}. N is a site set, Nie.N (1 is more than or equal to i and less than or equal to N) represents a site, and N is the total number of sites. Eτ＝{eijAnd j is more than or equal to 1 and less than or equal to n, and j is more than or equal to 1 and less than or equal to n is a directed adjacency matrix which represents an edge set under a tau time period. Each matrix element eijIs the value of from station n within the time period tauiBorrowing a car to a stop njThe number of cars returned. The greater the traffic between the stations, eijThe greater the weight of the edge.
The step 3 comprises the following steps:
public bicycle network G adopting certain community division algorithmτAnd carrying out community division, and dividing a series of sites with close borrowing and returning association into the same community. Since the purpose of the invention is to compare the result difference of the community division algorithm of a plurality of public bike sites, a plurality of different division algorithms such as an infomap algorithm, a louvain algorithm, a combo algorithm and the like can be adopted. The community division algorithm divides the site into k communities, and the division result is Cτ＝{ciI is less than or equal to 1 and less than or equal to k. c. CiRepresenting a community containing a plurality of sites. The clustering center position of a community is the average value of the longitude and latitude of all the sites in the community. Since the network is divided by adopting various community division algorithms, the division results are distinguished by different subscripts, such as
And
respectively showing the division results obtained by the method A and the method B.
The step 4 comprises the following steps:
step 4.1: obtaining a partition result by adopting a first community partition algorithm
To pair
And randomly distributing a color value to each community, and caching the communities and the corresponding color relationship. These colors are the basis for subsequent color assignments.
Step 4.2: obtaining a division result by adopting another community division algorithm
For the
The aggregation of each community inClass center at
And finding a clustering center with the closest longitude and latitude, and taking a color value of the clustering center. If it is not
Is greater than the number of communities
The color value is randomly assigned to the rest communities. When there is a new division result, always
The result of (c) is used as a reference for color assignment.
Step 4.3: and designing a clustering scatter diagram to show the geographical position distribution of the station community division. Dots are drawn on a hundred degree map to represent a site based on the latitude and longitude of the site. And based on the color assignment results of the step 4.1 and the step 4.2, the sites belonging to the same community are drawn in the same color, so that the areas with similar geographic positions keep consistent color as much as possible.
The step 5 comprises the following steps:
and designing a clustering association graph to show the flow association of the social intervals. The sites belonging to the same community in the graph are abstracted to be a clustering center, and a round point is drawn on a Baidu map. The size of the dots indicates the number of sites within the community. The clustering centers are connected by arcs, and the color and thickness of the arcs simultaneously code the flow size of the community. The thicker the arc, the larger the flow rate in the representative section. The colors are coded by gradient colors (dark-medium-light), and the darker the color, the larger the flow rate correlation of the social section, and the lighter the color, the smaller the flow rate correlation of the social section.
The step 6 comprises the following steps:
and designing a community division comparison graph, and visualizing the result difference of the community division by adopting a plurality of circular containing graphs. In the circular containing diagram, a small circle point represents a station, and the small circle point belonging to the same community is surrounded by a larger circle at the outer circle. When a certain small dot is clicked, the site information corresponding to the small dot is displayed. The method specifically comprises the following steps:
step 6.1: based on the result of the division
A first circle containing map is generated. The color of the dots in the graph is consistent with the color assignment of the sites in the cluster scatter diagram.
Step 6.2: based on the result of the division
Another circle containing figure is generated. In the figure, the color assignment of the dots is consistent with that of the first circle containing figure, and the site division (the surrounding situation of the outer circle) of the dots is according to
Is divided. Can be observed according to the color of the dots in the figure
The sites are divided into the same community, and after being subdivided by another algorithm, the sites are distributed in different communities. When a new division result exists, the color assignment of the dots in the generated new circular containment map is always summed
The result of (c) is kept consistent, and the outer circle of the small dots is specified according to the new site division result.
The step 7 comprises the following steps:
when the analyst clicks a certain station in step 4, a loan/return amount distribution map of the station is displayed for analyzing the use condition of each station. The borrowing amount distribution map represents the number of vehicles borrowed from the center station and returned to the areas located at different azimuths and distances. The return vehicle amount distribution chart shows the number of vehicles which are borrowed from the peripheral area and returned to the center station. The two distribution maps are in radial layout, the circle center represents a station to be analyzed, other stations related to the central station are aggregated according to distance and angle, and the aggregated information is visually encoded in the mode of an annular map. The angle of the ring encodes the direction, and the radius encodes the distance. Starting from the center of the ring towards the edge of the ring, the steps are increased by 1 km. Each sector within the ring represents the number of bicycle borrowings/returns in that direction. The color of the sector encodes the number of vehicles, and the darker the color, the greater the number of borrowing/returning vehicles within the current distance and angle range. The sector of the borrowed vehicle amount distribution diagram is coded from light color to dark color, and the sector of the borrowed vehicle amount distribution diagram is coded from light color to dark color. When a sector is clicked, a new map is displayed. In the map, a central station is represented by a five-pointed star icon, the bicycle stations belonging to the sector are displayed on the map, and the color of the station also encodes the flow size corresponding to the station.
Examples
Fig. 2 shows a comparison of the results of community division for the Hangzhou city public bike station. For the result obtained by each algorithm, the three graphs from left to right are a cluster scatter diagram, a cluster association graph and a community division comparison graph respectively. Fig. 2a, 2b, 2c are the results of using the infomap, combo and louvain algorithms, respectively. And finding out that the sites with adjacent positions are divided into the same community from the clustering scatter diagram. The clustering correlation diagram shows that most of the social intervals are not close, the connecting lines are light-colored, only part of the social intervals are strong in correlation, and the connecting lines are dark-colored or light-colored. Since the infomap algorithm is the first algorithm to be adopted, the site colors in the community partition comparison graph are based on the assignment result of the algorithm. The community division comparison graph shows that the number of communities is the largest and the number of isolated points is more as a result of the infomap division. The result of the combo algorithm is consistent with the infomap, and partial small areas are merged together. The partitioning result of the luvain algorithm is the coarsest, except that the results of 3 regions and the infomap method are basically consistent, other regions are combined into 4 large regions, and therefore a circle in the community partitioning comparison graph comprises a plurality of points with different colors. From the comparison results of the algorithms, it can be known that the three algorithms can be identified because the lower sand, the Binjiang and the southwest city are relatively small in association with other areas because the lower sand, the Binjiang and the southwest city are relatively far away from other areas in the geographical position. The louvain algorithm tends to divide the regions together, i.e., there is a resolution limit (resolution limit) problem.
Fig. 3 shows the distribution diagram of the amount of the borrowed/returned vehicles at different stations. The "10-14 pier stone bridge" station in fig. 3a is a singular station divided into a single community by using the infomap algorithm, and it can be known from its borrowing/returning amount distribution map that a station is divided into a community because the station is not frequently used (the borrowing amount is 0, and the returning amount is also very small). The "Langler" site (labeled in FIG. 2 (a)) shown in FIG. 3b, although very far from other sites in the community, is still classified in the same community, and as can be seen from the borrowing/returning volume distribution map, most of the traffic goes to the northwest direction 3km-4km away. When the deepest sector area in the vehicle borrowing amount distribution map is clicked, the map on the right side is obtained. From the map it can be seen that most people borrowing vehicles from the station are riding their bicycles to the opposite bank of the west lake.
Claims (6)
1. A visual analysis method for comparing community division of public bicycle stations comprises the following steps:
step 1: collecting public bicycle data and preprocessing the data;
step 2: constructing a public bicycle network, and representing the flow relation among public bicycle stations;
and step 3: based on the constructed public bicycle network, dividing the stations in the network into communities by adopting a community division algorithm;
and 4, step 4: designing a clustering scatter diagram, and visually displaying the geographical position distribution of the station community division by adopting a new color assignment strategy;
and 5: designing a clustering correlation diagram to visually display the bicycle flow correlation of the social interval;
step 6: designing a circular-contained community division comparison graph, and comparing differences of dividing the sites into different communities by adopting different algorithms; the community division comparison graph adopts a plurality of circular inclusion graphs to visualize the difference of the community division results, in the circular inclusion graphs, one small circle point represents one site, the small circle point belonging to the same community is surrounded by a larger circle at the outer circle, and when a certain small circle point is clicked, the site information corresponding to the small circle point is displayed;
the step 3 comprises the following steps:
public bicycle network G adopting various community division algorithmsτDividing communities, dividing a series of sites with close borrowing and returning association into the same community, dividing the sites into k communities by a plurality of community division algorithms, and using C as a division resultτ＝{ci1 is less than or equal to i and less than or equal to k; c. CiRepresenting a community, wherein the community comprises a plurality of sites, the clustering center position of the community is the average value of the longitude and latitude of all the sites in the community, and the division result is obtained
The step 4 comprises the following steps:
step 4.1: to pair
Randomly distributing a color value to each community in the system, and caching the community and the corresponding color relation;
step 4.2: for the
The cluster center of each community in
Finding a clustering center with the closest longitude and latitude, and taking the color value of the clustering center if the color value is the same as the color value of the clustering center
Is greater than the number of communities
The number of communities in the list is used for randomly giving color values to the rest communities;
step 4.3: designing a clustering scatter diagram, and displaying the geographical position distribution of the station community division; drawing a circular point on a map to represent the site based on the longitude and latitude of the site; based on the color assignment results of the step 4.1 and the step 4.2, sites belonging to the same community are drawn in the same color, so that areas with similar geographic positions keep consistent in color;
the step 6 comprises the following steps:
step 6.1: based on the result of the division
Generating a first circular containing graph, wherein the color of the small dots in the graph is consistent with the color assignment of the sites in the clustering scatter diagram;
step 6.2: based on the result of the division
Generating another circle containing diagram; the color assignment of the dots in the figure is consistent with the color of the first circle containing the figure, and the division of the dots in the site is according to
Is divided.
2. A visual analysis method for comparing community divisions of public bicycle sites as claimed in claim 1, wherein:
further comprising step 7: when the analyst clicks a certain station in step 4, a car borrowing/returning amount distribution map of the station is displayed, and people who borrow cars from the most frequent car returning area and which area of the car borrowing area of the station return the cars to the central station, so that the analyst can understand the community division result.
3. A visual analysis method for comparing community divisions of public bicycle sites as claimed in claim 1, wherein: the step 1 comprises the following steps:
step 1.1: acquiring a public bike data set comprising:
the bicycle lease data table stores the lease information of all users, and each lease record journeyRec is expressed as follows:
journeyRec＝[userID,bikeID,cardNo,startStation,startTime,returnStation,returnTime]
wherein userID is user ID, bikeID is vehicle ID, cardNO is user card number, startStation is station for borrowing vehicle, startTime is time for borrowing vehicle, return station is station for returning vehicle, and return time is time for returning vehicle;
the station information table stores information of bicycle stations, each station record stationRec being represented as follows:
stationRec＝[stationID,stationName,stationAddr,longitude,latitude]
wherein, stationID is site ID, stationName is site name, stationAddr is site address, longitude is longitude and latitude is latitude;
step 1.2: aggregating the lease records, aggregating data of each site based on journeyRec, counting the car borrowing amount of a certain site in unit time by taking the hour as a unit, and storing, wherein each aggregated record is represented as journeyAggreRec:
juoureyaggrrec [ startDate, starthome, startStation, endStation, bikeNum ] where startDate denotes the date of car borrowing, starthome denotes the hour of car borrowing, startStation and endStation denote the station IDs of car borrowing and returning, bikeNum denotes the number of vehicles the user borrows from the startStation and returns to the endStation within a certain hour (starthome) of a certain day (startDate);
step 1.3: and calculating the distance and the angle between every two stations, expressing the distance by using the linear distance between the two stations, and storing the result.
4. A visual analysis method for comparing community divisions of public bicycle sites as claimed in claim 1, wherein: the step 2 comprises the following steps:
building a public bicycle network GτThe network nodes are stations, the borrowing and returning relation of the vehicles among the stations is a directed edge of the network, and tau represents a certain analysis time period; gτ＝{N,EτN is site set, NiE.g. N represents a site, i is more than or equal to 1 and less than or equal to N, and N is the total number of sites; eτ＝{eijI is more than or equal to 1 and less than or equal to n, j is more than or equal to 1 and less than or equal to n is a directed adjacency matrix which represents an edge set under a tau time period, and each matrix element eijIs the value of from station n within the time period tauiBorrowing a car to a stop njThe number of cars returned.
5. A visual analysis method for comparing community divisions of public bicycle sites as claimed in claim 1, wherein: the step 5 comprises the following steps:
the sites belonging to the same community in the clustering association graph are abstracted to be a clustering center and drawn on the map by a round point; the size of the dots represents the number of the sites in the community; connecting the clustering centers by using arcs, wherein the colors and the thicknesses of the arcs simultaneously code the flow sizes of the social intervals; the thicker the arc, the larger the flow rate between the representative areas; colors are coded by gradient colors, wherein darker colors indicate a larger flow rate correlation in the social section, and lighter colors indicate a smaller flow rate correlation in the social section.
6. A visual analysis method for comparing community divisions of public bicycle sites as claimed in claim 2, wherein: the step 7 comprises the following steps:
the borrowing/returning vehicle amount distribution diagram adopts radial layout, wherein the circle center represents a station to be analyzed, other stations related to a central station aggregate according to distance and angle, the aggregated information is visually encoded in the mode of an annular diagram, the angle encoding direction and the radius encoding distance of a ring are increased progressively from the center of the ring to the edge of the ring according to the step length of 1 km; each sector within the ring represents the number of bicycle borrowings/returns in that direction; the color of the sector codes the number of vehicles, and the darker the color is, the larger the number of the vehicles borrowed/returned in the current distance and angle range is; the sector of the borrowing distribution map is coded from light color to dark color, and the sector of the returning distribution map is coded from light color to dark color; when a certain sector area is clicked, a new map is displayed; in the map, a central station is represented by a five-pointed star icon, the station points belonging to the sector are displayed on the map, and the color of the station also encodes the flow size corresponding to the station.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810092381.9A CN108280550B (en) | 2018-01-30 | 2018-01-30 | Visual analysis method for comparing community division of public bicycle stations |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810092381.9A CN108280550B (en) | 2018-01-30 | 2018-01-30 | Visual analysis method for comparing community division of public bicycle stations |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108280550A CN108280550A (en) | 2018-07-13 |
| CN108280550B true CN108280550B (en) | 2020-06-09 |
Family
ID=62805811
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810092381.9A Active CN108280550B (en) | 2018-01-30 | 2018-01-30 | Visual analysis method for comparing community division of public bicycle stations |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108280550B (en) |
Families Citing this family (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109272170B (en) * | 2018-10-11 | 2019-08-30 | 北京市交通信息中心 | A kind of traffic zone dividing system based on Louvain algorithm |
| CN109254984B (en) * | 2018-10-16 | 2020-10-23 | 杭州电子科技大学 | Visual analysis method for perceiving city dynamic structure evolution law based on OD data |
| CN109447143A (en) * | 2018-10-22 | 2019-03-08 | 南京航空航天大学 | Air traffic region Partition Analysis system |
| CN109447144A (en) * | 2018-10-22 | 2019-03-08 | 南京航空航天大学 | The clustering method divided for air traffic region |
| CN109903125A (en) * | 2019-01-28 | 2019-06-18 | 东南大学 | Shared bicycle based on OD data borrow also with park spatial and temporal distributions method for visualizing |
| CN110119482B (en) * | 2019-05-13 | 2021-04-06 | 杭州电子科技大学 | Visual analysis method for crowd travel mode based on POI and multi-source mobile data set |
| CN113052408B (en) * | 2019-12-10 | 2024-02-23 | 杭州海康威视数字技术股份有限公司 | Method and device for community aggregation |
| CN111489064B (en) * | 2020-03-27 | 2023-05-12 | 湖南大学 | Df-PBS (direct-flow-coupled system) -oriented public bicycle station dynamic planning method and system |
| CN112491639B (en) * | 2020-09-29 | 2022-10-18 | 南京大学 | Anycast region division measuring method based on domain name resolution |
| CN112669595B (en) * | 2020-12-10 | 2022-07-01 | 浙江大学 | Network taxi booking flow prediction method based on deep learning |
| CN112784423A (en) * | 2021-01-28 | 2021-05-11 | 河北师范大学 | Urban area feature analysis method based on complex network |
| CN112907103B (en) * | 2021-03-09 | 2022-08-12 | 北京邮电大学 | Method for sharing dynamic supply and demand balance of single vehicle |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103077157A (en) * | 2013-01-22 | 2013-05-01 | 清华大学 | Method and device for visualizing text set similarity |
| US8700756B2 (en) * | 2011-05-03 | 2014-04-15 | Xerox Corporation | Systems, methods and devices for extracting and visualizing user-centric communities from emails |
| CN105654387A (en) * | 2015-03-17 | 2016-06-08 | 重庆邮电大学 | Time-varying network community evolution visualization method introducing quantitative index |
| CN105959132A (en) * | 2016-04-21 | 2016-09-21 | 重庆邮电大学 | Community layout visualization method facing large scale network |
| CN106296350A (en) * | 2016-08-04 | 2017-01-04 | 杭州电子科技大学 | A kind of visual analyzing city public bicycle system borrows the method for also pattern |
| CN106294679A (en) * | 2016-08-08 | 2017-01-04 | 大连理工大学 | A kind of method for visualizing carrying out website cluster based on subway data |
| CN106874349A (en) * | 2016-12-26 | 2017-06-20 | 深圳市位和科技有限责任公司 | Multidimensional data analysis method and system based on interactive visual |
-
2018
- 2018-01-30 CN CN201810092381.9A patent/CN108280550B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8700756B2 (en) * | 2011-05-03 | 2014-04-15 | Xerox Corporation | Systems, methods and devices for extracting and visualizing user-centric communities from emails |
| CN103077157A (en) * | 2013-01-22 | 2013-05-01 | 清华大学 | Method and device for visualizing text set similarity |
| CN105654387A (en) * | 2015-03-17 | 2016-06-08 | 重庆邮电大学 | Time-varying network community evolution visualization method introducing quantitative index |
| CN105959132A (en) * | 2016-04-21 | 2016-09-21 | 重庆邮电大学 | Community layout visualization method facing large scale network |
| CN106296350A (en) * | 2016-08-04 | 2017-01-04 | 杭州电子科技大学 | A kind of visual analyzing city public bicycle system borrows the method for also pattern |
| CN106294679A (en) * | 2016-08-08 | 2017-01-04 | 大连理工大学 | A kind of method for visualizing carrying out website cluster based on subway data |
| CN106874349A (en) * | 2016-12-26 | 2017-06-20 | 深圳市位和科技有限责任公司 | Multidimensional data analysis method and system based on interactive visual |
Non-Patent Citations (1)
| Title |
|---|
| "大规模社交网络社区发现及可视化算法";赵润乾，吴渝，陈昕;《计算机辅助设计与图形学学报》;20170228;第29卷(第2期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108280550A (en) | 2018-07-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108280550B (en) | Visual analysis method for comparing community division of public bicycle stations | |
| Alexander et al. | Assessing the impact of real-time ridesharing on urban traffic using mobile phone data | |
| CN110298500B (en) | Urban traffic track data set generation method based on taxi data and urban road network | |
| Bao et al. | Spatial analysis of bikeshare ridership with smart card and POI data using geographically weighted regression method | |
| CN104902438B (en) | A kind of statistical method and its system based on mobile communication terminal analysis passenger flow characteristic information | |
| CN108170793A (en) | Dwell point analysis method and its system based on vehicle semanteme track data | |
| CN105206046A (en) | Big-data-based taxi service station site selection and feasibility evaluation method | |
| CN110716935A (en) | Track data analysis and visualization method and system based on online taxi appointment travel | |
| Mondal et al. | Crime hotspot detection using statistical and geospatial methods: a case study of Pune City, Maharashtra, India | |
| CN113988476A (en) | Dynamic assessment prediction method for road transportation safety risk | |
| CN106651027A (en) | Internet regular bus route optimization method based on social network | |
| CN107590766A (en) | A kind of method of discrimination of the land used combination form related to road traffic accident risk | |
| Yu et al. | Discovering spatiotemporal patterns and urban facilities determinants of cycling activities in Beijing | |
| CN112418696A (en) | Method and device for constructing urban traffic dynamic knowledge map | |
| Wang et al. | Digital roadway interactive visualization and evaluation network applications to WSDOT operational data usage. | |
| CN111125285A (en) | Animal geographic zoning method based on species spatial distribution relation | |
| CN112365595A (en) | Analysis method for identifying traffic accident multi-point based on alarm data point element | |
| Liu et al. | Exploring the spatially heterogeneous effect of the built environment on ride-hailing travel demand: A geographically weighted quantile regression model | |
| Nasri et al. | A multi-dimensional multi-level approach to measuring the spatial structure of US metropolitan areas | |
| Sun et al. | Analyzing spatiotemporal daily travel source carbon emissions based on taxi trajectory data | |
| Guo et al. | Spatiotemporal heterogeneity of the shared e-scooter–public transport relationships in Stockholm and Helsinki | |
| Francis et al. | Unsupervised feature extraction of aerial images for clustering and understanding hazardous road segments | |
| Verma et al. | Extracting spatiotemporal commuting patterns from public transit data | |
| Steenberghen et al. | Support study on data collection and analysis of active modes use and infrastructure in Europe | |
| CN117391335A (en) | Urban individual travel chain reconstruction method under MaaS environment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |