Public service facility blind area and low-efficiency area identification and facility addition method and system based on knowledge graph

Abstract

The application discloses a kind of public service facilities blind area and inefficient area identification and facility addition method and system based on knowledge graph, the present situation data of target city is integrated, and target city graph is constructed with spatial form element and present situation public service facility element as entity, and spatial distance is relation, and the service range subgraph of each public service facility is filtered based on target city graph, and then service blind area and inefficient area are identified.The application also aims at service blind area and inefficient area, and the entity node in multidimensional element intelligent filtering subgraph is selected as the site of new facility, and the result is interactively displayed.The application can help to accurately grasp the correlation characteristics between population and space elements and service facilities by constructing multidimensional knowledge graph, and can more scientifically and objectively identify service blind area and inefficient area.Meanwhile, the comprehensive site selection based on multidimensional elements of the application is more reliable, the optimization of sub-item site selection is more comprehensive, and the three-dimensional interactive display formed is more stereoscopic and intuitive.

Description

TECHNICAL FIELD

[0001] The application relates to a knowledge graph-based public service facility blind area and low-efficiency area identification and facility addition method and system, and belongs to the fields of knowledge graph and geographic information processing. BACKGROUND

[0002] Public service facilities (cultural facilities, education and scientific research facilities, medical and health care facilities, etc.) play an important role in improving the quality of life of residents, optimizing resource allocation and promoting the sustainable development of cities. By comprehensively analyzing the diversified needs and complex structure of the city, the optimization of resources can be achieved, and the continuous optimization and improvement of the urban public service system can be promoted.

[0003] At present, most of the public service facility site selection research relies on the analysis and decision of indicators such as supply and demand relationship and accessibility. However, due to the complex factors involved in the site selection of public service facilities, the commonly used model algorithms such as K-means clustering and shortest path algorithm cannot fully consider the multi-dimensional factors and complex urban structure, and such models rely on partial indicators for facility layout and site selection, which is difficult to fully consider the multi-dimensional factors and complex urban structure. Therefore, the scientificity and efficiency of the site selection of public service facilities are limited, and it is also difficult to fully solve the actual problems. SUMMARY

[0004] The first purpose of the present application is to provide a knowledge graph-based public service facility blind area and low-efficiency area identification method and system, which can integrate the current city data through the knowledge graph, identify the service blind area and low-efficiency area of the newly added public service facilities, and more objectively and accurately understand the current situation. The second purpose of the present application is to provide a knowledge graph-based public service facility addition method and system, which can intelligently recommend the site selection point of the newly added facilities and interactively display the identified service blind area and low-efficiency area, and improve the scientificity and efficiency of the site selection of public service facilities.

[0005] Technical scheme: The purpose of the present application can be achieved by the following technical scheme:

[0006] In the first aspect, the present application provides a knowledge graph-based public service facility blind area and low-efficiency area identification method, which comprises the following steps:

[0007] Obtain the current data of the target city, including the spatial form and spatial position data of the city, the current public service facility data and the population activity data in units of blocks; take the spatial form elements and the current public service facility elements as entities of the knowledge graph, and take the spatial distance between the entities as the relationship, to construct the graph of the target city;

[0008] Based on the constructed target city graph, the path length from each public service facility entity to all other spatial form element entities is calculated, entities with path length less than the service range of the public service facility are screened as the service range subgraph of each public service facility;

[0009] All entities not covered by the service range subgraph of the public service facility are identified and marked as service blind areas, and the corresponding entities and relationships constitute the present facility service blind area subgraph;

[0010] In the service range subgraph of each public service facility, the total population corresponding to the block entities contained therein is calculated, and the subgraph with population greater than the service carrying capacity of the corresponding facility is identified and marked as the present facility service inefficient area.

[0011] Further, the construction of the target city graph comprises:

[0012] Firstly, a graph mode layer is constructed, the entities are defined as spatial form elements and present public service facilities, the entity relationship is defined as the spatial distance between entities, the entity attributes of spatial form elements are defined as their spatial form, spatial position and population, and the entity attribute values of spatial form elements are area, perimeter, shape index, longitude and latitude of geometric center and population, the entity attributes of present public service facilities are their service range and service carrying capacity, and the entity attribute values of public service facilities are the service range radius size and the maximum number of service carrying capacity;

[0013] Secondly, a graph data layer is constructed, and the data is cleaned and arranged into a structured data format of triplets of "entity-relationship-entity" and "entity-attribute-attribute value";

[0014] Finally, the structured triplets in the data layer are imported into the system using a knowledge graph operating system, and the target city graph is established according to the rules defined in the mode layer.

[0015] Further, the path length calculation method from each public service facility entity to all other spatial form element entities is as follows: based on a GIS platform, a road network topology structure is established by importing road network data, a shortest path algorithm is adopted, the public service facility entity is taken as the starting point, and the other spatial form element entity is taken as the end point, and the path length from the public service facility entity to the other spatial form element entity is calculated.

[0016] In the second aspect, the present application provides a public service facility adding method based on a knowledge graph, which comprises the following steps after the present facility service blind area and the present facility service inefficient area are obtained by the public service facility blind area and inefficient area identification method based on a knowledge graph:

[0017] For the service blind area of the existing facilities, according to the demand degree, influence, accessibility and comprehensive screening of four types of entity nodes in the existing facility service blind area subgraph as the new facility site; Among them, the block with the highest daily activity number is taken as the demand degree entity node; The node with the largest characteristic vector centrality is taken as the influence entity node; The node with the shortest average value of the path to each block is taken as the accessibility entity node; The point with the highest comprehensive value is taken as the comprehensive entity node by using AHP hierarchical analysis method;

[0018] For the service inefficient area of the existing facilities, the same method is used to screen four types of entity nodes from the service range subgraph of the corresponding public service facilities, judge the spatial relationship between the entity nodes and the existing public service facilities, if they are located in the same block or the distance is less than the set value, delete the entity node, select the second entity node and judge again, until the spatial position relationship condition is met, the entity node meeting the condition is taken as the new facility site of the existing facility service inefficient area;

[0019] The three-dimensional space mapping is carried out by using the constructed target city graph and the geographic coordinates of the nodes, and the three-dimensional model of the target city is combined to form a three-dimensional sand table; The four types of facility site selection points and related calculation results are interactively displayed in different colors; The user determines the final site selection of public service facilities by interactive selection and operation.

[0020] Further, the screening method of the four types of entity nodes comprises:

[0021] The daily activity number of each block is calculated by combining the street crowd activity data collected by the video probe for multiple consecutive days, and the block entity node with the highest daily activity number is marked as the demand degree entity node;

[0022] The characteristic vector centrality of each node in the subgraph is calculated by using the eigenvalue decomposition algorithm, and the entity node with the largest characteristic vector centrality is taken as the influence entity node, and the characteristic vector centrality formula is calculated as follows:

[0023]

[0024] Wherein, C(i) represents the characteristic centrality of entity node i, λ is the maximum eigenvalue of the adjacency matrix, N(i) represents the neighbor entity node set of entity node i in the subgraph, A ij The connection weight between entity node i and entity node j in the adjacency matrix is set as the ratio of the degree product of adjacent nodes to the sum of the degrees of all nodes, which reflects the connection activity degree of the node;

[0025] The Dijkstra algorithm is used to calculate the path length of each node in the subgraph to all blocks, and the average value of the path length is calculated, and the entity node with the shortest average value of the path length is selected as the reachability entity node;

[0026] The AHP hierarchical analysis method is used to determine the weights of the influence, reachability and demand degree, and the comprehensive calculation formula is as follows:

[0027] S i =ω1v 1i +ω2ν 2i +ω3ν 3i

[0028] Wherein ω1、ω2、ω3 respectively represent the weights of influence, reachability and demand degree, ν 1i , ν 2i , ν 3i Respectively, the score of entity node i under the three criteria of influence, reachability and demand degree, and then select the entity node with the highest comprehensive value as the comprehensive entity node.

[0029] Thirdly, the application provides a public service facility blind area and low efficiency area identification system based on a knowledge graph, comprising:

[0030] A target city graph construction module is used to obtain target city status data, including city space form and space position data, status public service facility data and population activity data in units of blocks;Space form elements and status public service facility elements are used as entities of the knowledge graph, and the spatial distance between entities is used as the relationship, to construct a target city graph;

[0031] A public service facility service range subgraph screening module is used to calculate the path length from each public service facility entity to all other space form element entities based on the constructed target city graph, and to screen entities with path length less than the service range of the public service facility as the service range subgraph of each public service facility;

[0032] A service blind area identification module is used to identify all entities not covered by the service range subgraph of the public service facility and mark them as service blind areas, and the corresponding entities and relationships constitute the status facility service blind area subgraph;

[0033] And a service low efficiency area identification module is used to calculate the total population of the block entities contained in the service range subgraph of each public service facility, to identify the subgraph with population greater than the service carrying capacity of the corresponding facility and mark it as the status facility service low efficiency area.

[0034] In a fourth aspect, the present application provides a public service facility adding system based on a knowledge graph, comprising each module in the public service facility blind area and inefficient area identification system based on a knowledge graph, and

[0035] A service blind area facility adding module is configured to screen four types of entity nodes in a current facility service blind area subgraph as new facility site selection points according to demand degree, influence, accessibility and comprehensiveness; wherein, the block with the highest daily activity number is taken as the demand degree entity node; the node with the maximum characteristic vector centrality is taken as the influence entity node; the node with the shortest average path to each block is taken as the accessibility entity node; the point with the highest comprehensive value is taken as the comprehensive entity node by using the AHP hierarchical analysis method;

[0036] A service inefficient area facility adding module is configured to screen four types of entity nodes from the service range subgraph of the corresponding public service facility according to the same method, judge the spatial relationship between the entity node and the current public service facility, if they are located in the same block or the distance is less than a set value, delete the entity node, select the second entity node and judge again until the spatial position relationship condition is met, and take the entity node meeting the condition as the new facility site selection point of the current facility service inefficient area.

[0037] An interactive display module is configured to perform three-dimensional space mapping by using the constructed target city graph and the geographic coordinates of the nodes thereof, form a three-dimensional sand table in combination with the three-dimensional model of the target city, and interactively display the four types of facility site selection points and the related calculation results in different colors; and a user determines the final site selection of the public service facility by interactive selection and operation.

[0038] In a fifth aspect, the present application provides a computer system comprising a memory, a processor and a computer program / instruction stored on the memory and executable on the processor, wherein the computer program / instruction is executed by the processor to realize the steps of the public service facility blind area and inefficient area identification method based on a knowledge graph, or the steps of the public service facility adding method based on a knowledge graph.

[0039] In a sixth aspect, the present application provides a computer readable storage medium, wherein the computer readable storage medium stores a computer program, and the computer program is executed by a processor to realize the steps of the public service facility blind area and inefficient area identification method based on a knowledge graph, or the steps of the public service facility adding method based on a knowledge graph.

[0040] In a seventh aspect, the present application provides a computer program product comprising computer programs / instructions, which, when executed by a processor, implement the steps of the method for identifying blind areas and inefficient areas of public service facilities based on a knowledge graph, or the steps of the method for adding public service facilities based on a knowledge graph.

[0041] Advantages: Compared with the prior art, the present application has the following advantages:

[0042] 1. The present application helps to accurately grasp the correlation characteristics between the urban population and space elements and the public service facilities by constructing a multi-dimensional knowledge graph of the target city, not only can identify the service blind area of the public service facilities, but also can identify the service inefficient area by combining the facility carrying capacity and the corresponding population number, so that the identification process is more comprehensive.

[0043] 2. The present application can determine possible facility site selection points through crowd activity data collected by video, accessibility data obtained by various algorithms, etc., solve the problem of large error in traditional facility site selection determined according to part of the index, make the whole process more objective, intelligent and automatic.

[0044] 3. The present application selects the site of the public service facility intelligently based on the knowledge graph, and obtains a relatively optimal solution by interactive selection and optimization through a three-dimensional sand table, which improves the speed by 300% compared with the traditional manual site selection method.

[0045] 4. The present application displays and outputs the site selection list by interacting with the final facility site selection, refines the traditional conceptual site selection guidance of pictures, texts, etc., and cooperates with the three-dimensional city model, so that the subsequent site selection process is more intuitive, efficient and controllable, and the guidance for specific projects is stronger. BRIEF DESCRIPTION OF DRAWINGS

[0046] Figure 1 is a method flowchart of embodiment 1 of the present application.

[0047] Figure 2 is a method flowchart of embodiment 2 of the present application.

[0048] Figure 3 is a method flowchart of embodiment 3 of the present application.

[0049] Figure 4 is a schematic diagram of the shortest path algorithm. DETAILED DESCRIPTION

[0050] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0051] Example 1

[0052] like Figure 1 As shown in the figure, an embodiment of the present invention discloses a method for identifying blind spots and inefficient areas of public service facilities based on knowledge graphs, which includes the following steps:

[0053] Step S101: Obtain current data of the target city, including urban spatial form and location data, current public service facility data, and population activity data at the street level; construct the target city map by using spatial form elements and current public service facility elements as entities in the knowledge graph and the spatial distance between entities as relationships.

[0054] Specifically, the construction of the target city map includes:

[0055] First, construct the graph pattern layer, define entities as spatial morphological elements and existing public service facilities, define entity relationships as spatial distances between entities, define the entity attributes of spatial morphological elements as their spatial morphology, spatial location and population size, define the entity attribute values ​​of spatial morphological elements as area, perimeter, shape index, latitude and longitude of geometric center and population size, define the entity attributes of existing public service facilities as their service range and service carrying capacity, and define the entity attribute values ​​of public service facilities as the size of their service range radius and the maximum number of people they can carry.

[0056] Secondly, a graph data layer is constructed, and the cleaned data is organized into a structured data format of "entity-relationship-entity" and "entity-attribute-attribute value" triples;

[0057] Finally, using the knowledge graph operating system, the structured triples in the data layer are imported into the system, and the target city graph is built according to the rules defined in the schema layer.

[0058] Step S102: Based on the constructed target city map, calculate the path length from each public service facility entity to all other spatial morphological element entities, and filter out entities with path lengths shorter than the service range of the public service facilities as service range sub-maps of each public service facility.

[0059] Specifically, the path length calculation method of each public service facility entity to all other spatial form element entities is: based on the GIS platform, the road network topology structure is established by importing the road network data, the shortest path algorithm is adopted, the public service facility entity is taken as the starting point, and the other spatial form element entity is taken as the end point, and the path length of the public service facility entity to the other spatial form element entity is calculated.

[0060] Step S103, all entities not covered by the service range subgraph of the public service facility are identified and marked as service blind area, and the corresponding entities and relationships constitute the current facility service blind area subgraph.

[0061] Step S104, in the service range subgraph of each public service facility, the sum of the population corresponding to the block entities contained is calculated, and the subgraph with a population greater than the service carrying capacity of the corresponding facility is identified and marked as the current facility service inefficient area.

[0062] Embodiment 2

[0063] As shown in Figure 2 The public service facility adding method based on the knowledge graph disclosed in the embodiment of the application is implemented after the current facility service blind area and the current facility service inefficient area in the embodiment 1, and the newly added facility site is recommended for the service blind area and the service inefficient area, and the facility site selection result is interactively optimized and output displayed. Specifically, the following steps are included:

[0064] Step S201, for the current facility service blind area, four types of entity nodes in the current facility service blind area subgraph are screened as newly added facility site selection points according to demand, influence, accessibility and comprehensiveness; wherein the block with the highest daily activity number is taken as the demand entity node; the node with the maximum characteristic vector centrality is taken as the influence entity node; the node with the shortest average path to each block is taken as the accessibility entity node; and the point with the highest comprehensive value is taken as the comprehensive entity node by using the AHP hierarchical analysis method.

[0065] Specifically, the screening method of the above four types of entity nodes includes:

[0066] Combined with the block crowd activity data collected by the video probe for continuous multiple days, the daily activity number of each block is calculated, and the block entity node with the highest daily activity number is marked as the demand entity node;

[0067] The characteristic value decomposition algorithm is used to calculate the characteristic vector centrality of each node in the subgraph, and the entity node with the maximum characteristic vector centrality is taken as the influence entity node, and the characteristic vector centrality formula is calculated as follows:

[0068]

[0069] Wherein, C(i) represents the characteristic centrality of entity node i, λ is the maximum eigenvalue of the adjacency matrix, N(i) represents the set of neighbor entity nodes of entity node i in the subgraph, A ij represents the connection weight between entity node i and entity node j in the adjacency matrix, and the connection weight is set as the ratio of the degree product of adjacent nodes to the sum of the degrees of all nodes, which reflects the connection activity of the node;

[0070] The Dijkstra algorithm is used to calculate the path length of each node in the subgraph to all blocks, and the average value of the path length is calculated, and the entity node with the shortest average value of the path is selected as the reachability entity node;

[0071] The AHP hierarchical analysis method is used to determine the weights of influence, reachability and demand degree, and the comprehensive calculation formula is as follows:

[0072] S i =ω1v 1i +ω2v 2i +ω3v 3i

[0073] Wherein ω1, ω2, ω3 represent the weights of influence, reachability and demand degree respectively, v 1i , v 2i , v 3i are the scores of entity node i under the three criteria of influence, reachability and demand degree, and then the entity node with the highest comprehensive value is selected as the comprehensive entity node.

[0074] Step S202, for the low-efficiency area of the existing facility service, the same method is used to select four types of entity nodes from the service range subgraph of the corresponding public service facility, and the spatial relationship between the entity node and the existing public service facility is judged. If it is located in the same block or the distance is less than a set value, the entity node is deleted, and the second entity node is selected for judgment again until the spatial position relationship condition is met, and the entity node meeting the condition is selected as the new facility site of the low-efficiency area of the existing facility service.

[0075] Step S203, use the constructed target city map and the geographic coordinates of its nodes to perform three-dimensional space mapping, combine the three-dimensional model of the target city to form a three-dimensional sand table; the four types of facility site selected and the related calculation results are interactively displayed in different colors; the user selects and operates through interaction to determine the final site selection of the public service facility.

[0076] Example 3

[0077] As Figure 3As shown, the embodiment of the present application takes Nanjing as an example to illustrate the detailed implementation process of the public service facility blind area and inefficient area identification and facility addition method based on knowledge graph, and the specific steps are as follows:

[0078] Step S301, Nanjing city data acquisition and graph construction. Acquire the present situation data of Nanjing, and integrate the spatial form and spatial position data, population activity data, and present situation public service facility data of Nanjing city respectively. Take the spatial form elements and present situation public service facility elements as entities of the knowledge graph, and take the spatial distance between entities as the relationship. Construct the Nanjing city graph through the knowledge graph operation system.

[0079] Specifically, the present situation data is obtained from Nanjing Urban Planning Bureau, including: Nanjing city spatial form and spatial position data, population activity data, and present situation public service facility data. The Nanjing city spatial form and spatial position data includes: spatial form and spatial position of blocks, plots, buildings, roads, traffic hubs, mountains, and water bodies. The population activity data includes: block population data. The present situation public service facility data includes: service range, service carrying capacity, and spatial position of public service facilities. Then, start to construct the Nanjing city graph. First, construct the graph mode layer. Define the entity as the spatial form elements and present situation public service facilities in the planning site data, define the entity relationship as the spatial distance between entities, define the entity attributes of the spatial form elements as their spatial form, spatial position, and population number, and define the entity attribute values of the spatial form elements as area, perimeter, shape index, longitude and latitude of geometric center, and population number. The entity attributes of the present situation public service facilities are their service range and service carrying capacity, and the entity attribute values of the public service facilities are the service range radius size and the service carrying maximum number. The spatial form elements include blocks, plots, buildings, roads, traffic hubs, mountains, and water bodies. The spatial distance between entities is generated based on the GIS platform through road network data and present situation public service facility data analysis and calculation. The service range radius size and the service carrying maximum number of the public service facilities are determined according to the specific value size of the public service facility category level according to the “Urban Public Service Facility Planning Standard” (GB50442-2008). Second, construct the graph data layer. Clean the planning site data and present situation public service facility data, and arrange the cleaned data into the structured data format of triples. The triple structure is “entity-relationship-entity” and “entity-attribute-attribute value”. Finally, use the knowledge graph operation system, import the structured triples in the data layer into the system, and establish the target city graph according to the rules defined in the mode layer.

[0080] Step S302: Identification of Service Blind Spots and Inefficient Service Areas of Existing Facilities. Based on the Nanjing city map constructed in Step 1, the path length from each public service facility entity to all other spatial morphological element entities is calculated using a DGX computer workstation. Nodes with path lengths shorter than the standard service range of the facility are selected as service range sub-maps for each public service facility. All nodes not covered by these sub-maps are identified and marked as service blind spots. The method for calculating the path length from each public service facility entity to all other spatial morphological element entities is based on a GIS platform. This involves importing road network data to establish a road network topology and employing a shortest path algorithm (such as...). Figure 4 Starting from the public service facility entity and ending at other spatial elements, the path length from the public service facility entity to other spatial elements is calculated. Within each public service facility's service area sub-map, the total population corresponding to the included street nodes is calculated. Sub-maps with populations exceeding the facility's service capacity are identified and marked as inefficient service areas.

[0081] Specifically, the following steps are taken: For the Nanjing city map constructed in Step 1, a road network topology is established by importing road network data from a GIS platform. Using the shortest path algorithm, the path length from public service facility entities to other spatial morphological elements is calculated, starting from the public service facility entity and ending at other spatial morphological elements. Entities with path lengths shorter than the service radius of each existing public service facility are selected. This selected set of entities is marked as the service range sub-map of the public service facility. All entities not covered by the service range sub-map are marked as the service blind spot of the existing facility, and the corresponding entities and relationships constitute the service blind spot sub-map. Then, using the service range sub-maps of each public service facility, street entities within the service range are selected. The population quantity attribute of each street entity is extracted, and the total population corresponding to all street entities is calculated. Referring to the "Urban Public Service Facility Planning Standards," if the total population within a sub-map exceeds the service carrying capacity attribute of the public service facility, the sub-map of that public service facility is marked as an inefficient service area.

[0082] Step S303, service blind area and service low efficiency area add facility site selection. For the service blind area in step two, according to the demand degree, influence, accessibility, comprehensive screening of four types of entity nodes in the subgraph as the new facility site selection point. Among them, the street block with the highest daily activity number is selected as the demand degree entity node; the node with the largest eigenvector centrality calculated by the eigenvalue decomposition algorithm is selected as the influence entity node; the node with the shortest path average value to each street block calculated by the Dijkstra algorithm is selected as the accessibility entity node; the point with the highest comprehensive value is selected as the comprehensive entity node by using the AHP hierarchical analysis method. The above four types of entity nodes are selected as the new facility site selection point of the service blind area of the present situation facility. For the service low efficiency area in step two, repeat the above steps to screen four types of nodes, judge the spatial relationship between the nodes and the present situation public service facility, if they are located in the same street block or the distance is less than 100 meters, delete the node, select the node in the second order instead, and select the entity node as the new facility site selection point of the service low efficiency area of the present situation facility.

[0083] Specifically as follows: for the service blind area of the present situation facility in step two, according to the demand degree, influence, accessibility, comprehensive screening of four types of entity nodes in the subgraph as the new facility site selection point.

[0084] Combined with the street crowd activity data collected by the video probe for seven consecutive days, the daily activity number of each street block is calculated, and the street block entity node with the highest daily activity number is marked as the demand degree entity node.

[0085] The eigenvector centrality of each node in the service blind area graph of the present situation facility is calculated by using the eigenvalue decomposition algorithm, and the entity node with the largest eigenvector centrality is selected as the influence entity node. The formula for calculating the eigenvector centrality is as follows:

[0086]

[0087] Among them, N(i) represents the neighbor entity node set of entity node i in the service blind area graph of the present situation facility, A ij The connection weight between entity node i and entity node j in the adjacency matrix is represented by A The path length of each node in the service blind area subgraph of the present situation facility to all street blocks is calculated by using the Dijkstra algorithm, and the average value of the path length is calculated. The entity node with the shortest path average value is selected as the accessibility entity node. According to the importance of influence, accessibility and demand degree, the two-by-two comparison matrix of the criterion layer is constructed by using the AHP hierarchical analysis method and the expert scoring results. For each comparison matrix, the eigenvector is calculated and normalized to obtain the weight vector. The consistency ratio CR is calculated. If CR is less than 0.1, the consistency test is passed; otherwise, the comparison matrix needs to be adjusted. Finally, the comprehensive calculation formula is as follows:

[0088] S i = ω1v 1i + ω2v 2i + ω3v 3i

[0089] Wherein ω1, ω2, ω3 respectively represent the weight of influence, accessibility, demand degree, v 1i , v 2i , v 3i It is the weight of entity node i under the influence of influence, accessibility, demand degree three criteria, then select the highest comprehensive value of entity node as the comprehensive entity node. The above four kinds of entity nodes are selected as the new facility site of the service blind area of the present situation facility. Then, for the inefficient area of the present situation facility in step two, repeat step three to screen four kinds of entity nodes, and judge the spatial position relationship between the entity node and the present situation public service facility, if the entity node and the present situation public service facility are in the same block or the distance between the two is less than 100 meters, then delete the entity node, and select the second entity node to judge again, until the spatial position relationship condition is met, and the entity node is selected as the new facility site of the inefficient area of the present situation facility.

[0090] Step S304, facility site selection result interactive optimization and output display. The geographical coordinates of the Nanjing city atlas and its nodes constructed in step one are mapped in three-dimensional space, combined with the three-dimensional model of Nanjing city to form a three-dimensional sand table. The four types of site selection points and related calculation results obtained by step three are interactively displayed in different colors. The user determines the final site selection of the public service facility by interactive selection and operation, and integrates the above content into a report in the form of word / excel for output printing.

[0091] Specifically as follows: the geographical coordinates of the Nanjing city atlas and its entity nodes constructed in step one are mapped in three-dimensional space, and combined with the three-dimensional model of Nanjing city to form a three-dimensional sand table of Nanjing city. The new facility site selection points of the present situation facility service blind area and the new facility site selection points of the present situation facility service inefficient area and related calculation results obtained by step three are interactively displayed in different colors on the three-dimensional sand table of Nanjing city. The related calculation results include the demand degree, influence, accessibility, comprehensive value and percentage. The user performs interactive selection and operation on the three-dimensional sand table of Nanjing city to determine the final site selection of the public service facility, and integrates the above content into a report in the form of word / excel for output printing.

[0092] Example 4

[0093] The public service facility blind area and inefficient area identification system based on knowledge graph disclosed by the embodiment of the application comprises:

[0094] a target city graph construction module, configured to acquire target city status data, including city spatial form and spatial position data, status public service facility data, and population activity data in units of blocks; and construct a target city graph by taking spatial form elements and status public service facility elements as entities of the knowledge graph and taking spatial distances between the entities as relationships;

[0095] a public service facility service range subgraph screening module, configured to calculate path lengths from each public service facility entity to all other spatial form element entities based on the constructed target city graph, and screen entities with path lengths less than the service range of the public service facility as service range subgraphs of the public service facility;

[0096] a service blind area identification module, configured to identify all entities not covered by the service range subgraphs of the public service facility and mark them as service blind areas, and the corresponding entities and relationships form status facility service blind area subgraphs;

[0097] and a service inefficient area identification module, configured to calculate the total number of populations corresponding to block entities in the service range subgraphs of each public service facility, identify subgraphs with population numbers greater than the service carrying capacity of the corresponding facility, and mark them as status facility service inefficient areas.

[0098] This embodiment is based on the same inventive concept as embodiment 1, and specific details can be referred to in embodiment 1, which will not be described here.

[0099] Embodiment 5

[0100] The public service facility addition system based on the knowledge graph disclosed in the embodiments of the present application includes the modules in embodiment 4, and the following modules:

[0101] a service blind area facility addition module, configured to screen four types of entity nodes in the status facility service blind area subgraph as new facility site selection points according to demand, influence, accessibility, and comprehensiveness; wherein, the block with the highest number of daily activities is taken as the demand entity node; the node with the maximum characteristic vector centrality is taken as the influence entity node; the node with the shortest average path to each block is taken as the accessibility entity node; the point with the highest comprehensive value is taken as the comprehensive entity node by using the AHP hierarchical analysis method;

[0102] The service inefficient area facility adding module is used for screening four types of entity nodes from the service range subgraph of the corresponding public service facility for the current facility service inefficient area, judging the spatial relationship between the entity node and the current public service facility, deleting the entity node if the entity node is located in the same block as the current public service facility or the distance is less than a set value, selecting the entity node in the second order to judge again until the spatial position relationship condition is met, and taking the entity node meeting the condition as a new facility site selection point of the current facility service inefficient area.

[0103] The interactive display module is used for carrying out three-dimensional space mapping by using the constructed target city graph and the geographic coordinates of nodes thereof, combining a three-dimensional model of the target city to form a three-dimensional sand table, and interactively displaying the four types of facility site selection points and related calculation results in different colors.

[0104] The embodiment and the embodiment 2 are based on the same inventive concept, and specific details are referred to the embodiment 2, and will not be repeated.

[0105] Embodiment 6

[0106] The computer system disclosed by the embodiment of the present application comprises a memory, a processor and a computer program / instruction stored on the memory and executable on the processor, and the computer program / instruction is executed by the processor to realize the steps of the method in the method embodiments.

[0107] Embodiment 7

[0108] The computer readable storage medium disclosed by the embodiment of the present application stores a computer program, and the computer program is executed by the processor to realize the steps of the method in the method embodiments.

[0109] Embodiment 8

[0110] The computer program product disclosed by the embodiment of the present application comprises a computer program / instruction, and the computer program / instruction is executed by the processor to realize the steps of the method in the method embodiments.

Claims

1. A method for identifying blind spots and inefficient areas of public service facilities based on knowledge graphs, characterized in that, The method includes the following steps: Acquire current status data of the target city, including urban spatial morphology and spatial location data, current public service facility data, and population activity data at the street level; construct the target city map by using spatial morphology elements and current public service facility elements as entities in the knowledge graph and spatial distance between entities as relationships. Based on the constructed target city map, the path length from each public service facility entity to all other spatial morphological element entities is calculated, and entities with path lengths shorter than the service range of the public service facilities are selected as the service range sub-maps of each public service facility. Identify all entities not covered by the service scope subgraph of public service facilities and mark them as service blind spots. The corresponding entities and relationships constitute the current facility service blind spot subgraph. In the service area sub-map of each public service facility, calculate the total population corresponding to the block entities it contains, identify sub-maps with a population greater than the service carrying capacity of the corresponding facility, and mark them as inefficient service areas of the current facility. The construction of the target city map includes: First, construct the graph pattern layer, define entities as spatial morphological elements and existing public service facilities, define entity relationships as spatial distances between entities, define the entity attributes of spatial morphological elements as their spatial morphology, spatial location and population size, define the entity attribute values ​​of spatial morphological elements as area, perimeter, shape index, latitude and longitude of geometric center and population size, define the entity attributes of existing public service facilities as their service range and service carrying capacity, and define the entity attribute values ​​of public service facilities as the size of their service range radius and the maximum number of people they can carry. Secondly, a graph data layer is constructed, and the cleaned data is organized into a structured data format of "entity-relationship-entity" and "entity-attribute-attribute value" triples; Finally, using the knowledge graph operating system, the structured triples in the data layer are imported into the system, and the target city graph is built according to the rules defined in the schema layer.

2. The method for identifying blind spots and inefficient areas of public service facilities based on knowledge graphs according to claim 1, characterized in that, The path length from each public service facility entity to all other spatial morphological element entities is calculated as follows: Based on the GIS platform, a road network topology is established by importing road network data. Using the shortest path algorithm, the path length from the public service facility entity to other spatial morphological element entities is calculated, with the public service facility entity as the starting point and other spatial morphological element entities as the ending point.

3. A method for adding public service facilities based on knowledge graphs, characterized in that, After obtaining the current service blind spots and inefficient service areas of public service facilities according to the knowledge graph-based identification method for public service facilities as described in claim 1 or 2, the method includes the following steps: To address the service blind spots of existing facilities, four types of entity nodes in the sub-map of the service blind spots of existing facilities are selected as new facility site selection points based on demand, influence, accessibility, and comprehensiveness. Among them, the street with the highest average daily number of active people is selected as the demand entity node; the node with the largest eigenvector centrality is selected as the influence entity node; the node with the shortest average path to each street is selected as the accessibility entity node; and the point with the highest comprehensiveness value is selected as the comprehensive entity node using the analytic hierarchy process (AHP). For areas where existing facilities serve inefficiently, the same method is used to select four types of entity nodes from the service range sub-map of the corresponding public service facilities. The spatial relationship between the entity node and the existing public service facilities is determined. If the entity node is located in the same block as the existing public service facilities or the distance is less than the set value, the entity node is deleted. The entity node in the second priority is selected and judged again until the spatial location relationship condition is met. The entity node that meets the condition is used as the site selection point for new facilities in areas where existing facilities serve inefficiently. The system utilizes the constructed target city map and the geographic coordinates of its nodes to perform three-dimensional spatial mapping, and combines this with the three-dimensional model of the target city to form a three-dimensional sand table. The four types of facility site selection points and related calculation results are interactively displayed in different colors. Users can determine the final site selection of public service facilities through interactive selection and operation.

4. The method for adding public service facilities based on knowledge graphs according to claim 3, characterized in that, The methods for filtering the four types of entity nodes include: By combining street crowd activity data collected by video cameras over several consecutive days, the average daily number of people active in each street is calculated, and the street entity node with the highest average daily number of people active is marked as the demand entity node. Using the eigenvalue decomposition algorithm, the eigenvector centrality of each node in the subgraph is calculated. The entity node with the largest eigenvector centrality is taken as the influential entity node. The formula for calculating eigenvector centrality is as follows: ； in, Indicate the eigencentrality of entity node i. The largest eigenvalue of the adjacency matrix. This represents the set of neighboring entity nodes of entity node i in the subgraph. This represents the connection weight between entity node i and entity node j in the adjacency matrix. The connection weight is set as the ratio of the product of the degrees of adjacent nodes to the sum of the degrees of all nodes, reflecting the degree of connection activity of the nodes. Dijkstra's algorithm is used to calculate the path length from each node in the subgraph to all blocks, and the average path length is calculated. The entity node with the shortest average path length is selected as the reachability entity node. Using the Analytic Hierarchy Process (AHP), the weights of influence, accessibility, and demand are determined, resulting in the following comprehensive calculation formula: ； in , , , represent the weights of influence, accessibility, and demand, respectively. , , The entity node i is scored according to the three criteria of influence, accessibility, and demand, and then the entity node with the highest comprehensive score is selected as the comprehensive entity node.

5. A knowledge graph-based system for identifying blind spots and inefficient areas in public service facilities, characterized in that, include: The target city mapping module is used to acquire current data of the target city, including urban spatial morphology and location data, current public service facility data, and population activity data at the street level. Using spatial morphological elements and existing public service facilities as entities in a knowledge graph, and spatial distance between entities as relationships, a target city graph is constructed. The construction of the target city graph includes: First, constructing a graph schema layer, defining entities as spatial morphological elements and existing public service facilities, defining entity relationships as spatial distances between entities, defining the entity attributes of spatial morphological elements as their spatial form, spatial location, and population size, and defining the entity attribute values ​​as area, perimeter, shape index, latitude and longitude of the geometric center, and population size; defining the entity attributes of existing public service facilities as their service area and service carrying capacity, and defining the entity attribute values ​​as the radius of their service area and the maximum number of people they can serve; Second, constructing a graph data layer, cleaning and organizing the data into a structured data format of "entity-relationship-entity" and "entity-attribute-attribute value" triples; Finally, using a knowledge graph operating system, importing the structured triples from the data layer into the system, and constructing the target city graph according to the rules defined in the schema layer. The public service facility service area sub-map filtering module is used to calculate the path length from each public service facility entity to all other spatial morphological element entities based on the constructed target city map, and filter entities whose path length is less than the service area of ​​the public service facility as the service area sub-map of each public service facility. The service blind spot identification module is used to identify all entities not covered by the service scope sub-graph of public service facilities and mark them as service blind spots. The corresponding entities and relationships constitute the current facility service blind spot sub-graph. And a service inefficiency zone identification module, used to calculate the total population of the block entities contained in each public service facility's service area sub-map, identify sub-maps where the population is greater than the corresponding facility's service carrying capacity and mark them as current facility service inefficiency zones.

6. A public service facility addition system based on knowledge graphs, characterized in that, Includes the modules in the knowledge graph-based public service facility blind spot and inefficient area identification system according to claim 5, and: The service blind spot facility addition module is used to select four types of entity nodes in the existing facility service blind spot submap as new facility site selection points based on demand, influence, accessibility, and comprehensiveness. Among them, the street with the highest average daily activity is selected as the demand entity node; the node with the largest eigenvector centrality is selected as the influence entity node; the node with the shortest average path to each street is selected as the accessibility entity node; and the point with the highest comprehensiveness value is selected as the comprehensive entity node using the AHP analytic hierarchy process. The module for adding facilities in inefficient service areas is used to select four types of entity nodes from the service range sub-map of the corresponding public service facilities for inefficient service areas of existing facilities using the same method. It determines the spatial relationship between the entity node and the existing public service facilities. If the entity node is located in the same block as the existing public service facilities or the distance is less than the set value, the entity node is deleted. The entity node in the second priority is selected and judged again until the spatial location relationship condition is met. The entity node that meets the condition is used as the site selection point for the new facility in the inefficient service area of ​​existing facilities. The interactive display module is used to perform three-dimensional spatial mapping using the constructed target city map and the geographic coordinates of its nodes, and to form a three-dimensional sand table by combining it with the three-dimensional model of the target city; the four types of facility site selection points and related calculation results are interactively displayed in different colors; users can determine the final site selection of public service facilities through interactive selection and operation.

7. A computer system comprising a memory, a processor, and computer programs / instructions stored in the memory and executable on the processor, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the method according to any one of claims 1-4.

8. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1-4.

9. A computer program product comprising a computer program / instructions, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the method according to any one of claims 1-4.

Images