Methods, devices, equipment and media for predicting the spatiotemporal characteristics of population movement

Abstract

This application relates to a method, apparatus, device, and medium for predicting the spatiotemporal characteristics of population flow. The method includes: determining the change in population building scale based on changes in building occupancy and building area; determining the change in outflow of people in adjacent time periods based on changes in population building scale, and determining the change in inflow of people in adjacent time periods based on fusion weights and a comprehensive importance index; calculating the dynamic variables for the next time period based on a dynamic variable iterative model, and correcting the dynamic variables for the next time period based on a correction model to determine the corrected dynamic variables for the next time period; inputting the corrected dynamic variables for the next time period, the changes in outflow of people in the initial time period, and the changes in inflow of people into the spatiotemporal characteristic prediction model of population flow to determine the outflow and inflow of people within a preset time range, and determining the spatiotemporal characteristics of population flow based on the outflow and inflow of people. This application can improve the interpretability of the population flow prediction model.

Description

Technical Field

[0001] This application relates to the field of data processing, and in particular to a method for predicting the spatiotemporal characteristics of population movement, a corresponding device, electronic equipment, and a computer-readable storage medium. Background Technology

[0002] In an era of rapid urbanization, the demand for population mobility is increasing. Dynamic prediction of the inflow and outflow characteristics of a region can guide policymakers to make targeted adjustments to the spatiotemporal characteristics of population mobility. Population mobility characteristics are also an important basis for calculating road traffic flow and the pollutants emitted.

[0003] Traditional prediction models do not take into account the spatiotemporal heterogeneity of crowd flow. Their input dependent variables are static variables, such as point of interest density, road density, and population density, which leads to relatively fixed prediction results. Or they only predict crowd flow for a single mode of transportation, such as buses and taxis.

[0004] The existing prediction models do not take into account the spatiotemporal heterogeneity of crowd flow, and their input dependent variables are static variables, resulting in relatively fixed prediction results or predictions only for crowd flow of a single mode of transportation. The applicant has made corresponding explorations to address this problem. Summary of the Invention

[0005] The purpose of this application is to solve the above-mentioned problems by providing a method, device, electronic device and computer-readable storage medium for predicting the spatiotemporal characteristics of population flow.

[0006] To achieve the various objectives of this application, the following technical solution is adopted:

[0007] A method for predicting the spatiotemporal characteristics of population flow, proposed to meet one of the purposes of this application, includes:

[0008] In response to the command to predict the spatiotemporal characteristics of population flow, determine the change in the number of people in buildings and the building area corresponding to each building category, and determine the change in population building scale based on the change in the number of people in buildings and the building area corresponding to each building category.

[0009] The change in the number of people leaving each building category in adjacent time periods is determined based on the change in population and building scale, and the change in the number of people entering each building category in adjacent time periods is determined based on the fusion weight and the comprehensive importance index.

[0010] The dynamic variables for the next period are calculated based on the dynamic variable iterative model. The dynamic variables for the next period are then corrected based on the correction model to determine the corrected dynamic variables for the next period. The dynamic variables represent the changes in population and building scale corresponding to each building category.

[0011] The revised changes in population building size for each building category in the next time period, the changes in outflow and inflow for each building category in the initial time period are input into the spatiotemporal feature prediction model for population flow to determine the outflow and inflow for each building category within a preset time range. Based on the outflow and inflow for each building category, the spatiotemporal features of population flow are determined, and the spatiotemporal feature prediction of population flow for each building category is completed.

[0012] Optionally, the step of determining the change in population and building size based on the change in the number of people in buildings corresponding to each building category and the building area includes:

[0013] Obtain the change in the number of people in buildings corresponding to each building category and the building area corresponding to each building category in adjacent time periods;

[0014] The change in population and building size for each building category is determined by multiplying the change in the number of people in each building category for adjacent time periods by the building area for each building category.

[0015] Optionally, the step of determining the change in the number of people leaving each building category in adjacent time periods based on the change in population building size includes:

[0016] A spatiotemporal characteristic prediction model for determining the outflow population change corresponding to each building category based on population and building scale changes;

[0017] The changes in population and building size in adjacent time periods are input into the spatiotemporal feature prediction model of the change in outflow population to determine the change in outflow population corresponding to each building category in adjacent time periods.

[0018] Optionally, the step of determining the change in the number of visitors corresponding to each building category in adjacent time periods based on the fusion weight and the comprehensive importance index includes:

[0019] In response to the instruction to construct a spatiotemporal characteristic prediction model for changes in the number of inflows, the importance comprehensive index, fusion weight, and community traffic convenience index corresponding to each building category are determined.

[0020] A spatiotemporal feature prediction model for the change in the number of inflows is constructed based on the comprehensive importance index, fusion weight, and community transportation convenience index corresponding to each building category.

[0021] Based on the spatiotemporal feature prediction model, the change in the number of people entering each building category is determined.

[0022] Optionally, the steps for determining the comprehensive importance index corresponding to each building category include:

[0023] Using the change in population size of buildings as the feature variable and the change in the number of people entering each building category in adjacent time periods as the target variable, the change in population size of buildings and the change in the number of people entering each building category in adjacent time periods are input into the random forest model to determine the importance index corresponding to each building category. The importance index includes one or more of the average accuracy decline coefficient and the average Gini coefficient.

[0024] The product of the average accuracy decrease coefficient and the average Gini coefficient, after being normalized, is used as the comprehensive importance index corresponding to each building category.

[0025] Optionally, the dynamic variable iterative model is The For the original t o +T time period and t o The change in the number of people in building type m in community i during time period i, the As a coefficient, the S im Let m be the building area of ​​the m-th type of building in community i. Let be the change in the number of people in building type m in community i between time period t and time period tT. This is the intercept.

[0026] Optionally, the modified model is as well as Among them, the For the corrected t o The change in the number of people in building type m in community i between time period +T and time period t, the For t o Time period and t o -Change in the number of people in cell i during time period T.

[0027] A population flow spatiotemporal characteristic prediction device provided for another purpose of this application includes:

[0028] The population change determination module is configured to respond to the spatiotemporal characteristics prediction command of population flow, determine the change in the number of people in buildings and the building area corresponding to each building category, and determine the change in population building scale based on the change in the number of people in buildings and the building area corresponding to each building category.

[0029] The module for determining the change in the number of people in transit is configured to determine the change in the number of people leaving each building category in adjacent time periods based on the change in population and building scale, and to determine the change in the number of people entering each building category in adjacent time periods based on the fusion weight and the comprehensive importance index.

[0030] The dynamic variable determination module is configured to calculate the dynamic variables for the next time period based on the dynamic variable iterative model, correct the dynamic variables for the next time period based on the correction model, and determine the corrected dynamic variables for the next time period. The dynamic variables represent the changes in population and building scale corresponding to each building category.

[0031] The crowd flow prediction module is configured to input the corrected changes in population and building size corresponding to each building category in the next time period, the changes in outflow and inflow of each building category in the initial time period, into the crowd flow spatiotemporal feature prediction model to determine the outflow and inflow of each building category within a preset time range. Based on the outflow and inflow of each building category, the module determines the spatiotemporal features of crowd flow and completes the prediction of the spatiotemporal features of crowd flow corresponding to each building category.

[0032] An electronic device provided for another purpose of this application includes a central processing unit and a memory, the central processing unit being configured to invoke and run a computer program stored in the memory to perform the steps of the population flow spatiotemporal characteristic prediction method of this application.

[0033] A computer-readable storage medium is provided for another purpose of this application, which stores, in the form of computer-readable instructions, a computer program implemented according to the method for predicting the spatiotemporal characteristics of population flow, which, when called by a computer, executes the steps included in the corresponding method.

[0034] Compared to existing technologies, this application addresses the problems of existing prediction models failing to consider the spatiotemporal heterogeneity of crowd flow, using static input variables, resulting in relatively fixed prediction results, or only predicting crowd flow for a single mode of transportation. This application offers the following advantages, including but not limited to:

[0035] First, the prediction method is based on a spatiotemporal geographic weighted regression model, which takes into account the spatiotemporal heterogeneity of population movement and uses dynamic independent variables (population size change and transportation convenience) as input variables, thus making up for the shortcomings of previous prediction models that used static independent variables as input variables, resulting in relatively fixed prediction results.

[0036] Secondly, the spatiotemporal characteristic prediction model for changes in the number of inflows combines importance comprehensive indicators and fusion weights to emphasize the impact of each variable on the target value and improve the interpretability of the model.

[0037] Third, the dynamic variable iterative model contained in this application significantly reduces the amount of data required in the prediction stage, which can greatly improve the prediction efficiency of spatiotemporal characteristics of the population, and greatly improve the prediction speed while ensuring its prediction accuracy.

[0038] Fourth, this application overcomes the limitation of previous prediction models that are only applicable to the prediction of population flow under a single mode of transportation. It can predict the spatiotemporal characteristics of population flow under multiple modes of transportation. Attached Figure Description

[0039] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein:

[0040] Figure 1 This is a flowchart illustrating the method for predicting the spatiotemporal characteristics of population flow in the embodiments of this application;

[0041] Figure 2 This is a schematic diagram of the population flow spatiotemporal characteristic prediction device in the embodiments of this application;

[0042] Figure 3 This is a schematic diagram of the structure of the computer device in the embodiments of this application. Detailed Implementation

[0043] The embodiments of this application are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain this application, and should not be construed as limiting this application.

[0044] Those skilled in the art will understand that, unless specifically stated otherwise, the singular forms “a,” “an,” “the,” and “the” used herein may also include the plural forms. It should be further understood that the term “comprising” as used in this application means the presence of the stated features, integers, steps, operations, elements, and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof. It should be understood that when we say an element is “connected” or “coupled” to another element, it can be directly connected or coupled to the other element, or there may be intermediate elements. Furthermore, “connected” or “coupled” as used herein can include wireless connections or wireless coupling. The term “and / or” as used herein includes all or any units and all combinations of one or more associated listed items.

[0045] It will be understood by those skilled in the art that, unless otherwise defined, all terms used herein (including technical and scientific terms) have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains. It should also be understood that terms such as those defined in general dictionaries should be understood to have the same meaning as in the context of the prior art, and should not be interpreted in an idealized or overly formal sense unless specifically defined as herein.

[0046] Those skilled in the art will understand that the terms "client," "terminal," and "terminal device" as used herein include both devices that receive wireless signals, devices that only possess wireless signal receiver capabilities without transmission capabilities, and devices with receiving and transmitting hardware, devices that have receiving and transmitting hardware capable of bidirectional communication over a bidirectional communication link. Such devices may include: cellular or other communication devices such as personal computers or tablets, having single-line displays, multi-line displays, or cellular or other communication devices without multi-line displays; PCS (Personal Communications Service) that can combine voice, data processing, fax, and / or data communication capabilities; PDA (Personal Digital Assistant) that may include a radio frequency receiver, pager, internet / intranet access, web browser, notepad, calendar, and / or GPS (Global Positioning System) receiver; and conventional laptops and / or handheld computers or other devices that have and / or include radio frequency receivers. As used herein, "client," "terminal," and "terminal device" can be portable, transportable, installed in a means of transportation (air, sea, and / or land), or suitable and / or configured to operate locally and / or in a distributed manner, operating in any other location on Earth and / or in space. "Client," "terminal," and "terminal device" as used herein can also be a communication terminal, an internet access terminal, or a music / video playback terminal, such as a PDA, a MID (Mobile Internet Device), and / or a mobile phone with music / video playback capabilities, or a smart TV, set-top box, etc.

[0047] The hardware referred to by the names "server," "client," and "service node" in this application is essentially an electronic device with the equivalent capabilities of a personal computer. It is a hardware device with the necessary components revealed by the von Neumann architecture, such as a central processing unit (including an arithmetic logic unit and a control unit), memory, input devices, and output devices. The computer program is stored in its memory, and the central processing unit loads the program stored in the secondary storage into the main memory to run it, execute the instructions in the program, and interact with the input and output devices to complete specific functions.

[0048] It should be noted that the concept of "server" used in this application can also be extended to the case of server clusters. Based on the network deployment principles understood by those skilled in the art, the servers should be logically divided. Physically, these servers can be independent of each other but accessible through interfaces, or they can be integrated into a single physical computer or a computer cluster. Those skilled in the art should understand this flexibility and should not use it to constrain the implementation of the network deployment method in this application.

[0049] One or more of the technical features of this application, unless explicitly specified herein, can be deployed on a server and accessed by a client remotely calling the online service interface provided by the server, or can be directly deployed and run on a client for access.

[0050] Unless otherwise specified, the neural network models referenced or potentially referenced in this application may be deployed on a remote server and invoked remotely on the client, or deployed on a client with the capability to invoke directly. In some embodiments, when running on the client, the corresponding intelligence may be acquired through transfer learning in order to reduce the requirements on the client's hardware resources and avoid excessive consumption of the client's hardware resources.

[0051] Unless otherwise specified, all data involved in this application may be stored remotely on a server or on a local terminal device, as long as it is suitable for use by the technical solution of this application.

[0052] Those skilled in the art will understand that although the various methods in this application are described based on the same concept and thus present commonality among them, they can be performed independently unless otherwise specified. Similarly, the various embodiments disclosed in this application are all based on the same inventive concept; therefore, concepts expressed in the same way, as well as concepts that are appropriately changed for convenience but are expressed differently, should be understood equivalently.

[0053] Unless otherwise expressly stated, the various embodiments disclosed in this application can be combined in a cross-cutting manner to flexibly construct new embodiments, as long as such combination does not depart from the inventive spirit of this application and can meet the needs of the prior art or solve a certain deficiency in the prior art. Those skilled in the art should be aware of such modifications.

[0054] Please see Figure 1 In one embodiment of the population flow spatiotemporal characteristic prediction method of this application, the method includes:

[0055] Step S10: Respond to the command for predicting the spatiotemporal characteristics of population flow, determine the change in the number of people in buildings and the building area corresponding to each building category, and determine the change in population building scale based on the change in the number of people in buildings and the building area corresponding to each building category.

[0056] The terminal device can respond to commands predicting the spatiotemporal characteristics of population flow, determine the changes in the number of people in buildings corresponding to each building category, and the building area. Based on the changes in the number of people in buildings corresponding to each building category and the building area, it determines the changes in population building scale. First, based on the time step, community, and building division, it calculates the changes in the number of people entering and leaving the community, the total number of people, and the changes in the number of people corresponding to each building category in adjacent time periods. Then, it multiplies the changes in the number of people corresponding to each building category in adjacent time periods with their corresponding building areas to obtain the changes in population building scale. The expression is as follows:

[0057] ΔSPBA imt =ΔN imt ×S im

[0058] Among them, the change in population and building scale ΔSPBA between adjacent time periods imt ;ΔN imt S represents the change in the number of people in building type m in community i between time period t and time period tT; im Let be the building area of ​​the m-th type of building in community i.

[0059] Step S20: Determine the change in the number of people leaving each building category in adjacent time periods based on the change in population building scale, and determine the change in the number of people entering each building category in adjacent time periods based on the fusion weight and importance comprehensive index.

[0060] Based on the changes in population and building scale, the changes in the number of people leaving each building category in adjacent time periods are determined. That is, a spatiotemporal characteristic prediction model for the changes in the number of people leaving is determined based on the changes in population and building scale. This spatiotemporal characteristic prediction model for the changes in the number of people leaving uses the changes in the number of people leaving in adjacent time periods as the dependent variable and the changes in population and building scale as the independent variable. The expression of the spatiotemporal characteristic prediction model for the changes in the number of people leaving is as follows:

[0061]

[0062] Wherein, ΔP it ΔSPBA represents the change in the number of people leaving cell i between time period t and time period tT; and the change in the building size of the m-th type of population in cell i between adjacent time periods. imt ; ω is the intercept of cell i in time period t, and x and y are the centroid coordinates of that cell;im (x,y,t) represents the regression coefficient of the population building size of the m-th type of building in the i-th neighborhood during time period t; ε it Let be the residual of cell i during time period t.

[0063] The variables, spatial geographic information (centroid coordinates of the residential area), and temporal information are input into a spatiotemporal weighted regression model for calculation. If there are n time periods, u residential areas, and k building categories, and each residential area in each time period is considered a sample, then there are a total of n×u sample points. The parameters in the model... The solution formula is as follows:

[0064]

[0065] In the formula, Let i be the parameter matrix of cell i during time period t, and the matrix size is (k+1)×1; (k+1); Let u be the matrix of independent variables, and let its size be (u×n)×(k+1). The dependent variable matrix has a size of (u×n)×1; W i (x,y,t)=diag[α ij α ij+1 ...... α i(u×n )] is the spatiotemporal matrix of the sample points of cell i in time period t; the size of the matrix is ​​(u×n)×(u×n).

[0066] αij is the spatiotemporal weighting coefficient, and its expression is as follows:

[0067]

[0068] In the formula, Let i be the spatial distance between sample point i and sample point j. h is the time distance between sample point i and sample point j. s h t These are spatial bandwidth and temporal bandwidth, respectively.

[0069] Spatial bandwidth h s and time bandwidth h t The value of is determined by the minimum cross-validation method, as shown in the following expression:

[0070]

[0071] In the formula, f j f j (h s ,h t ) represent the true value at sample point j and the value within the spatial bandwidth h, respectively. s and time bandwidth h t The predicted value is below.

[0072] Furthermore, based on the community's traffic convenience index, fusion weight, and comprehensive importance index, the change in the number of people entering each building category in adjacent time periods is determined. A spatiotemporal characteristic prediction model for the change in the number of people entering is then determined by combining the fusion weight and the comprehensive importance index through spatiotemporal geographic weighted regression. This spatiotemporal characteristic prediction model for the change in the number of people entering is expressed as follows:

[0073]

[0074] In the formula, ΔA it The change in the number of people entering cell i between time period t and time period tT; COT it For the convenience of transportation in the community during time period t; κ represents the combined weight of the change in population and building area of ​​community i during time period t and the traffic convenience index; imt ψ represents the importance of the change in the size of building type m in community i during time period t to the change in population outflow. im (x,y,t) These are the regression coefficients of the m-th type of population building in community i during time period t and the regression coefficients of the traffic convenience of community i during time period t, respectively. The intercept is cell i in time period t; δ it Let t be the residual of cell i during time period t.

[0075] In some embodiments, the community traffic convenience index is calculated as follows: the community traffic convenience during time period t can be reflected by the road area ratio and the traffic flow status during time period tT. The extracted built environment data and road traffic flow are input into the expression, which is as follows:

[0076]

[0077] In the formula, This is the ratio of the road area within community i to the total area of ​​the community. It is the ratio of the sum of the capacity of all roads in community i to the sum of the traffic volume of all roads during time period tT, where l is the road number.

[0078] In some embodiments, the steps for determining the comprehensive importance index are as follows: using building size population change ΔSPBA imt As a characteristic variable, the change in the number of people flowing into each building type in adjacent time periods, ΔA it The target variable is the change in population based on building size ΔSPBA. imt and the change in the number of people entering each building type during adjacent time periods, ΔA itThe data is input into a random forest model to determine the importance index corresponding to each building category. The importance index includes one or more of the mean accuracy decline coefficient (MDA) and the mean Gini coefficient (MDG). The product of the mean accuracy decline coefficient and the mean Gini coefficient is normalized and used as the comprehensive importance index corresponding to each building category.

[0079] The formula for normalizing the product of the average accuracy decrease coefficient and the average Gini coefficient is as follows:

[0080]

[0081] In the formula, MDA imt ×MDG imt The product of the average accuracy decrease and the average Gini coefficient for the change in building size of the m-th population category in neighborhood i during time period t, (MDA) it ×MDG it ) min Let MDA be the minimum of the products of the average accuracy decrease and the average Gini coefficient of the population and building size changes in community i during time period t. it ×MDG it ) max The maximum value is the product of the average accuracy decrease and the average Gini coefficient of the population and building scale changes of various types in community i during time period t.

[0082] In some embodiments, the steps for determining the fusion weights are as follows: Calculate the weighted average change in population and building area using the importance composite index obtained through the above steps. and the weighted average change in population and buildings Convenience of transportation in the community (COT) it Change in the number of people entering the country compared to adjacent time periods ΔA it correlation coefficient The weighted calculation of changes in building population and community transportation convenience indicators is based on the calculation of the combined weights. The expression is as follows:

[0083]

[0084]

[0085] In the formula, These are the correlation coefficients between the weighted average change in population and building area of ​​community i from time period tT to time period t and the change in population inflow of community i during time period t, respectively, and the correlation coefficients between the traffic convenience index of community i from time period tT to time period t and the change in population inflow of community i during time period t.

[0086] Finally, the parameters of the prediction model for the spatiotemporal characteristics of changes in the number of inflows are solved, using the same method as the prediction model for changes in the number of outflows.

[0087] Step S30: Calculate the dynamic variables for the next period based on the dynamic variable iterative model, correct the dynamic variables for the next period based on the correction model, and determine the corrected dynamic variables for the next period. The dynamic variables represent the changes in population and building scale corresponding to each building category.

[0088] The dynamic variables of the initial time period t0 are input into the crowd flow characteristic model to obtain... The outflow and inflow numbers for cell i during the following time period are obtained using the following formulas:

[0089]

[0090]

[0091] In the formula, The number of people leaving cell i during time period t0+T; This represents the number of people leaving cell i during time period t0;

[0092] The number of people entering cell i during time period t0+T; This represents the number of people entering cell i during time period t0.

[0093] In some embodiments, the dynamic variables for the next time period before correction, ΔN, are calculated based on the dynamic variable iterative model. imt-T There is a certain degree of correlation, and the size of the building is also an important factor affecting the change in the number of people in the building. The dynamic variable iterative model formula is expressed as follows:

[0094]

[0095] In the formula, For the original t o +T time period and t o Change in the number of people in building type m in community i during time period; S is the coefficient; im Let m be the building area of ​​the m-th type of building in community i; This represents the change in the number of people in building type m in community i between time period t and time period tT. This is the intercept.

[0096] The dynamic variables for the next time period are corrected based on the correction model to determine the corrected dynamic variables for the next time period. The correction model is expressed as follows:

[0097]

[0098]

[0099] In the formula, This represents the change in the number of people in building type m in community i between time period t0+T and time period t; This represents the change in the number of people in cell i between time period t0 and time period t0-T.

[0100] Step S40: Input the corrected changes in population building size corresponding to each building category in the next time period, the changes in outflow and inflow of each building category in the initial time period into the spatiotemporal feature prediction model of population flow, so as to determine the outflow and inflow of each building category within a preset time range, and determine the spatiotemporal features of population flow based on the outflow and inflow of each building category, and complete the prediction of the spatiotemporal features of population flow corresponding to each building category.

[0101] After correcting the dynamic variables for the next time period, the changes in population building size corresponding to each building category in the next time period, the changes in outflow and inflow of each building category in the initial time period, and the changes in inflow of each building category are input into the spatiotemporal feature prediction model of population flow. The above steps are repeated until the outflow and inflow of each building category within the preset time range are determined. Based on the outflow and inflow of each building category, the spatiotemporal features of population flow are determined, and the spatiotemporal feature prediction of population flow corresponding to each building category is completed.

[0102] As can be seen from the above embodiments, compared with the prior art, this application addresses the problems in the prior art where the prediction model does not consider the spatiotemporal heterogeneity of crowd flow, the input dependent variable is a static variable, resulting in relatively fixed prediction results, or only predicting crowd flow for a single mode of transportation. This application has, but is not limited to, the following beneficial effects:

[0103] First, the prediction method is based on a spatiotemporal geographic weighted regression model, which takes into account the spatiotemporal heterogeneity of population movement and uses dynamic independent variables (population size change and transportation convenience) as input variables, thus making up for the shortcomings of previous prediction models that used static independent variables as input variables, resulting in relatively fixed prediction results.

[0104] Secondly, the spatiotemporal characteristic prediction model for changes in the number of inflows combines importance comprehensive indicators and fusion weights to emphasize the impact of each variable on the target value and improve the interpretability of the model.

[0105] Third, the dynamic variable iterative model contained in this application significantly reduces the amount of data required in the prediction stage, which can greatly improve the prediction efficiency of spatiotemporal characteristics of the population, and greatly improve the prediction speed while ensuring its prediction accuracy.

[0106] Fourth, this application overcomes the limitation of previous prediction models that are only applicable to the prediction of population flow under a single mode of transportation. It can predict the spatiotemporal characteristics of population flow under multiple modes of transportation.

[0107] Based on any embodiment of this application, the step of determining the change in population and building scale according to the change in the number of people in buildings corresponding to each building category and the building area includes:

[0108] Obtain the change in the number of people in buildings corresponding to each building category and the building area corresponding to each building category in adjacent time periods;

[0109] The change in population and building size for each building category is determined by multiplying the change in the number of people in each building category for adjacent time periods by the building area for each building category.

[0110] Specifically, the population-building scale change is determined based on the changes in the number of people in buildings corresponding to each building category and the building area. First, based on the time step, community, and building division, the changes in the inflow and outflow of people in the community, the total population change, and the population change corresponding to each building category are calculated in adjacent time periods. Then, the population-building scale change is obtained by multiplying the population change corresponding to each building category in adjacent time periods by its corresponding building area. The expression is as follows:

[0111] ΔSPBA imt =ΔN imt ×S im

[0112] Among them, the change in population and building scale ΔSPBA between adjacent time periods imt ;ΔN imt S represents the change in the number of people in building type m in community i between time period t and time period tT; im Let be the building area of ​​the m-th type of building in community i.

[0113] Based on any embodiment of this application, the step of determining the change in the number of outflowing people corresponding to each building category in adjacent time periods according to the change in population building size includes:

[0114] A spatiotemporal characteristic prediction model for determining the outflow population change corresponding to each building category based on population and building scale changes;

[0115] The changes in population and building size in adjacent time periods are input into the spatiotemporal feature prediction model of the change in outflow population to determine the change in outflow population corresponding to each building category in adjacent time periods.

[0116] Specifically, a spatiotemporal characteristic prediction model for the change in the number of people leaving the city is determined based on changes in population and building scale. This model uses the change in the number of people leaving the city in adjacent time periods as the dependent variable and the change in population and building scale as the independent variable. The expression for the spatiotemporal characteristic prediction model for the change in the number of people leaving the city is as follows:

[0117]

[0118] Wherein, ΔP it ΔSPBA represents the change in the number of people leaving cell i between time period t and time period tT; and the change in the building size of the m-th type of population in cell i between adjacent time periods. imt ; ω is the intercept of cell i in time period t, and x and y are the centroid coordinates of that cell; im (x,y,t) represents the regression coefficient of the population building size of the m-th type of building in the i-th neighborhood during time period t; ε it Let be the residual of cell i during time period t;

[0119] The changes in population and building size in adjacent time periods are input into the spatiotemporal feature prediction model of the change in outflow population to determine the change in outflow population corresponding to each building category in adjacent time periods.

[0120] Based on any embodiment of this application, the step of determining the change in the number of people entering each building category in adjacent time periods according to the fusion weight and the comprehensive importance index includes:

[0121] In response to the instruction to construct a spatiotemporal characteristic prediction model for changes in the number of inflows, the importance comprehensive index, fusion weight, and community traffic convenience index corresponding to each building category are determined.

[0122] A spatiotemporal feature prediction model for the change in the number of inflows is constructed based on the comprehensive importance index, fusion weight, and community transportation convenience index corresponding to each building category.

[0123] Based on the spatiotemporal feature prediction model, the change in the number of people entering each building category is determined.

[0124] Based on any embodiment of this application, the step of determining the comprehensive importance index corresponding to each building category includes:

[0125] Using the change in population size of buildings as the feature variable and the change in the number of people entering each building category in adjacent time periods as the target variable, the change in population size of buildings and the change in the number of people entering each building category in adjacent time periods are input into the random forest model to determine the importance index corresponding to each building category. The importance index includes one or more of the average accuracy decline coefficient and the average Gini coefficient.

[0126] The product of the average accuracy decrease coefficient and the average Gini coefficient, after being normalized, is used as the comprehensive importance index corresponding to each building category.

[0127] Specifically, the steps to determine the comprehensive importance index are as follows: using the change in population based on building size ΔSPBA imt As a characteristic variable, the change in the number of people flowing into each building type in adjacent time periods, ΔA it The target variable is the change in population based on building size ΔSPBA. imt and the change in the number of people entering each building type during adjacent time periods, ΔA it The data is input into a random forest model to determine the importance index corresponding to each building category. The importance index includes one or more of the mean accuracy decline coefficient (MDA) and the mean Gini coefficient (MDG). The product of the mean accuracy decline coefficient and the mean Gini coefficient is normalized and used as the comprehensive importance index corresponding to each building category.

[0128] The formula for normalizing the product of the average accuracy decrease coefficient and the average Gini coefficient is as follows:

[0129]

[0130] In the formula, MDA imt ×MDG imt The product of the average accuracy decrease and the average Gini coefficient for the change in building size of the m-th population category in neighborhood i during time period t, (MDA) it ×MDG it ) min Let MDA be the minimum of the products of the average accuracy decrease and the average Gini coefficient of the population and building size changes in community i during time period t. it ×MDG it ) max The maximum value is the product of the average accuracy decrease and the average Gini coefficient of the population and building scale changes of various types in community i during time period t.

[0131] Based on any embodiment of this application, the dynamic variable iteration model is as follows: The For the original t o +T time period and t o The change in the number of people in building type m in community i during time period i, the As a coefficient, the S im Let m be the building area of ​​the m-th type of building in community i. Let be the change in the number of people in building type m in community i between time period t and time period tT. This is the intercept.

[0132] As can be seen from the above embodiments, the dynamic variable iterative model included in this application significantly reduces the amount of data required in the prediction stage, greatly improves the prediction efficiency of spatiotemporal characteristics of the population, and greatly improves the prediction speed while ensuring its prediction accuracy.

[0133] Based on any embodiment of this application, the modified model is as follows: as well as Among them, the For the corrected t o The change in the number of people in building type m in community i between time period +T and time period t, the For t o Time period and t o -Change in the number of people in cell i during time period T.

[0134] Please see Figure 2This application provides a population flow spatiotemporal characteristic prediction device, comprising a population change determination module 1100, a flow population change determination module 1200, a dynamic variable determination module 1300, and a population flow prediction module 1400. The population change determination module 1100 is configured to respond to a population flow spatiotemporal characteristic prediction command, determine the change in the number of people in buildings corresponding to each building category and the building area, and determine the population-building scale change based on the change in the number of people in buildings corresponding to each building category and the building area. The flow population change determination module 1200 is configured to determine the change in the number of people leaving each building category in adjacent time periods based on the population-building scale change, and determine the change in the number of people entering each building category in adjacent time periods based on fusion weights and importance comprehensive indicators. The dynamic variable determination module 1300 is configured to calculate the dynamic variables for the next time period based on a dynamic variable iterative model, and adjust the dynamic variables based on the modified model. The dynamic variables for the next time period are corrected to determine the corrected dynamic variables for the next time period. The dynamic variables represent the changes in population and building scale corresponding to each building category. The population flow prediction module 1400 is configured to input the corrected changes in population and building scale corresponding to each building category for the next time period, the changes in outflow and inflow of people corresponding to each building category for the initial time period into the population flow spatiotemporal feature prediction model to determine the outflow and inflow of people corresponding to each building category within a preset time range. Based on the outflow and inflow of people corresponding to each building category, the spatiotemporal features of population flow are determined, and the spatiotemporal feature prediction of population flow corresponding to each building category is completed.

[0135] Based on any embodiment of this application, please refer to Figure 3 Another embodiment of this application also provides an electronic device, which can be implemented by a computer device, such as... Figure 3 The diagram shows the internal structure of a computer device. The computer device includes a processor, a computer-readable storage medium, a memory, and a network interface connected via a system bus. The computer-readable storage medium stores an operating system, a database, and computer-readable instructions. The database may store control information sequences. When the computer-readable instructions are executed by the processor, the processor can implement a method for predicting the spatiotemporal characteristics of crowd flow. The processor of the computer device provides computing and control capabilities, supporting the operation of the entire computer device. The memory of the computer device may store computer-readable instructions. When the computer-readable instructions are executed by the processor, the processor can execute the spatiotemporal characteristic prediction method for crowd flow of this application. The network interface of the computer device is used for communication with a terminal. Those skilled in the art will understand that… Figure 3The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0136] In this embodiment, the processor is used to execute... Figure 2 The system contains the specific functions of each module and its sub-modules, and the memory stores the program code and various data required to execute these modules or sub-modules. The network interface is used for data transmission between the user terminal and the server. In this embodiment, the memory stores the program code and data required to execute all modules / sub-modules in the population flow spatiotemporal characteristic prediction device of this application, and the server can call the server's program code and data to execute the functions of all sub-modules.

[0137] This application also provides a storage medium storing computer-readable instructions, which, when executed by one or more processors, cause the one or more processors to perform the steps of the population flow spatiotemporal feature prediction method described in any embodiment of this application.

[0138] This application also provides a computer program product, including a computer program / instructions that, when executed by one or more processors, implement the steps of the population flow spatiotemporal feature prediction method described in any embodiment of this application.

[0139] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments of this application can be implemented by a computer program instructing related hardware. This computer program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the methods described above. The aforementioned storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc.

[0140] The above description is only a partial embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

[0141] In summary, as can be seen from the above embodiments, compared with the prior art, this application addresses the problems of existing prediction models not considering the spatiotemporal heterogeneity of crowd flow, having static input dependent variables leading to relatively fixed prediction results, or only predicting crowd flow for a single mode of transportation. This application includes, but is not limited to, the following beneficial effects:

[0142] First, the prediction method is based on a spatiotemporal geographic weighted regression model, which takes into account the spatiotemporal heterogeneity of population movement and uses dynamic independent variables (population size change and transportation convenience) as input variables, thus making up for the shortcomings of previous prediction models that used static independent variables as input variables, resulting in relatively fixed prediction results.

[0143] Secondly, the spatiotemporal characteristic prediction model for changes in the number of inflows combines importance comprehensive indicators and fusion weights to emphasize the impact of each variable on the target value and improve the interpretability of the model.

[0144] Third, the dynamic variable iterative model contained in this application significantly reduces the amount of data required in the prediction stage, which can greatly improve the prediction efficiency of spatiotemporal characteristics of the population, and greatly improve the prediction speed while ensuring its prediction accuracy.

[0145] Fourth, this application overcomes the limitation of previous prediction models that are only applicable to the prediction of population flow under a single mode of transportation. It can predict the spatiotemporal characteristics of population flow under multiple modes of transportation.

Claims

1. A method for predicting the spatiotemporal characteristics of population flow, characterized in that, include: In response to the command to predict the spatiotemporal characteristics of population flow, determine the change in the number of people in buildings and the building area corresponding to each building category, and determine the change in population building scale based on the change in the number of people in buildings and the building area corresponding to each building category. The change in outflow of people for each building category in adjacent time periods is determined based on the change in population and building size. The change in inflow of people for each building category in adjacent time periods is determined based on the community traffic convenience index, integration weight, and comprehensive importance index. The expression for the community traffic convenience index is as follows: ； in, This is the ratio of the road area within community i to the total area of ​​the community. It is the ratio of the sum of the capacity of all roads within community i to the sum of the traffic volume of all roads during time period tT. Number the road; The dynamic variables for the next time period are calculated based on the dynamic variable iterative model. The dynamic variables for the next time period are then corrected based on the correction model to determine the corrected dynamic variables for the next time period. These dynamic variables represent the changes in population and building scale corresponding to each building category. The dynamic variable iterative model is as follows: The Before the correction Time period and Time period Community No. Changes in the number of people in buildings of this type, the , As a coefficient, the for Community No. The building area of ​​the type of building, the for Time period and Time period Community No. Changes in the number of people in buildings of this type The intercept is; the corrected model is as well as , wherein For the revised Time period and Time period Community No. Changes in the number of people in buildings of this type, the for Time period and Time period Changes in the number of residents in the community; The revised changes in population building size for each building category in the next time period, the changes in outflow and inflow for each building category in the initial time period are input into the spatiotemporal feature prediction model for population flow to determine the outflow and inflow for each building category within a preset time range. Based on the outflow and inflow for each building category, the spatiotemporal features of population flow are determined, and the spatiotemporal feature prediction of population flow for each building category is completed.

2. The method for predicting the spatiotemporal characteristics of population flow according to claim 1, characterized in that, The steps for determining the change in population and building size based on the changes in the number of people and building area corresponding to each building category include: Obtain the change in the number of people in buildings corresponding to each building category and the building area corresponding to each building category in adjacent time periods; The change in population and building size for each building category is determined by multiplying the change in the number of people in each building category for adjacent time periods by the building area for each building category.

3. The method for predicting the spatiotemporal characteristics of population flow according to claim 1, characterized in that, The steps for determining the change in the number of people leaving each building category in adjacent time periods based on the change in population and building size include: A spatiotemporal characteristic prediction model for determining the outflow population change corresponding to each building category based on population and building scale changes; The changes in population and building size in adjacent time periods are input into the spatiotemporal feature prediction model of the change in outflow population to determine the change in outflow population corresponding to each building category in adjacent time periods.

4. The method for predicting the spatiotemporal characteristics of population flow according to claim 1, characterized in that, The steps for determining the changes in the number of people entering each building category in adjacent time periods based on the community's traffic convenience index, integration weight, and comprehensive importance index include: In response to the instruction to construct a spatiotemporal characteristic prediction model for changes in the number of inflows, the importance comprehensive index, fusion weight, and community traffic convenience index corresponding to each building category are determined. A spatiotemporal feature prediction model for the change in the number of inflows is constructed based on the comprehensive importance index, fusion weight, and community transportation convenience index corresponding to each building category. Based on the spatiotemporal feature prediction model, the change in the number of people entering each building category is determined.

5. The method for predicting the spatiotemporal characteristics of population flow according to claim 4, characterized in that, The steps for determining the comprehensive importance index corresponding to each building category include: Using the change in population size of buildings as the feature variable and the change in the number of people entering each building category in adjacent time periods as the target variable, the change in population size of buildings and the change in the number of people entering each building category in adjacent time periods are input into the random forest model to determine the importance index corresponding to each building category. The importance index includes one or more of the average accuracy decline coefficient and the average Gini coefficient. The product of the average accuracy decrease coefficient and the average Gini coefficient, after being normalized, is used as the comprehensive importance index corresponding to each building category.

6. A device for predicting the spatiotemporal characteristics of crowd flow, characterized in that, include: The population change determination module is configured to respond to the spatiotemporal characteristics prediction command of population flow, determine the change in the number of people in buildings and the building area corresponding to each building category, and determine the change in population building scale based on the change in the number of people in buildings and the building area corresponding to each building category. The module for determining the change in the number of people moving in and out is configured to determine the change in the number of people moving out for each building category in adjacent time periods based on the change in population and building size, and to determine the change in the number of people moving in for each building category in adjacent time periods based on the community's traffic convenience index, integration weight, and comprehensive importance index. The expression for the community's traffic convenience index is as follows: ； in, This is the ratio of the road area within community i to the total area of ​​the community. It is the ratio of the sum of the capacity of all roads within community i to the sum of the traffic volume of all roads during time period tT. Number the road; The dynamic variable determination module is configured to calculate the dynamic variables for the next time period based on a dynamic variable iterative model, correct the dynamic variables for the next time period based on a correction model, and determine the corrected dynamic variables for the next time period. The dynamic variables represent the changes in population and building scale corresponding to each building category. The dynamic variable iterative model is... The Before the correction Time period and Time period Community No. Changes in the number of people in buildings of this type, the , As a coefficient, the for Community No. The building area of ​​the type of building, the for Time period and Time period Community No. Changes in the number of people in buildings of this type The intercept is; the corrected model is as well as , wherein For the revised Time period and Time period Community No. Changes in the number of people in buildings of this type, the for Time period and Time period Changes in the number of residents in the community; The crowd flow prediction module is configured to input the corrected changes in population and building size corresponding to each building category in the next time period, the changes in outflow and inflow of each building category in the initial time period, into the crowd flow spatiotemporal feature prediction model to determine the outflow and inflow of each building category within a preset time range. Based on the outflow and inflow of each building category, the module determines the spatiotemporal features of crowd flow and completes the prediction of the spatiotemporal features of crowd flow corresponding to each building category.

7. An electronic device comprising a central processing unit and a memory, characterized in that, The central processing unit is used to invoke and run a computer program stored in the memory to perform the steps of the method as described in any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, It stores, in the form of computer-readable instructions, a computer program implemented according to any one of claims 1 to 5, which, when invoked by a computer, executes the steps included in the corresponding method.

Images