Underground space value evaluation method fusing multi-source space big data

Abstract

The invention relates to the technical field of asset management, in particular to an underground space value evaluation method fusing multi-source space big data, and the method comprises the steps: obtaining the multi-source space big data in real time, carrying out the time-space fusion processing of the obtained multi-source space big data, and generating a time-space data set of a target underground space; calculating a plurality of evaluation indexes of the target underground space based on the spatio-temporal data set, and predicting the value of the target underground space by using the plurality of evaluation indexes; and comprehensively evaluating the plurality of evaluation indexes through a machine learning model, and correcting the predicted value based on a comprehensive evaluation result to obtain a final evaluation value.

Description

Technical Field

[0001] This invention relates to the field of asset management technology, and in particular to a method for assessing the value of underground space by integrating multi-source spatial big data. Background Technology

[0002] Assessing the value of underground space is a comprehensive issue involving multiple disciplines and factors. Assessment dimensions include economic value (the most common), socio-environmental benefits, and strategic location value. Multi-source spatial big data has revolutionized underground space valuation, transforming the traditional "static, partial, and hypothetical" analysis into a "dynamic, panoramic, and predictable" precise decision support. Multi-source big data primarily consists of urban operational big data, underground space structure big data, environmental big data, and socio-economic and point-of-interest (POI) data. Urban operational big data can identify pedestrian flow patterns and activity patterns; for example, Wi-Fi data can describe the scale and duration of pedestrian flow in above-ground functional areas at different times. Combining POI data with pedestrian flow data allows for analysis of the matching needs between underground space and above-ground functional areas.

[0003] Existing technologies typically employ a multi-layered system approach to assess the value of underground space using multi-source big data. Specifically, this involves: collecting and fusing multi-source big data through a data layer; analyzing the multi-source big data using models in an analysis layer; and visually outputting the assessment results through an application layer to support decision-making.

[0004] However, the evaluation process suffers from problems such as the use of a single model, isolated and static evaluation indicators, and an inability to capture the relationships between different evaluation indicators, resulting in inaccurate value assessment. Summary of the Invention

[0005] This invention improves the reliability and accuracy of value estimation by calculating multiple evaluation indicators and using the fusion analysis of these indicators to predict the value of target underground space.

[0006] The technical solution proposed in this invention is: a method for assessing the value of underground space by integrating multi-source spatial big data, the method comprising: Real-time acquisition of multi-source spatial big data; spatiotemporal fusion processing of the acquired multi-source spatial big data to generate a spatiotemporal dataset of the target underground space; Based on spatiotemporal datasets, multiple evaluation indicators of the target underground space are calculated, and the value of the target underground space is predicted using these multiple evaluation indicators. The machine learning model is used to comprehensively evaluate multiple evaluation indicators, and the predicted value is corrected based on the comprehensive evaluation results to obtain the final evaluation value.

[0007] Preferably, the real-time acquisition of multi-source spatial big data, and the spatiotemporal fusion processing of the acquired multi-source spatial big data to generate a spatiotemporal dataset of the target underground space, includes: Acquire spatial data, including urban GIS road network data and BIM drawing data of the target underground space; Acquire dynamic pedestrian flow data, including Wi-Fi or Bluetooth probe data and mobile phone signaling data; Acquire environmental and operational data, including data on the distribution of business interest points and business types, sales / rental data, and weather data; The process of performing spatiotemporal fusion processing on the obtained multi-source spatial big data to generate a spatiotemporal dataset of the target underground space includes: Unified time reference means unifying the acquired spatial data, dynamic pedestrian flow data, and environmental and operational data under the same urban coordinate system; The unified spatial benchmark aligns the acquired spatial data, dynamic pedestrian flow data, and environmental and operational data to UTC+8, i.e., Beijing time, on the timeline, with a granularity of 15 minutes. After preprocessing multi-source spatial big data with unified time and spatial benchmarks, a spatiotemporal dataset for spatial value assessment is constructed.

[0008] Preferably, the step of calculating multiple evaluation indicators of the target underground space based on spatiotemporal datasets, and using these multiple evaluation indicators to predict the value of the target underground space, includes: Calculate the accessibility index of the target underground space within its local area network; Calculate the contribution index of the structural characteristics of the target underground space to the attractiveness of pedestrian flow; Predict the flow of people entering the target underground space and quantify its consumption conversion indicators; The value of target underground space is predicted using accessibility indicators, consumption conversion indicators, and attractiveness contribution indicators.

[0009] Preferably, the accessibility index of the target underground space within its local network includes: Construct a three-dimensional spatial network of the area where the target underground space is located, including: The origin of the flow of people is taken as the source point. Using transfer points or intersections of routes in the transportation network as connection points The target underground space to be evaluated is used as the entrance set as the target point. The key node set of a three-dimensional spatial network is constructed using source points, connection points, and target points. The connection paths of key nodes on the same horizontal plane are used as horizontal edges; each horizontal edge The toll cost is used as the weight of that horizontal edge. ; Using the connection paths of key points on planes at different heights as vertical edges, each vertical edge... The passage cost is used as the weight of the vertical edge. ; The set of critical edges for constructing a three-dimensional spatial network is formed using horizontal and vertical edges. A graph that forms a three-dimensional spatial network using the set of key nodes and the set of key edges. ,in, Represents the set of key points. Denotes the set of key edges; where, ; In the figure The calculation of minimum cumulative toll cost includes: Dijkstra's algorithm, the shortest path algorithm, is used to find the sequence of paths from each source point to the destination point that minimizes the travel cost. ;in, Indicates the first From the source point to the first A set of paths to a target point; , , , These represent the number of source points and the number of target points, respectively. Therefore, the minimum cumulative toll cost is ;in, Representing a path One of the edges in; For the three-dimensional reachability index of a target point ;in, Indicates the first The weights of each source point Indicates the first Distance attenuation coefficient for each mode of transportation; Indicates the weight of traffic modes. Indicates the number of traffic modes; Integrating multiple source points to obtain multiple overall target underground space three-dimensional accessibility indicators .

[0010] Preferably, the method for calculating the accessibility index of the target underground space within its local network further includes: Considering traffic conditions at different times, the distance decay index for each mode of transportation changes over time, that is: For a target point 3D accessibility index of time ; express Time of the first Distance attenuation coefficient for each mode of transportation; Obtaining the time series of 3D accessibility indicators ;in, Indicates the evaluation period; Based on the time series of three-dimensional accessibility indicators, the three-dimensional accessibility indicators of multiple overall target underground spaces are recalculated to obtain dynamic accessibility indicators. .

[0011] Preferably, the contribution index of the structural characteristics of the target underground space to the attractiveness of pedestrian flow includes: Obtain the size index of the target underground space, which reflects the capacity of the target underground space, including: Calculate the size index of the target underground space ,in, Represents the normalized function. This indicates the total leasable area of ​​the target underground space. Indicates the net height of the public area of ​​the target underground space; Indicates the weight of area indicators. Indicates the weight of the height indicator; Obtain the connectivity efficiency index of the target underground space, including: Based on the BIM drawing data of the target underground space, the passageway portion in the target underground space plan is divided into multiple grids, with the center of each grid being a node, and nodes connected by straight line segments; the topological depth of any node is calculated. ,in Represents a node To the node The shortest topological step count; Integration level of each node ;in, Indicates the number of nodes; The average integration degree of the target underground space is then: ; Obtain the functional mixing index of the target underground space, including: Acquire data on shops within the target underground space and categorize the shops; Statistics on the proportion of each type of shop area to the total leasable area. ;in, express The area of ​​the shop-like premises; based on Calculate the entropy value of the distribution of shop types. ; Attracting attractiveness contribution indicators ;in, , and These represent the normalized scale exponent, average integration degree, and entropy value. , , This represents the weighting coefficient for the fusion of indicators.

[0012] Preferably, the prediction of pedestrian flow into the target underground space and the quantification of its consumption conversion indicators include: Predict the flow of people entering the target underground space, including: Based on dynamic pedestrian flow data Passing the target point at all times Nearby pedestrian traffic ; The interception rate at the target point ; People flow from the destination Utility function when entering the target underground space ; Represents a constant term. , , Indicates the weighting coefficient. Indicates the characteristics of a time period. ; Indicates the weather condition index. or Utility function for those who have not entered the target underground space ; Predict from the target point The flow of people entering the target underground space is ; The total number of people entering the entire target underground space is ; Quantifying consumption conversion indicators of people entering the target underground space, including: Consumption conversion index ;in, Indicates consumption. This indicates entry into the target underground space; This indicates the probability of entering the target underground space for consumption; This represents the consumer propensity index; ; The constant term of the consumer index, , , This represents the weighting coefficient of the consumer index. Indicates real-time crowd density. Indicates the area of ​​the target underground space; The predicted number of consumers entering the target underground space is: ; The method of predicting the value of target underground space using accessibility indicators, consumption conversion indicators, and attractiveness contribution indicators includes: get Forecasted consumption of similar shops ; in, Average spending per person at any time , This represents the adjustment coefficient for the characteristics of the time period. express Average sales volume and premium coefficient of similar shops ; Indicates premium weight; The value of target underground space is predicted based on consumption expenditure, including: Calculate the annual revenue of the target underground space ;in, This represents the annual spending of all shops. Indicates the rent-to-price ratio. This represents the annual operating expenses of all shops; ; express The average number of operating days per year for a shop; The commercial value of the target underground space is ,in, Indicates the capitalization rate. , Indicates the risk-free interest rate. This represents the risk coefficient of underground space, and is usually taken as a value greater than 1. Indicates market risk premium; It represents the long-term growth rate, which is the sum of the inflation rate and the urban development premium.

[0013] Preferably, the step of comprehensively evaluating multiple evaluation indicators using a machine learning model, and correcting the predicted value based on the comprehensive evaluation results to obtain the final evaluation value includes: , , Construct indicator vectors ; The index vector is input into a pre-trained deep neural network, which outputs a comprehensive evaluation index. ,include: The deep neural network includes an input layer, a feature extraction layer, an attention layer, and an output layer; Attention weights of attention layer ;in, Represents the index vector of the first Embedded representation of each indicator; , , Indicates learnable parameters; Output comprehensive evaluation index ;in, Represents the index vector of the first One indicator; The final assessed value is ,in, This represents the correction factor. This represents the benchmark index.

[0014] Preferably, it also includes: optimizing three-dimensional accessibility based on the structural characteristics of the target underground space, including: Will Extended to the cost of traveling from the source point to the target point and reaching the target location within the target underground space. ,in, Indicates from the target point To the target location inside the underground space Navigation costs; target location Nodes in the grid As an alternative location to the target location, i.e. ; in, Indicates the effective pathfinding speed. ; Indicates the base walking speed. Gain coefficient; optimized stereo reachability ; The value of the target underground space is re-predicted using optimized three-dimensional accessibility, consumption conversion indicators, and attractiveness contribution indicators.

[0015] A computer-readable storage medium storing a computer program that is executed by a processor to implement the aforementioned method for assessing the value of underground space by integrating multi-source spatial big data.

[0016] The beneficial effects of this invention are: 1. This invention constructs a three-dimensional accessibility system that accurately reflects the spatial relationships of a three-dimensional city. Based on time-varying three-dimensional accessibility, it can accurately describe the impact of travel costs (time) at different time periods on the three-dimensional accessibility index. Furthermore, considering that the coupling relationship between the structural characteristics of the target underground space and the three-dimensional accessibility index affects the final value assessment, the endpoint of the arrival cost assessment is expanded from reaching the target point to reaching the interior of the "target underground space," thereby obtaining the full-chain cost from external transportation to internal navigation. This more accurately reflects the arrival cost of the target underground space and ensures the accuracy of the final value estimate.

[0017] 2. This invention uses a deep learning network to learn the nonlinear relationships between different evaluation indicators and outputs a comprehensive evaluation index. The comprehensive evaluation is used to correct the value prediction value to reduce valuation error. This can mitigate the adverse effects of evaluation indicator errors on valuation when data is scarce, thus reducing the evaluation process's dependence on data. Attached Figure Description

[0018] Figure 1 This is a flowchart of a method for assessing the value of underground space that integrates multi-source spatial big data, as proposed in this invention. Detailed Implementation

[0019] The following description is intended to disclose the present invention and enable those skilled in the art to implement it. The preferred embodiments described below are merely examples, and other obvious modifications will occur to those skilled in the art. The basic principles of the invention defined in the following description can be applied to other embodiments, modifications, improvements, equivalents, and other technical solutions that do not depart from the spirit and scope of the invention.

[0020] It is understood that the term "a" should be understood as "at least one" or "one or more," that is, in one embodiment, the number of an element can be one, while in another embodiment, the number of the element can be multiple, and the term "a" should not be understood as a limitation on the number. Example

[0021] refer to Figure 1 The technical solution provided by this invention is: a method for assessing the value of underground space by integrating multi-source spatial big data, the method comprising: Step 1: Acquire multi-source spatial big data in real time, perform spatiotemporal fusion processing on the acquired multi-source spatial big data, and generate a spatiotemporal dataset of the target underground space. This includes the following steps: Acquire spatial data, including urban GIS road network data and BIM drawing data of the target underground space; Acquire dynamic pedestrian flow data, including Wi-Fi / Bluetooth probe data and mobile phone signaling data; Acquire environmental and operational data, including data on the distribution of business interest points and business types, sales / rental data, and weather data; The process of performing spatiotemporal fusion processing on the obtained multi-source spatial big data to generate a spatiotemporal dataset of the target underground space includes: Unified time reference means unifying the acquired spatial data, dynamic pedestrian flow data, and environmental and operational data under the same urban coordinate system; The unified spatial benchmark aligns the acquired spatial data, dynamic pedestrian flow data, and environmental and operational data to UTC+8, i.e., Beijing time, on the timeline, with a granularity of 15 minutes. After preprocessing multi-source spatial big data with unified time and spatial benchmarks, a spatiotemporal dataset for spatial value assessment is constructed.

[0022] Step 2: Calculate multiple evaluation indicators for the target underground space based on the spatiotemporal dataset, and use these indicators to predict the value of the target underground space. This includes the following steps: Step 2.1: Calculate the accessibility index of the target underground space in its regional network to quantify the regional advantages of the target underground space.

[0023] Specifically, the origin of the pedestrian flow is taken as the source point. For example, the center of a residential community, the entrance of an office building, or a bus stop.

[0024] Using transfer points or intersections of routes in the transportation network as connection points For example, subway platforms, intersections, and junctions of underground passages.

[0025] The target underground space to be evaluated is the set of entrances and the target point. For example, entrances and exits of underground shopping malls, subway turnstiles, and entrances to sunken plazas that connect to the ground.

[0026] The key node set of a three-dimensional spatial network is constructed using source points, connection points, and target points; The horizontal edges are defined by the connecting paths of key nodes on the same horizontal plane; for example, sidewalks, streets, and underpasses. The toll cost is used as the weight of that horizontal edge. In this embodiment, travel time is selected as the travel cost.

[0027] The vertical edges are the paths connecting key points at different heights on a plane; for example, stairs, escalators, and elevators. The passage cost is used as the weight of the vertical edge. .

[0028] The critical edge set of a three-dimensional spatial network is constructed using horizontal and vertical edges, and the graph of the three-dimensional spatial network is constructed using the critical node set and the critical edge set. ,in, Represents the set of key points. Denotes the set of key edges; where, .

[0029] In the figure The minimum cumulative toll cost is calculated as follows: Dijkstra's algorithm, the shortest path algorithm, is used to find the sequence of paths from each source point to the destination point that minimizes the travel cost. ;in, Indicates the first From the source point to the first A set of paths to a target point; , , , These represent the number of source points and the number of target points, respectively. Therefore, the minimum cumulative toll cost is ;in, Representing a path One of the edges in; We use the classic "potential model" in geography to aggregate the contributions of all source points to the target point. This model posits that the accessibility of a location is the sum of the influences of all opportunity points (source points) around it, and that this influence diminishes as distance (travel cost) increases.

[0030] For the three-dimensional reachability index of a target point ;in, Indicates the first The weight of a source point, such as population density or number of jobs, determines its importance or potential contribution in accessibility calculations. Indicates the first The distance attenuation coefficient for each mode of transportation determines the sensitivity of travel costs to accessibility; the larger the value, the smaller the accessibility. For walking, the value is usually 1.5-2.5. Indicates the weight of transportation modes (walking, cycling, driving, etc.). Indicates the number of traffic modes.

[0031] Integrating multiple source points, multiple overall target underground space three-dimensional accessibility indicators .

[0032] Step 2.2: Calculate the contribution index of the structural characteristics of the target underground space to the attraction of pedestrian flow, in order to quantify the internal attractiveness of the target underground space. The goal of this step is to quantify the contribution of the space's own structural characteristics, that is, its physical form and functional layout, to attracting pedestrian flow. Specifically, this includes the following steps: Obtain the size index of the target underground space, which reflects the capacity of the target underground space, including: Calculate the size index of the target underground space ,in, This represents the standardization function (Min-Max standardization function). This indicates the total leasable area of ​​the target underground space. Indicates the net height of the public area of ​​the target underground space; This indicates the weight of the area indicator (e.g., 0.7). This indicates the weight of the height indicator (e.g., 0.3). The clear height of public areas can be obtained from the BIM drawing data of the target underground space.

[0033] The connectivity efficiency index within the target underground space is obtained to reflect the degree of boundary movement of people within the underground space. This is quantified using space syntax theory, including: Based on the BIM drawing data of the target underground space, the passage section in the plan of the target underground space is divided into multiple grids, with the center of each grid being a node. The nodes are connected by straight line segments, thus simplifying the plan of the target underground space into an axis diagram.

[0034] Calculate the topology depth of any node ,in Represents a node To the node The shortest topological step count; Integration level of each node ;in, Indicates the number of nodes. The higher the value, the closer the node is to the center in the target underground space plane, and the easier it is to reach. The average integration degree of the target underground space is calculated as follows: ; Obtain the functional mix index of the target underground space to reflect the diversity of business types, as diversity promotes dwell time and cross-consumption, including: Acquire data on shops within the target underground space and categorize the shops; Statistics on the proportion of each type of shop area to the total leasable area. ;in, express The area of ​​the shop-like premises; based on Calculate the entropy value of the distribution of shop types. The larger the entropy value, the more uniform the distribution of shop types, that is, the more uniform the distribution of business formats and the higher the degree of mixing.

[0035] Attracting attractiveness contribution indicators ;in, , and These represent the normalized scale exponent, average integration degree, and entropy value. , , This represents the weighting coefficient for the fusion of indicators.

[0036] Step 2.3: Predict the flow of people entering the target underground space and quantify its consumption conversion indicators to predict the potential for conversion of people into consumption.

[0037] Not all pedestrians passing the target point will enter the target underground space. This step aims to quantify the probability of entry and the conversion rate to consumption. Based on the discrete choice model of random effects theory, it is assumed that when pedestrian n passes near the entrance, he faces two mutually exclusive choices: enter the underground space (denoted as option ). a ) and continue passing through (denoted as the plan) b Each choice has a utility, consisting of deterministic and random components. For entering the underground space, the utility is... For those who venture underground, the effect is... .in, , Indicates a definite item. , This represents a random item. Based on this, step 2.3 specifically includes the following steps: Predict the flow of people entering the target underground space, including: Based on dynamic pedestrian flow data Passing the target point at all times Nearby pedestrian traffic ; The interception rate at the target point That is, the probability that the flow of people will enter the target underground space. ; People flow from the destination Utility function when entering the target underground space ; Represents a constant term. , , Indicates the weighting coefficient. Indicates the characteristics of a time period. The lunch break is 1, the breakfast break is 0, and the dinner break is 0.5.

[0038] Indicates the weather condition index. or That is, 0 for sunny days and 1 for rainy days. This is because the probability of people entering is relatively higher on rainy days compared to sunny days.

[0039] Utility function for not entering the target underground space Uncertainties are not considered in this embodiment.

[0040] Predict from the target point The flow of people entering the target underground space is The total number of people entering the target underground space is .

[0041] Quantifying consumption conversion indicators of people entering the target underground space, including: Because consumption conversion is a binary outcome—consumption is 1 and no consumption is 0—its probability follows a logistic regression function, specifically: Consumption conversion index ;in, Indicates consumption. This indicates entry into the target underground space; This indicates the probability of entering the target underground space for consumption; This represents the consumer propensity index; ; The constant term of the consumer index, , , This represents the weighting coefficient of the consumer index. Indicates real-time crowd density. Indicates the area of ​​the target underground space; The predicted number of consumers entering the target underground space is: .

[0042] Step 2.4: Predict the value of the target underground space using accessibility indicators, consumption conversion indicators, and attractiveness contribution indicators. This specifically includes the following steps: get Forecasted consumption of similar shops ; in, Average spending per person at any time , This represents the adjustment coefficient for time period characteristics (e.g., during lunchtime, for restaurants, the coefficient is 0.3). express Average sales volume and premium coefficient of similar shops ; This represents the premium weight, which can be obtained through regression analysis of historical sales data. In other words, the higher the historical sales, the higher the premium weight.

[0043] The value of target underground space is predicted based on consumption expenditure, including: Calculate the annual revenue of the target underground space ;in, This represents the annual spending of all shops. Indicates the rent-to-price ratio. This represents the annual operating expenses of all shops; ; express The average number of operating days per year for a shop; The commercial value of the target underground space is ,in, Indicates the capitalization rate. , Indicates the risk-free interest rate. This indicates the risk coefficient of underground space (greater than 1). Indicates market risk premium; It represents the long-term growth rate, which is the sum of the inflation rate and the urban development premium.

[0044] Step 3: Use a machine learning model to comprehensively evaluate multiple evaluation indicators, and revise the predicted value based on the comprehensive evaluation results to obtain the final evaluation value.

[0045] , , Construct indicator vectors ; The index vector is input into a pre-trained deep neural network, which outputs a comprehensive evaluation index. ,include: The deep neural network includes an input layer, a feature extraction layer with 128 neurons, an attention layer, and an output layer; Attention weights of attention layer ;in, Represents the index vector of the first Embedded representation of each indicator; , , Indicates learnable parameters; Output comprehensive evaluation index ;in, Represents the index vector of the first One indicator; The final assessed value is ,in, This represents the correction factor. This represents the benchmark index.

[0046] By using deep neural networks to non-linearly couple the various indicators, the problem of neglecting the non-linear correlation between indicators when using direct linear coupling is solved. At the same time, attention weights represent the contribution of each indicator, making the generation of the comprehensive evaluation index interpretable.

[0047] Example 2: In Example 1, the three-dimensional accessibility index is a fixed value, meaning it only considers the three-dimensional accessibility index for a fixed time period, which is suitable for scenarios with simple traffic conditions. For scenarios with complex traffic conditions (where the target underground space is located in a busy area), it is necessary to consider the time-varying distance decay index of each mode of transportation. Therefore, based on Example 1, we propose the following technical solution: For a target point 3D accessibility index of time ; express Time of the first Distance attenuation coefficient for each mode of transportation; Obtaining the time series of 3D accessibility indicators ;in, Indicates the evaluation period; Based on the time series of three-dimensional accessibility indicators, the three-dimensional accessibility indicators of multiple overall target underground spaces are recalculated to obtain dynamic accessibility indicators. . Example

[0048] In Example 1, the calculation of vertical accessibility did not consider the coupling between vertical accessibility and the underground space structure. For example, entrance layout and internal passage structure can improve vertical accessibility, thereby ultimately increasing commercial value, especially in the early design and later renovation processes. Therefore, we propose a technical solution to optimize vertical accessibility based on the structural characteristics of the target underground space, specifically: Will Extended to the cost of traveling from the source point to the target point and reaching the target location within the target underground space. ,in, Indicates from the target point To the target location inside the underground space Navigation costs; target location Nodes in the grid As an alternative location to the target location, i.e. ; in, Indicates the effective pathfinding speed. ; This indicates the basic walking speed, such as 80 meters per minute.

[0049] Gain coefficient, representing the proportion by which the topology increases walking speed; The higher the altitude, the more straightforward the layout of the interior space, and the less detours are required.

[0050] Optimized 3D accessibility The value of the target underground space is re-predicted using optimized three-dimensional accessibility, consumption conversion indicators, and attractiveness contribution indicators, as detailed in the previous steps.

[0051] The present invention also provides a computer-readable storage medium storing a computer program, which is executed by a processor to implement the aforementioned method for assessing the value of underground space by integrating multi-source spatial big data.

[0052] The processes described above with reference to the flowcharts in the embodiments disclosed in this invention can be implemented as computer software programs. The embodiments disclosed in this invention include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication component, and / or installed from a removable medium. When the computer program is executed by a central processing unit (CPU), it performs the functions defined in the methods of this application. It should be noted that the computer-readable medium described above in this application can be a computer-readable signal medium or a computer-readable storage medium, or any combination of the two. Computer-readable storage media can be, for example, but not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatuses, or devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wire segments, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this application, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in connection with an instruction execution system, apparatus, or device. In this application, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can also be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on a computer-readable medium may be transmitted using any suitable medium, including but not limited to: wireless segments, wire segments, optical fibers, RF, etc., or any suitable combination thereof.

[0053] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0054] Those skilled in the art should understand that the embodiments of the present invention described above and shown in the accompanying drawings are merely examples and do not limit the present invention. The purpose of the present invention has been fully and effectively achieved. The functions and structural principles of the present invention have been shown and explained in the embodiments. Without departing from the principles described, the implementation of the present invention may have any changes or modifications.

Claims

1. A method for assessing the value of underground space by integrating multi-source spatial big data, characterized in that, The method includes: Real-time acquisition of multi-source spatial big data; spatiotemporal fusion processing of the acquired multi-source spatial big data to generate a spatiotemporal dataset of the target underground space; Based on spatiotemporal datasets, multiple evaluation indicators of the target underground space are calculated, and the value of the target underground space is predicted using these multiple evaluation indicators. The machine learning model is used to comprehensively evaluate multiple evaluation indicators, and the predicted value is corrected based on the comprehensive evaluation results to obtain the final evaluation value.

2. The method for assessing the value of underground space by integrating multi-source spatial big data as described in claim 1, characterized in that, The real-time acquisition of multi-source spatial big data, followed by spatiotemporal fusion processing of the acquired multi-source spatial big data to generate a spatiotemporal dataset of the target underground space, includes: Acquire spatial data, including urban GIS road network data and BIM drawing data of the target underground space; Acquire dynamic pedestrian flow data, including Wi-Fi or Bluetooth probe data and mobile phone signaling data; Acquire environmental and operational data, including data on the distribution of business interest points and business types, sales / rental data, and weather data; The process of performing spatiotemporal fusion processing on the obtained multi-source spatial big data to generate a spatiotemporal dataset of the target underground space includes: Unified time reference means unifying the acquired spatial data, dynamic pedestrian flow data, and environmental and operational data under the same urban coordinate system; The unified spatial benchmark aligns the acquired spatial data, dynamic pedestrian flow data, and environmental and operational data to UTC+8, i.e., Beijing time, on the timeline, with a granularity of 15 minutes. After preprocessing multi-source spatial big data with unified time and spatial benchmarks, a spatiotemporal dataset for spatial value assessment is constructed.

3. The method for assessing the value of underground space by integrating multi-source spatial big data according to claim 2, characterized in that, The process involves calculating multiple evaluation indicators for the target underground space based on a spatiotemporal dataset, and using these indicators to predict the value of the target underground space, including: Calculate the accessibility index of the target underground space within its local area network; Calculate the contribution index of the structural characteristics of the target underground space to the attractiveness of pedestrian flow; Predict the flow of people entering the target underground space and quantify its consumption conversion indicators; The value of target underground space is predicted using accessibility indicators, consumption conversion indicators, and attractiveness contribution indicators.

4. The method for assessing the value of underground space by integrating multi-source spatial big data according to claim 3, characterized in that, The accessibility index of the target underground space within its local network includes: Construct a three-dimensional spatial network of the area where the target underground space is located, including: The origin of the flow of people is taken as the source point. Using transfer points or intersections of routes in the transportation network as connection points The target underground space to be evaluated is used as the entrance set as the target point. The key node set of a three-dimensional spatial network is constructed using source points, connection points, and target points. The connection paths of key nodes on the same horizontal plane are used as horizontal edges; each horizontal edge The toll cost is used as the weight of that horizontal edge. ; Using the connection paths of key points on planes at different heights as vertical edges, each vertical edge... The passage cost is used as the weight of the vertical edge. ; The set of critical edges for constructing a three-dimensional spatial network is formed using horizontal and vertical edges. A graph that forms a three-dimensional spatial network using the set of key nodes and the set of key edges. ,in, Represents the set of key points. Denotes the set of key edges; where, ; In the figure The calculation of minimum cumulative toll cost includes: Dijkstra's algorithm, the shortest path algorithm, is used to find the sequence of paths from each source point to the destination point that minimizes the travel cost. ;in, Indicates the first From the source point to the first A set of paths to a target point; , , , These represent the number of source points and the number of target points, respectively. Therefore, the minimum cumulative toll cost is ;in, Representing a path One of the edges in; For the three-dimensional reachability index of a target point ;in, Indicates the first The weights of each source point Indicates the first Distance attenuation coefficient for each mode of transportation; Indicates the weight of traffic modes. Indicates the number of traffic modes; Integrating multiple source points to obtain multiple overall target underground space three-dimensional accessibility indicators .

5. The method for assessing the value of underground space by integrating multi-source spatial big data according to claim 4, characterized in that, The accessibility index of the target underground space within its local network also includes: Considering traffic conditions at different times, the distance decay index for each mode of transportation changes over time, that is: For a target point 3D accessibility index of time ; express Time of the first Distance attenuation coefficient for each mode of transportation; Obtaining the time series of 3D accessibility indicators ;in, Indicates the evaluation period; Based on the time series of three-dimensional accessibility indicators, the three-dimensional accessibility indicators of multiple overall target underground spaces are recalculated to obtain dynamic accessibility indicators. .

6. The method for assessing the value of underground space by integrating multi-source spatial big data as described in claim 5, characterized in that, The calculation of the contribution of the structural characteristics of the target underground space to the attractiveness of pedestrian flow includes: Obtain the size index of the target underground space, which reflects the capacity of the target underground space, including: Calculate the size index of the target underground space ,in, Represents the normalized function. This indicates the total leasable area of ​​the target underground space. Indicates the net height of the public area of ​​the target underground space; Indicates the weight of area indicators. Indicates the weight of the height indicator; Obtain the connectivity efficiency index of the target underground space, including: Based on the BIM drawing data of the target underground space, the passageway portion in the target underground space plan is divided into multiple grids, with the center of each grid being a node, and nodes connected by straight line segments; the topological depth of any node is calculated. ,in Represents a node To the node The shortest topological step count; Integration level of each node ;in, Indicates the number of nodes; The average integration degree of the target underground space is then: ; Obtain the functional mixing index of the target underground space, including: Acquire data on shops within the target underground space and categorize the shops; Statistics on the proportion of each type of shop area to the total leasable area. ;in, express The area of ​​the shop-like premises; based on Calculate the entropy value of the distribution of shop types. ; Attracting attractiveness contribution indicators ;in, , and These represent the normalized scale exponent, average integration degree, and entropy value. , , This represents the weighting coefficient for the fusion of indicators.

7. The method for assessing the value of underground space by integrating multi-source spatial big data as described in claim 6, characterized in that, The predicted flow of people entering the target underground space and the quantification of its consumption conversion indicators include: Predict the flow of people entering the target underground space, including: Based on dynamic pedestrian flow data Passing the target point at all times Nearby pedestrian traffic ; The interception rate at the target point ; People flow from the destination Utility function when entering the target underground space ; Represents a constant term. , , Indicates the weighting coefficient. Indicates the characteristics of a time period. ; Indicates the weather condition index. or Utility function for those who have not entered the target underground space ; Predict from the target point The flow of people entering the target underground space is ; The total number of people entering the entire target underground space is ; Quantifying consumption conversion indicators of people entering the target underground space, including: Consumption conversion index ;in, Indicates consumption. This indicates entry into the target underground space; This indicates the probability of entering the target underground space for consumption; This represents the consumer propensity index; ; The constant term of the consumer index, , , This represents the weighting coefficient of the consumer index. Indicates real-time crowd density. Indicates the area of ​​the target underground space; The predicted number of consumers entering the target underground space is: ; The method of predicting the value of target underground space using accessibility indicators, consumption conversion indicators, and attractiveness contribution indicators includes: get Forecasted consumption of similar shops ; in, Average spending per person at any time , This represents the adjustment coefficient for the characteristics of the time period. express Average sales volume and premium coefficient of similar shops ; Indicates premium weight; The value of target underground space is predicted based on consumption expenditure, including: Calculate the annual revenue of the target underground space ;in, This represents the annual spending of all shops. Indicates the rent-to-price ratio. This represents the annual operating expenses of all shops; ; express The average number of operating days per year for a shop; The commercial value of the target underground space is ,in, Indicates the capitalization rate. , Indicates the risk-free interest rate. Indicates the risk coefficient of underground space. Indicates market risk premium; It represents the long-term growth rate, which is the sum of the inflation rate and the urban development premium.

8. The method for assessing the value of underground space by integrating multi-source spatial big data according to claim 7, characterized in that, The process of comprehensively evaluating multiple evaluation indicators using a machine learning model, and then revising the predicted value based on the comprehensive evaluation results to obtain the final evaluation value includes: , , Construct indicator vectors ; The index vector is input into a pre-trained deep neural network, which outputs a comprehensive evaluation index. ,include: The deep neural network includes an input layer, a feature extraction layer, an attention layer, and an output layer; Attention weights of attention layer ;in, Represents the index vector of the first Embedded representation of each indicator; , , Indicates learnable parameters; Output comprehensive evaluation index ;in, Represents the index vector of the first One indicator; The final assessed value is ,in, This represents the correction factor. This represents the benchmark index.

9. The method for assessing the value of underground space by integrating multi-source spatial big data according to claim 8, characterized in that, Also includes: Optimize three-dimensional accessibility based on the structural characteristics of the target underground space, including: Will Extended to the cost of traveling from the source point to the target point and reaching the target location within the target underground space. ,in, Indicates from the target point To the target location inside the underground space Navigation costs; target location Nodes in the grid As an alternative location to the target location, i.e. ; in, Indicates the effective pathfinding speed. ; Indicates the base walking speed. Gain coefficient; optimized stereo reachability ; The value of the target underground space is re-predicted using optimized three-dimensional accessibility, consumption conversion indicators, and attractiveness contribution indicators.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, which is executed by a processor to implement the underground space value assessment method that integrates multi-source spatial big data as described in any one of claims 1-9.

Images