CN115994313B - Crowd movement modeling method and device based on access place clustering - Google Patents
Crowd movement modeling method and device based on access place clustering Download PDFInfo
- Publication number
- CN115994313B CN115994313B CN202310281628.2A CN202310281628A CN115994313B CN 115994313 B CN115994313 B CN 115994313B CN 202310281628 A CN202310281628 A CN 202310281628A CN 115994313 B CN115994313 B CN 115994313B
- Authority
- CN
- China
- Prior art keywords
- access
- individual
- individuals
- place
- exploration
- 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
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Abstract
The application discloses a crowd movement modeling method and device based on access place clustering, which belong to the field of track prediction, and the method comprises the following steps: acquiring access sites and access site clusters of a plurality of individuals in a crowd based on historical crowd movement track data; extracting crowd moving features according to access places, access place clusters and historical crowd moving tracks, and generating model parameters of a crowd moving model according to the crowd moving features, wherein the crowd moving features comprise conditional probability distribution among all access places of individuals and access frequency sequences of the access place clusters to which the individuals belong; inputting the set number of the simulated individuals and the target time length into a crowd movement model to generate movement tracks of a plurality of the simulated individuals in the target time length. According to the method and the device, not only can the movement track of the individual be predicted, but also the access mode of the individual between access site clusters and inside the access site clusters can be predicted, the inherent connection between the access site and the access site clusters is repeated, and the defect of the existing model is overcome.
Description
Technical Field
The application relates to the technical field of track prediction, in particular to a crowd movement modeling method and device based on access place clustering.
Background
In recent decades, as the availability of large crowd movement track data such as call record data, mobile device positioning data, social media card punching data and the like is higher and higher, crowd movement track processing methods and crowd movement modeling methods are rapidly developed, and more applications are obtained in solving the practical problems of urban planning, traffic engineering, disease transmission control and the like. The crowd movement modeling method can simulate the movement tracks of individuals in the city, and enables the movement tracks to accord with the basic rules of crowd movement found in the empirical data on the individual level and the crowd level, so that key characteristics of the urban crowd movement are reproduced. If the movement rule of the crowd cannot be accurately predicted, key characteristics of the movement of the crowd are reproduced, and accurate references cannot be provided for reasonable formulation of related policies.
A series of fine-grained models of crowd movement have been proposed in prior studies, with Exploration and Preference Return (EPR) models and variants thereof, such as d-EPR models and PEPR models, being the most influential and most widely used models of crowd movement. However, in the existing crowd movement model, individuals move between independent places, and the spatial distribution characteristics of the access places and the internal connection of the access places are completely ignored. In reality, however, access sites of an individual form clusters in physical space, which have rich semantic features that are generally associated with a certain class of behavior of the individual and constitute various types of activity spaces (e.g., living activity space, work activity space, entertainment activity space, etc.) in the daily life of the individual, and influence the movement of the individual. Modeling is performed on crowd movement by using the view angles of the access location clusters instead of the view angles of the access locations independent of each other, so that on one hand, the movement rule of individuals among the access location clusters and inside each access location cluster can be accurately reproduced, and on the other hand, the movement behaviors of the individuals can be linked with urban spaces, and the interaction process of the individuals and various urban spaces can be understood.
In summary, how to consider the influence of access location clustering on individual movement is one of the problems to be solved by the current crowd movement modeling method in crowd movement modeling.
Disclosure of Invention
The crowd movement modeling method and device based on the visit site clusters can not only predict the movement track of an individual, but also predict the visit modes of the individual between the visit site clusters and inside the visit site clusters, and reproduce the inherent connection between the visit site and the visit site clusters, so that the defect of the existing model is overcome.
An embodiment of a first aspect of the present application provides a crowd movement modeling method based on access location clustering, including the following steps: acquiring access sites and access site clusters of a plurality of individuals in a crowd based on historical crowd movement track data; extracting crowd movement features according to the access places, the access place clusters and the historical crowd movement tracks, and generating model parameters of a crowd movement model according to the crowd movement features, wherein the crowd movement features comprise access frequency sequences of all access places of individuals and conditional probability distribution among access frequency sequences of the access place clusters to which the individuals belong; inputting the set number of simulated individuals and the target duration into the crowd movement model, and generating movement tracks of a plurality of simulated individuals in the target duration through the crowd movement model.
Optionally, in an embodiment of the present application, the acquiring access locations and access location clusters of the plurality of individuals in the crowd based on the historical crowd movement trajectory data includes: determining a plurality of position record points in an individual track according to the historical crowd movement track data, calculating the current speeds of the position record points, and marking the position record points with the current speeds less than or equal to a preset speed threshold as stay position record points; clustering all the stay position record points, calculating single stay time of an individual in each position record point cluster, and determining access places of the individual according to the single stay time, wherein coordinates of the access places are centers of the corresponding position record point clusters; clustering all access places of the individuals to obtain the access place clusters.
Optionally, in one embodiment of the present application, extracting crowd movement features according to the visit location, the visit location cluster, and the historical crowd movement trajectory includes: calculating the residence time of each individual visit according to the visit sequence of each individual visit to each visit place, the arrival time and the departure time of each visit, and counting the residence time distribution of all individuals; calculating the moving step length between each visit of each individual according to the visit sequence of each individual to each visit place, and counting the moving step length distribution of all the individuals; counting the exploration probability of the individual under different access places according to the historical movement track of the individual; according to the historical movement track of the individual, counting random exploration probabilities of the individual under different access places; sorting all access places of the individual and the access frequencies of all access place clusters, and calculating the access frequency sorting of all access places of the individual and the conditional probability distribution among the access frequency sorting of the access place clusters to which the access places belong.
Optionally, in an embodiment of the present application, generating, by the crowd movement model, movement trajectories of a plurality of simulated individuals within the target duration includes: selecting an initial position of each simulation individual, and setting the initial time to be 0; extracting the residence time of the simulated individuals according to the crowd movement characteristics, and carrying out exploration behaviors or return behaviors according to the exploration probability of the simulated individuals; when the simulated individual is the exploration behavior, carrying out random exploration behavior or intra-cluster exploration behavior according to the random exploration probability of the simulated individual; when the simulated individuals conduct random exploration behaviors, extracting movement step length according to the crowd movement characteristics, and determining any exploration place of the simulated individuals outside the visit place cluster in the movement step length range for visit; when the simulated individuals perform intra-cluster exploration behaviors, determining a to-be-visited place cluster of the simulated individuals according to the visit frequency sequence of the visit place cluster, extracting the moving step length according to the crowd moving characteristics, and determining any exploration place of the simulated individuals in the to-be-visited place cluster and visiting the exploration place within the moving step length range; when the simulated individual is in a return behavior, determining a to-be-accessed place cluster of the simulated individual according to the access frequency of the access place cluster, and determining and accessing the to-be-accessed place of the simulated individual according to the access frequency of the access place in the to-be-accessed place cluster; calculating whether the total residence time of the simulation individual in a plurality of access places is larger than or equal to the target duration, if the total residence time is larger than or equal to the target duration, generating a moving track of the simulation individual according to the positions and the access sequence of the plurality of access places, and if the total residence time is smaller than the target duration, continuing to search or return according to the exploration probability of the simulation individual until the total residence time is larger than or equal to the target duration.
An embodiment of a second aspect of the present application provides a crowd movement modeling apparatus based on access location clustering, including: the acquisition module is used for acquiring access sites and access site clusters of a plurality of individuals in the crowd based on the historical crowd movement track data; the parameter generation module is used for extracting crowd movement characteristics according to the access places, the access place clusters and the historical crowd movement tracks, and generating model parameters of a crowd movement model according to the crowd movement characteristics, wherein the crowd movement characteristics comprise access frequency sequences of all access places of individuals and conditional probability distribution among the access frequency sequences of the access place clusters to which the individual access places belong; the track generation module is used for inputting the set number of the simulated individuals and the target time length into the crowd movement model, and generating movement tracks of the plurality of the simulated individuals in the target time length through the crowd movement model.
Optionally, in one embodiment of the present application, the acquiring module includes: the identification unit is used for determining a plurality of position record points in the individual track according to the historical crowd moving track data, calculating the current speed of the position record points, and marking the position record points with the current speed less than or equal to a preset speed threshold as stay position record points; the first clustering unit is used for clustering all the stay position record points, calculating single stay time of an individual in each position record point cluster, and determining access places of the individual according to the single stay time, wherein the coordinates of the access places are the centers of the corresponding position record point clusters; and the second clustering unit is used for clustering all the access places of the individuals to obtain the access place clusters.
Optionally, in one embodiment of the present application, the parameter generating module includes: the first extraction unit is used for calculating the stay time of each access of each individual according to the access sequence of the individual to each access place, the arrival time and the departure time of each access, and counting the stay time distribution of all the individuals; the second extraction unit is used for calculating the moving step length between each visit of each individual according to the visit sequence of the individual to each visit place and counting the moving step length distribution of all the individuals; the third extraction unit is used for counting the exploration probability of the individuals under different access places according to the historical movement track of the individuals; a fourth extraction unit, configured to count random exploration probabilities of individuals under different access location numbers according to historical movement tracks of the individuals; and a fifth extraction unit, configured to rank all access sites of the individual and access frequencies of all access site clusters, and calculate a conditional probability distribution between the access frequency ranks of all access sites of the individual and the access frequency ranks of the access site clusters to which the access sites belong.
Optionally, in one embodiment of the present application, the track generating module includes: the initial unit is used for selecting the initial position of each simulation individual and setting the initial time to be 0; the exploration unit is used for extracting the residence time of the analog individuals according to the crowd movement characteristics and carrying out exploration behaviors or return behaviors according to the exploration probability of the analog individuals; the exploration unit is further used for conducting random exploration behaviors or intra-cluster exploration behaviors according to random exploration probabilities of the simulated individuals when the simulated individuals are exploration behaviors; the exploration unit is further used for extracting a moving step length according to the crowd moving characteristics when the simulated individuals conduct random exploration behaviors, and determining any exploration place of the simulated individuals outside the visit place cluster and visiting the exploration place within the moving step length range; the exploration unit is further used for determining a to-be-visited place cluster of the analog individual according to the visit frequency sequence of the visit place cluster when the analog individual performs intra-cluster exploration behaviors, extracting the moving step length according to the crowd moving characteristics, and determining any exploration place of the analog individual in the to-be-visited place cluster and visiting the exploration place within the moving step length range; the exploration unit is further used for determining a to-be-accessed place cluster of the simulated individual according to the access frequency of the access place cluster when the simulated individual is in return behavior, and determining and accessing the to-be-accessed place of the simulated individual according to the access frequency of the access place in the to-be-accessed place cluster; the output unit is used for calculating whether the total residence time of the simulation individual at a plurality of access places is greater than or equal to the target duration, if the total residence time is greater than or equal to the target duration, generating a moving track of the simulation individual according to the positions and the access sequence of the plurality of access places, and if the total residence time is less than the target duration, continuing to perform exploration or return according to the exploration probability of the simulation individual until the total residence time is greater than or equal to the target duration.
According to the crowd movement modeling method and device based on the access location clustering, the access location clustering is identified from the individual movement track, the crowd movement characteristics are obtained by utilizing the movement behaviors of the individuals on the access location clustering level, the stay time, the movement step length, the variation situation of the number of the access locations followed by the exploration probability of the individuals and the variation situation of the number of the access locations followed by the random exploration probability of the individuals are utilized, the access frequency ordering of all the access locations of the individuals and the conditional probability distribution among the access frequency ordering of the access location clustering to which the access location is belonged are creatively considered, and the intrinsic relation between the access locations and the access location clustering is reflected by considering the conditional probability distribution; two important features of individual movement are characterized: the access sites belonging to the high-frequency access site cluster have higher probability of being accessed by individuals, and the access sites with higher access frequency are more likely to appear in the high-frequency access site cluster; and the existing model is proved to have obvious defects in describing the association relationship between the access places and the access place clusters to which the access places belong. The two-stage decision process of the movement of the individual under the view angle of the visit site clusters is further introduced, so that not only can the movement track of the individual be predicted, but also the visit modes of the individual between the visit site clusters and inside the visit site clusters can be predicted, the inherent connection between the visit site and the visit site clusters can be reproduced, and the defect of the existing model can be overcome.
Additional aspects and advantages of the application will be set forth in part in the description which follows and, in part, will be obvious from the description, or may be learned by practice of the application.
Drawings
The foregoing and/or additional aspects and advantages of the present application will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings, in which:
FIG. 1 is a flow chart of a crowd movement modeling method based on visit site clustering according to an embodiment of the present application;
FIG. 2 is a flowchart for obtaining access locations and access location clusters for a plurality of individuals in a crowd provided according to an embodiment of the present application;
FIG. 3 is a flow chart for extracting crowd moving features provided in accordance with an embodiment of the present application;
FIG. 4 is a flowchart of an individual movement trajectory simulation method provided according to an embodiment of the present application;
FIG. 5 is a schematic diagram of a crowd movement model provided according to an embodiment of the application;
FIG. 6 is an exemplary diagram of a crowd movement modeling apparatus based on access location clustering in accordance with an embodiment of the present application.
Detailed Description
Embodiments of the present application are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals refer to the same or similar elements or elements having the same or similar functions throughout. The embodiments described below by referring to the drawings are exemplary and intended for the purpose of explaining the present application and are not to be construed as limiting the present application.
Fig. 1 is a flowchart of a crowd movement modeling method based on access location clustering according to an embodiment of the application.
As shown in fig. 1, the crowd movement modeling method based on access place clustering comprises the following steps:
in step S101, access points and access point clusters of a plurality of individuals in a crowd are acquired based on historical crowd movement trajectory data.
The historical crowd movement track data can be crowd movement data of a specific city in a past period of time, and the crowd movement track data comprises places where individuals go, arrival and departure time and the like. According to the crowd moving track data, access places of a plurality of individuals in the crowd can be obtained.
In one embodiment of the present application, as shown in fig. 2, as a way of obtaining access locations and access location clusters of a plurality of individuals in a crowd, the method includes:
s1011, determining a plurality of position record points in the individual track according to the historical crowd movement track data, calculating the current speed of the position record points, and marking the position record points with the current speed less than or equal to a preset speed threshold as stay position record points.
Specifically, the embodiment of the application can extract the track of each individual in the historical crowd moving track data, and identify the position record point which is in a stay state in the track of the individual, wherein the track of the individual consists of a plurality of position record points with time stamps, and the states of the position record points comprise the stay state and the moving state.
Defining the trajectory of an individual as
Definitions->
Is a time-stamped location record. The state of the position record can be divided into a stay state and a moving state, and the identification is performed
The identification method is as follows:
setting a speed threshold
(recommended value is 1.3 m/s), scanning backward from the first position recording point, when position recording point +.>
Satisfy->
When the position recording point is in the stop state, the position recording point is in the moving state.
S1012, clustering all the stay position record points, calculating single stay time of the individual in each position record point cluster, and determining access places of the individual according to the single stay time, wherein coordinates of the access places are centers of the corresponding position record point clusters.
Specifically, the embodiment of the application may utilize a first clustering algorithm to cluster the position record points in all the stay states, count a single maximum stay time of an individual in each position record point cluster, and use the position record point cluster with the single maximum stay time longer than a preset duration as an access location of the individual, where coordinates of the access location are centers of the corresponding position record point clusters.
The first clustering algorithm may use a DBSCAN (Density-Based Spatial Clustering of Applications with Noise) algorithm, which is a Density-based clustering algorithm.
Given a maximum scan radius and a minimum number of samples (recommended values of 50 meters and 2, respectively), a first clustering algorithm is used to cluster all the position records of a given individual belonging to a stay state. And according to the clustering result, counting the single maximum residence time of the given individual in each position record point cluster. And (3) reserving all position record point clusters with the single maximum stay time longer than a preset time length (which can be set to be 5 minutes), wherein each rest position record point cluster represents one access place of a given individual, and the coordinates of the access place are the centers of the corresponding position record point clusters.
Based on the individual trajectories and the clustering results, the individual trajectories may be represented as a series of time-stamped access location records, and further, the arrival times, departure times, and access orders of a given individual to a series of access locations may be determined.
S1013, clustering all access places of the individuals to obtain the access place clusters.
The embodiment of the application can cluster all access places of the individual by using a second clustering algorithm to obtain access place clusters.
The second clustering algorithm may be the HDBSCAN (Hierarchical Density-Based Spatial Clustering of Applications with Noise) algorithm, which is a hierarchical clustering algorithm based on the DBSCAN algorithm.
And (3) clustering all access places of the given individual by using a second clustering algorithm to obtain a plurality of access place clusters of the given individual, wherein the given cluster extraction parameters and the minimum sample number (recommended values are 1 km and 2 respectively), and determining the subordinate relation between each access place and each access cluster of the given individual according to the clustering result of the access places.
In step S102, crowd movement features are extracted according to the access locations, the access location clusters, and the historical crowd movement tracks, and model parameters of a crowd movement model are generated according to the crowd movement features, wherein the crowd movement features include access frequency ranks of all access locations of individuals and conditional probability distributions among access frequency ranks of the access location clusters to which the individual access locations belong.
Based on crowd movement track data of a specific city, a series of individual access places and access place clusters are obtained, crowd movement characteristics can be extracted according to the access places, the access place clusters and the crowd movement track data, and then a plurality of parameters of a crowd movement model are determined according to the characteristics. The crowd movement features may include residence time, movement step length, change situation of the number of visit places followed by the exploration probability, change situation of the number of visit places followed by the ratio of the random exploration probability to the exploration probability, relationship between the visit frequency ranking of the visit places and the visit frequency ranking of the visit place cluster to which the visit places belong, and the like.
In one embodiment of the present application, as a way of extracting crowd moving features, as shown in fig. 3, extracting crowd moving features according to access places, access place clusters, and historical crowd moving tracks includes:
s1021, calculating the stay time of each individual visit according to the visit sequence of each visit place, the arrival time and the departure time of each visit, and counting the stay time distribution of all individuals.
S1022, calculating the moving step length between each visit of each individual according to the visit sequence of each visit place of the individual, and counting the moving step length distribution of all the individuals.
S1023, counting the exploration probability of the individuals under different access places according to the historical movement track of the individuals.
Exploration probability
The statistical method is as follows according to the change condition of the number S of the access places: defining each visit behavior of an individual to be divided into a exploring behavior and a returning behavior, and when the individual visits a visit place which is never visited, the visit behavior is called as the exploring behavior; conversely, an individual accesses a past, current location when the access is referred to as a return.
When the individual has been currently visitedDefining the probability of the search behavior of the individual at the next visit when the number of the query places is S as the search probability of the individual at the visit place is S 
。
For each individual, based on its historical movement track, statistics is made of the search probabilities of that individual given the number S of different access points
Is a value of (a).
S1024, counting random exploration probabilities of the individuals under different access places according to the historical movement tracks of the individuals.
Random exploration probability
The ratio of the search probability to the search probability is changed along with the number S of the access places, and the statistical method is as follows: defining each exploration behavior of an individual as random exploration behaviors and exploration behaviors in clusters, and when the access places visited by the individual do not belong to any current access place cluster when the individual performs the exploration behaviors, calling the exploration behaviors as random exploration behaviors; otherwise, when the visit location visited by the individual when the individual performs the exploration behavior belongs to a current cluster of the visit locations, the exploration behavior is called as an intra-cluster exploration behavior.
When the number of access places which the individual has currently moved to is S, defining the probability of selecting random exploration behaviors when the individual makes exploration behaviors in the next access as the random exploration probability of the individual when the number of access places is S
。
For each individual, based on its historical movement trajectories, statistics is made of the probability of random exploration for that individual given the number of different access points S 
Is a value of (a).
S1025, sorting all access places of the individual and the access frequencies of all access place clusters, and calculating the conditional probability distribution among the access frequency sorting of all access places of the individual and the access frequency sorting of the access place clusters to which the access places belong.
The relation between the access frequency ordering of the access places and the access frequency ordering of the access place clusters to which the access places belong is calculated by the following statistical method: defining the number of access times of an individual to a certain access point as the access frequency of the access point
The method comprises the steps of carrying out a first treatment on the surface of the Ordering all access places from high to low according to their access frequency, defining the ordering of a certain access place as the access frequency ordering of that access place +.>
。
Defining the sum of the access times of an individual to all access sites in a certain access site cluster as the access frequency of the access site cluster
The method comprises the steps of carrying out a first treatment on the surface of the All access place clusters are ordered according to the access frequency from high to low, and the order of a certain access place cluster is defined as the access frequency order of the access place cluster +.>
。
Counting access frequency ranking for all access sites of each individual
And the access frequency ordering of the access site clusters to which it belongs +.>
Calculating conditional probability distribution +.>
。
The crowd movement characteristics of the embodiment of the application creatively consider the access frequency ordering of all access sites of an individual and the conditional probability distribution among the access frequency ordering of the access site clusters to which the individual belongs, and the distribution can reflect the access sites and the inherent relations among the access site clusters; two important features characterizing individual movement: the access sites belonging to the high-frequency access site cluster have higher probability of being accessed by individuals, and the access sites with higher access frequency are more likely to appear in the high-frequency access site cluster; and the existing model is proved to have obvious defects in describing the association relationship between the access places and the access place clusters to which the access places belong.
Based on the extracted features, model parameters are determined according to the following method
Is a numerical value of (1):
based on residence time
Fitting the distribution of +.>
Middle->
Is a numerical value of (2);
based on residence time
Fitting the distribution of +.>
Middle->
Is a number of (2);
based on the number S of access places and the exploration probability
Fitting equation->
Middle->
And
is a numerical value of (2);
based on the number of access places S and the random exploration probability
Fitting equation->
Middle->
Is a numerical value of (2);
access frequency ordering based on access location
And the access frequency ordering of the access site clusters to which it belongs +.>
Conditional probability distribution->
Parameter->
And->
The fitting method of the numerical values of (2) is as follows:
access frequency ordering for a given access location
The value of (2) is +.>
Calculating probability distribution
；
Fitting at given different r l Exponential distribution at the time
The value of m;
based on
And m, fitting the equation->
Middle->
And->
Is a numerical value of (2).
In step S103, the set number of simulated individuals and the target duration are input into a crowd movement model, and movement tracks of a plurality of simulated individuals within the target duration are generated through the crowd movement model.
Optionally, in one embodiment of the present application, generating, by the crowd movement model, movement trajectories of a plurality of simulated individuals within a target duration includes:
Selecting an initial position of each simulation individual, and setting the initial time to be 0;
extracting the residence time of the simulated individuals according to the crowd movement characteristics, and carrying out exploration behaviors or return behaviors according to the exploration probability of the simulated individuals;
when the simulated individual is the exploration behavior, carrying out random exploration behavior or intra-cluster exploration behavior according to the random exploration probability of the simulated individual;
when the simulated individuals perform random exploration behaviors, extracting movement step length according to crowd movement characteristics, and determining any exploration place except the visit place clusters of the simulated individuals in the movement step length range for visit;
when the simulated individuals perform intra-cluster exploration behaviors, determining a to-be-visited place cluster of the simulated individuals according to the visit frequency sequence of the visit place cluster, extracting a moving step length according to crowd moving features, and determining any exploration place of the simulated individuals in the to-be-visited place cluster and performing visit within the moving step length range;
when the simulated individual is in a return behavior, determining a to-be-accessed place cluster of the simulated individual according to the access frequency of the access place cluster, and determining and accessing to-be-accessed places of the simulated individual according to the access frequency of the access places in the to-be-accessed place cluster;
Calculating whether the total residence time of the simulated individual in the multiple access places is greater than or equal to the target duration, if the total residence time is greater than or equal to the target duration, generating a moving track of the simulated individual according to the positions of the multiple access places and the access sequence, and if the total residence time is less than the target duration, continuing to explore or return according to the exploration probability of the simulated individual until the total residence time is greater than or equal to the target duration.
As shown in fig. 4, a flow of a method of generating movement trajectories for a number of virtual individuals over a period of time is illustrated. The specific operation steps are as follows:
setting the number of individuals to generate the movement track and a target duration:
for each virtual individual, randomly selecting a position point in a selected city as the initial position of the individual;
setting the initial time to be 0, and repeating the following steps until the total residence time of the individual exceeds the target duration;
step A: from distribution
Extracting residence time->
；
And (B) step (B): to individuals by
The probability of (a) explores the behavior (go to step C), otherwise, go back to the behavior (go to step F);
step C: when the individual performs exploring action
Random exploration of the probability of (to step D) ) The method comprises the steps of carrying out a first treatment on the surface of the Otherwise, performing intra-cluster exploration behaviors (turning to the step E);
step D: when individuals conduct random exploration behaviors, the individuals are firstly distributed
Extracting the step size->
The method comprises the steps of carrying out a first treatment on the surface of the Then, let the individual to go from the current position as the center, < > or->
Randomly selecting a place which does not belong to any existing access place cluster for access on the circular arc with the radius; if there is no place on the arc meeting the above-mentioned requirements, the step of moving is regenerated +.>
Repeating step D;
step E: when individuals conduct intra-cluster exploration behaviors, probability is used
Select explore access frequency ranking +.>
Clustering i for the access places; after determining the access place clusters to be explored, from the distribution +.>
Extracting the step size->
The method comprises the steps of carrying out a first treatment on the surface of the Then, let the individual to go from the current position as the center, < > or->
And randomly selecting a place belonging to the selected cluster for access on the circular arc of the radius. If there is no place on the arc meeting the above-mentioned requirements, the step of moving is regenerated +.>
Repeating step E;
step F: when an individual performs a return action, firstly, selecting an access place cluster to be returned according to the access frequency of all access place clusters, wherein the return probability is proportional to the access frequency of each access place cluster; then, the access points to be returned are determined according to the access frequencies of all access points in the access point cluster, and the return probability is proportional to the access frequency of each access point.
The application introduces a two-stage decision process of individual movement under the view of visit site clustering, namely, an individual firstly makes a movement decision at a clustering level to decide which visit site cluster to visit and then makes a movement decision at the visit site level to decide which specific visit site to visit. As shown in fig. 5, in the crowd movement model, the influence of access location clusters on individual movement exists in both the preference return phase and the exploration phase. In the preference return stage, an individual firstly decides which access place cluster to access according to the access frequency of all access place clusters, and then decides which specific access place to access according to the access frequency of all access places in the cluster: this mechanism embodies that the attractions of the access sites constituting the access site cluster affect the return probability of the access site cluster, and in addition, the frequency of accesses by an individual to a certain access site is affected not only by its own attraction, but also by the attractions of the access site cluster to which it belongs, in other words, by the attractions of other access sites in the vicinity thereof; similarly, during the exploration process, the individual first decides within which cluster to explore (or chooses to explore outside of all existing clusters), and then decides new places to explore within the chosen clusters based on the movement step size: this mechanism means that the individual selection of a new location to explore is not completely random as assumed in the EPR model, but is affected by the clustering of existing access locations.
In the test, the present model was tested on two people moving track data sets of guangzhou and houston in united states, and when the test was performed on the guangzhou, the number of individuals to generate moving tracks was set to 17000, and the target time period was set to 5 months, referring to the time length and the number of samples covered by the guangzhou, city data set. Based on the result of the parameter fitting of the Guangzhou market dataset, relevant parameters in the model are set as follows: β=0.618, ρ=0.697, γ=0.199, α=0.615, τ=0.595, ω= -2.848, ε=0.578; when testing on the houston data set, the number of individuals to generate the movement track is set to 17000, and the target period is set to 50 weeks, with reference to the length of time and the number of samples covered by the houston data set. Based on the result of the parameter fitting of the houston market dataset, the relevant parameters in the present model are set as follows: β=0.800, ρ=0.647, γ=0.273, α=0.416, τ=0.606, ω= -3.572, ε=0.819. Based on the test, the association relation of the access points in the human movement tracks of the two cities and the access frequency ordering of the clusters to which the access points belong can be obtained respectively, and the association relation of the access points in the human movement tracks generated by the model and the access frequency ordering of the clusters to which the access points belong can be found out to be more consistent with the characteristics presented by the data set by comparing the association relation of the access points in the human movement tracks generated by the model and the access frequency ordering of the clusters to which the access points belong: i.e. access points belonging to the high frequency access point cluster have a higher access frequency and access points with a higher access frequency are more likely to occur in the high frequency access point cluster. The existing EPR model has obvious defects in the relation of describing access places and the access frequency ordering of clusters to which the EPR model belongs, in the result generated by the EPR model, the access places of the clusters with different access frequencies are almost the same in composition, and the access places with higher access frequencies have higher probability of appearing in the clusters of the low-frequency access places. The results prove that the model can successfully reproduce the intrinsic relation between the access places and the access place clusters in the crowd movement, and compared with the existing EPR model, the model has obvious improvement. Secondly, by introducing access site clustering into the model, the model establishes a relation between the movement of an individual and the physical space in the city, and on the basis of the model, the relation between the movement behavior of a person and various city spaces can be realized, thereby being beneficial to further simulating the interaction process of urban residents and various city spaces.
According to the crowd movement modeling method based on the visit site clustering, the visit site clustering is identified from the individual movement track, the crowd movement characteristics are obtained by utilizing the movement behaviors of the individuals on the visit site clustering level, the stay time, the movement step length, the variation condition of the individual exploration probability follow-up visit site quantity, the variation condition of the individual random exploration probability follow-up visit site quantity, the visit frequency sequencing of all visit sites of the individuals and the conditional probability distribution among the visit frequency sequencing of the visit site clustering to which the visit site clusters belong are creatively considered, and the intrinsic relation between the visit sites and the visit site clustering in the crowd movement can be successfully reproduced by considering the conditional probability distribution, so that the important characteristics of the individual movement are characterized, and the defects of the existing model are filled. The two-stage decision process of the movement of the individual under the view angle of the visit site clusters is further introduced, so that not only can the movement track of the individual be predicted, but also the visit modes of the individual between the visit site clusters and inside the visit site clusters can be predicted, the inherent connection between the visit site and the visit site clusters can be reproduced, and the defect of the existing model can be overcome.
Next, a crowd movement modeling apparatus based on access location clustering according to an embodiment of the present application will be described with reference to the accompanying drawings.
FIG. 6 is an exemplary diagram of a crowd movement modeling apparatus based on access location clustering in accordance with an embodiment of the present application.
As shown in fig. 6, the crowd movement modeling apparatus 10 based on access place clustering includes: an acquisition module 100, a parameter generation module 200 and a trajectory generation module 300.
The acquisition module 100 is configured to acquire access locations and access location clusters of a plurality of individuals in a crowd based on historical crowd movement track data.
The parameter generation module 200 is configured to extract crowd movement features according to the access locations, the access location clusters and the historical crowd movement tracks, and generate model parameters of a crowd movement model according to the crowd movement features, where the crowd movement features include access frequency ranks of all access locations of the individual and conditional probability distribution among access frequency ranks of the access location clusters to which the individual belongs.
The track generation module 300 is configured to input the set number of simulated individuals and the target duration into a crowd movement model, and generate movement tracks of a plurality of simulated individuals within the target duration through the crowd movement model.
Optionally, in one embodiment of the present application, the acquiring module 100 includes: the identification unit is used for determining a plurality of position record points in the individual track according to the historical crowd movement track data, calculating the current speed of the position record points, and marking the position record points with the current speed less than or equal to a preset speed threshold as stay position record points; the first clustering unit is used for clustering all the stay position recording points, calculating single stay time of the individual in each position recording point cluster, and determining access places of the individual according to the single stay time, wherein the coordinates of the access places are the centers of the corresponding position recording point clusters; and the second clustering unit is used for clustering all the access places of the individuals to obtain access place clusters.
Optionally, in one embodiment of the present application, the parameter generation module 200 includes: the first extraction unit is used for calculating the stay time of each access of each individual according to the access sequence of the individual to each access place, the arrival time and the departure time of each access, and counting the stay time distribution of all the individuals; the second extraction unit is used for calculating the moving step length between each visit of each individual according to the visit sequence of the individual to each visit place and counting the moving step length distribution of all the individuals; the third extraction unit is used for counting the exploration probability of the individuals under different access places according to the historical movement track of the individuals; a fourth extraction unit, configured to count random exploration probabilities of individuals under different access location numbers according to historical movement tracks of the individuals; and a fifth extraction unit, configured to rank all access sites of the individual and access frequencies of all access site clusters, and calculate a conditional probability distribution between the access frequency ranks of all access sites of the individual and the access frequency ranks of the access site clusters to which the access sites belong.
Optionally, in one embodiment of the present application, the track generation module 300 includes: the initial unit is used for selecting the initial position of each simulation individual and setting the initial time to be 0; the exploration unit is used for extracting the residence time of the simulated individuals according to the crowd movement characteristics and carrying out exploration behaviors or return behaviors according to the exploration probability of the simulated individuals; the exploration unit is also used for carrying out random exploration behaviors or intra-cluster exploration behaviors according to the random exploration probability of the analog individual when the analog individual is the exploration behaviors; the exploration unit is also used for extracting a moving step length according to crowd moving characteristics when the simulated individuals conduct random exploration behaviors, and determining any exploration place except the visit place clusters of the simulated individuals and visiting the exploration places in the moving step length range; the searching unit is also used for determining a to-be-accessed place cluster of the simulated individual according to the access frequency sequence of the access place cluster when the simulated individual performs intra-cluster searching behaviors, extracting a moving step length according to crowd moving characteristics, and determining any searching place of the simulated individual in the to-be-accessed place cluster and accessing the to-be-accessed place cluster within the moving step length range; the exploration unit is also used for determining a to-be-accessed place cluster of the analog individual according to the access frequency of the access place cluster when the analog individual is in a return behavior, and determining the to-be-accessed place of the analog individual and accessing according to the access frequency of the access place in the to-be-accessed place cluster; the output unit is used for calculating whether the total residence time of the simulated individual in the multiple access places is greater than or equal to the target duration, if the total residence time is greater than or equal to the target duration, generating a moving track of the simulated individual according to the positions of the multiple access places and the access sequence, and if the total residence time is less than the target duration, continuing to perform exploration behaviors or return behaviors according to the exploration probability of the simulated individual until the total residence time is greater than or equal to the target duration.
It should be noted that the explanation of the foregoing embodiment of the crowd movement modeling method based on the access location cluster is also applicable to the crowd movement modeling device based on the access location cluster of the embodiment, and will not be repeated here.
According to the crowd movement modeling device based on the access location clustering, the access location clustering is identified from the individual movement track, the crowd movement characteristics are obtained by utilizing the movement behaviors of the individuals on the access location clustering level, the stay time, the movement step length, the variation condition of the number of the access locations followed by the exploration probability of the individuals and the variation condition of the number of the access locations followed by the random exploration probability of the individuals are utilized, the conditional probability distribution among the access frequency sequencing of all the access locations of the individuals and the access frequency sequencing of the access location clustering to which the access location is belonged is creatively considered, and the intrinsic relation between the access locations and the access location clustering in crowd movement can be successfully reproduced by considering the conditional probability distribution, so that the important characteristics of individual movement are characterized, and the defects of the existing model are filled. The two-stage decision process of the movement of the individual under the view angle of the visit site clusters is further introduced, so that not only can the movement track of the individual be predicted, but also the visit modes of the individual between the visit site clusters and inside the visit site clusters can be predicted, the inherent connection between the visit site and the visit site clusters can be reproduced, and the defect of the existing model can be overcome.
In the description of the present specification, a description referring to terms "one embodiment," "some embodiments," "examples," "specific examples," or "some examples," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the present application. In this specification, schematic representations of the above terms are not necessarily directed to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or N embodiments or examples. Furthermore, the different embodiments or examples described in this specification and the features of the different embodiments or examples may be combined and combined by those skilled in the art without contradiction.
Furthermore, the terms "first," "second," and the like, are used for descriptive purposes only and are not to be construed as indicating or implying a relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defining "a first" or "a second" may explicitly or implicitly include at least one such feature. In the description of the present application, the meaning of "N" is at least two, such as two, three, etc., unless explicitly defined otherwise.
Any process or method descriptions in flow charts or otherwise described herein may be understood as representing modules, segments, or portions of code which include one or more N executable instructions for implementing specific logical functions or steps of the process, and further implementations are included within the scope of the preferred embodiment of the present application in which functions may be executed out of order from that shown or discussed, including substantially concurrently or in reverse order, depending on the functionality involved, as would be understood by those reasonably skilled in the art of the embodiments of the present application.
Claims (6)
1. The crowd movement modeling method based on the access place clustering is characterized by comprising the following steps of:
acquiring access sites and access site clusters of a plurality of individuals in a crowd based on historical crowd movement track data;
extracting crowd movement features according to the access places, the access place clusters and the historical crowd movement tracks, and generating model parameters of a crowd movement model according to the crowd movement features, wherein the crowd movement features comprise access frequency sequences of all access places of individuals and conditional probability distribution among access frequency sequences of the access place clusters to which the individuals belong;
Inputting the set number of simulated individuals and target time length into the crowd movement model, and generating movement tracks of a plurality of simulated individuals in the target time length through the crowd movement model;
wherein, extracting crowd movement features according to the visit location, the visit location cluster, and the historical crowd movement track comprises:
calculating the residence time of each individual visit according to the visit sequence of each individual visit to each visit place, the arrival time and the departure time of each visit, and counting the residence time distribution of all individuals;
calculating the moving step length between each visit of each individual according to the visit sequence of each individual to each visit place, and counting the moving step length distribution of all the individuals;
counting the exploration probability of the individual under different access places according to the historical movement track of the individual;
according to the historical movement track of the individual, counting random exploration probabilities of the individual under different access places;
sorting all access places of the individual and the access frequencies of all access place clusters, and calculating the access frequency sorting of all access places of the individual and the conditional probability distribution among the access frequency sorting of the access place clusters to which the access places belong.
2. The method of claim 1, wherein the obtaining access locations and access location clusters for a plurality of individuals in the crowd based on the historical crowd movement trajectory data comprises:
determining a plurality of position record points in an individual track according to the historical crowd movement track data, calculating the current speeds of the position record points, and marking the position record points with the current speeds less than or equal to a preset speed threshold as stay position record points;
clustering all the stay position record points, calculating single stay time of an individual in each position record point cluster, and determining access places of the individual according to the single stay time, wherein coordinates of the access places are centers of the corresponding position record point clusters;
clustering all access places of the individuals to obtain the access place clusters.
3. The method of claim 1 or 2, wherein generating, by the crowd movement model, movement trajectories for a plurality of simulated individuals within the target time period comprises:
selecting an initial position of each simulation individual, and setting the initial time to be 0;
extracting the residence time of the simulated individuals according to the crowd movement characteristics, and carrying out exploration behaviors or return behaviors according to the exploration probability of the simulated individuals;
When the simulated individual is the exploration behavior, carrying out random exploration behavior or intra-cluster exploration behavior according to the random exploration probability of the simulated individual;
when the simulated individuals conduct random exploration behaviors, extracting movement step length according to the crowd movement characteristics, and determining any exploration place of the simulated individuals outside the visit place cluster in the movement step length range for visit;
when the simulated individuals perform intra-cluster exploration behaviors, determining a to-be-visited place cluster of the simulated individuals according to the visit frequency sequence of the visit place cluster, extracting the moving step length according to the crowd moving characteristics, and determining any exploration place of the simulated individuals in the to-be-visited place cluster and visiting the exploration place within the moving step length range;
when the simulated individual is in a return behavior, determining a to-be-accessed place cluster of the simulated individual according to the access frequency of the access place cluster, and determining and accessing the to-be-accessed place of the simulated individual according to the access frequency of the access place in the to-be-accessed place cluster;
calculating whether the total residence time of the simulation individual in a plurality of access places is larger than or equal to the target duration, if the total residence time is larger than or equal to the target duration, generating a moving track of the simulation individual according to the positions and the access sequence of the plurality of access places, and if the total residence time is smaller than the target duration, continuing to search or return according to the exploration probability of the simulation individual until the total residence time is larger than or equal to the target duration.
4. A crowd movement modeling apparatus based on access location clustering, comprising:
the acquisition module is used for acquiring access sites and access site clusters of a plurality of individuals in the crowd based on the historical crowd movement track data;
the parameter generation module is used for extracting crowd movement characteristics according to the access places, the access place clusters and the historical crowd movement tracks, and generating model parameters of a crowd movement model according to the crowd movement characteristics, wherein the crowd movement characteristics comprise access frequency sequences of all access places of individuals and conditional probability distribution among the access frequency sequences of the access place clusters to which the individual access places belong;
the track generation module is used for inputting the set number of the simulated individuals and the target duration into the crowd movement model, and generating movement tracks of a plurality of the simulated individuals in the target duration through the crowd movement model;
the parameter generation module comprises:
the first extraction unit is used for calculating the stay time of each access of each individual according to the access sequence of the individual to each access place, the arrival time and the departure time of each access, and counting the stay time distribution of all the individuals;
The second extraction unit is used for calculating the moving step length between each visit of each individual according to the visit sequence of the individual to each visit place and counting the moving step length distribution of all the individuals;
the third extraction unit is used for counting the exploration probability of the individuals under different access places according to the historical movement track of the individuals;
a fourth extraction unit, configured to count random exploration probabilities of individuals under different access location numbers according to historical movement tracks of the individuals;
and a fifth extraction unit, configured to rank all access sites of the individual and access frequencies of all access site clusters, and calculate a conditional probability distribution between the access frequency ranks of all access sites of the individual and the access frequency ranks of the access site clusters to which the access sites belong.
5. The apparatus of claim 4, wherein the acquisition module comprises:
the identification unit is used for determining a plurality of position record points in the individual track according to the historical crowd moving track data, calculating the current speed of the position record points, and marking the position record points with the current speed less than or equal to a preset speed threshold as stay position record points;
the first clustering unit is used for clustering all the stay position record points, calculating single stay time of an individual in each position record point cluster, and determining access places of the individual according to the single stay time, wherein the coordinates of the access places are the centers of the corresponding position record point clusters;
And the second clustering unit is used for clustering all the access places of the individuals to obtain the access place clusters.
6. The apparatus of claim 4 or 5, wherein the trajectory generation module comprises:
the initial unit is used for selecting the initial position of each simulation individual and setting the initial time to be 0;
the exploration unit is used for extracting the residence time of the analog individuals according to the crowd movement characteristics and carrying out exploration behaviors or return behaviors according to the exploration probability of the analog individuals;
the exploration unit is further used for conducting random exploration behaviors or intra-cluster exploration behaviors according to random exploration probabilities of the simulated individuals when the simulated individuals are exploration behaviors;
the exploration unit is further used for extracting a moving step length according to the crowd moving characteristics when the simulated individuals conduct random exploration behaviors, and determining any exploration place of the simulated individuals outside the visit place cluster and visiting the exploration place within the moving step length range;
the exploration unit is further used for determining a to-be-visited place cluster of the analog individual according to the visit frequency sequence of the visit place cluster when the analog individual performs intra-cluster exploration behaviors, extracting the moving step length according to the crowd moving characteristics, and determining any exploration place of the analog individual in the to-be-visited place cluster and visiting the exploration place within the moving step length range;
The exploration unit is further used for determining a to-be-accessed place cluster of the simulated individual according to the access frequency of the access place cluster when the simulated individual is in return behavior, and determining and accessing the to-be-accessed place of the simulated individual according to the access frequency of the access place in the to-be-accessed place cluster;
the output unit is used for calculating whether the total residence time of the simulation individual at a plurality of access places is greater than or equal to the target duration, if the total residence time is greater than or equal to the target duration, generating a moving track of the simulation individual according to the positions and the access sequence of the plurality of access places, and if the total residence time is less than the target duration, continuing to perform exploration or return according to the exploration probability of the simulation individual until the total residence time is greater than or equal to the target duration.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310281628.2A CN115994313B (en) | 2023-03-22 | 2023-03-22 | Crowd movement modeling method and device based on access place clustering |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310281628.2A CN115994313B (en) | 2023-03-22 | 2023-03-22 | Crowd movement modeling method and device based on access place clustering |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115994313A CN115994313A (en) | 2023-04-21 |
| CN115994313B true CN115994313B (en) | 2023-05-30 |
Family
ID=85993686
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310281628.2A Active CN115994313B (en) | 2023-03-22 | 2023-03-22 | Crowd movement modeling method and device based on access place clustering |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115994313B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106951903A (en) * | 2016-10-31 | 2017-07-14 | 浙江大学 | A kind of method for visualizing of crowd's movement law |
| CN113486927A (en) * | 2021-06-15 | 2021-10-08 | 北京大学 | Unsupervised track access place labeling method based on prior probability |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106407519B (en) * | 2016-08-31 | 2019-04-16 | 浙江大学 | A kind of modeling method of crowd's movement law |
| US11037064B2 (en) * | 2017-10-19 | 2021-06-15 | International Business Machines Corporation | Recognizing recurrent crowd mobility patterns |
| CN113505314A (en) * | 2021-07-27 | 2021-10-15 | 王程 | Position track analysis system for space-time complex network clustering |
| CN114580539A (en) * | 2022-03-04 | 2022-06-03 | 京东鲲鹏(江苏)科技有限公司 | Vehicle driving strategy processing method and device |
-
2023
- 2023-03-22 CN CN202310281628.2A patent/CN115994313B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106951903A (en) * | 2016-10-31 | 2017-07-14 | 浙江大学 | A kind of method for visualizing of crowd's movement law |
| CN113486927A (en) * | 2021-06-15 | 2021-10-08 | 北京大学 | Unsupervised track access place labeling method based on prior probability |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115994313A (en) | 2023-04-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Liu et al. | Modelling urban change with cellular automata: Contemporary issues and future research directions | |
| Openshaw | Developing GIS-relevant zone-based spatial analysis methods | |
| Long et al. | Mapping block-level urban areas for all Chinese cities | |
| Yang et al. | Predicting next location using a variable order Markov model | |
| Chen et al. | Identification of general fuzzy measures by genetic algorithms based on partial information | |
| CN113191023B (en) | Crowd-sourcing-aware task allocation and user recruitment model cross-validation method and system | |
| CN113158038B (en) | Interest point recommendation method and system based on STA-TCN neural network framework | |
| CN109146741A (en) | A kind of course method for pushing, device and equipment based on capability comparison | |
| Du et al. | A comparative approach to modelling multiple urban land use changes using tree-based methods and cellular automata: the case of Greater Tokyo Area | |
| KR20170030379A (en) | Method and system for personalized travel curation service | |
| CN107818334A (en) | A kind of mobile Internet user access pattern characterizes and clustering method | |
| CN103440199B (en) | Test bootstrap technique and device | |
| Liu et al. | Spatio-temporal topic models for check-in data | |
| Elgar et al. | Simulations of firm location decisions: Replicating office location choices in the Greater Toronto Area | |
| CN108182240A (en) | The new gaining rate prediction model training of point of interest and Forecasting Methodology, device and storage medium | |
| Geertman et al. | Spatial‐temporal specific neighbourhood rules for cellular automata land‐use modelling | |
| Wei et al. | Exploring pattern-aware travel routes for trajectory search | |
| Gonçalves et al. | Cartographic visualization of human trajectory data: Overview and analysis | |
| CN108665513A (en) | Drawing practice based on user behavior data and device | |
| CN109858024A (en) | A kind of source of houses term vector training method and device based on word2vec | |
| Bobkova | Towards a theory of natural occupation: Developing theoretical, methodological and empirical support for the relation between plot systems and urban processes | |
| Pizarro et al. | ABM RoutePlanner: An agent-based model simulation for suggesting preference-based routes in Spain | |
| CN115994313B (en) | Crowd movement modeling method and device based on access place clustering | |
| Wallace et al. | Towards explaining metaheuristic solution quality by data mining surrogate fitness models for importance of variables | |
| Wang et al. | Spatial evaluation of the accessibility of public service facilities in Shanghai: A community differentiation perspective |
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 |