A multi-granularity geographic entity matching method, device, equipment and medium

Abstract

The present application discloses a multi-granularity geographic entity matching method, device, equipment and medium, and relates to the field of geographic information technology. The method comprises: obtaining data to be matched in a study area; the data to be matched comprises geographic entity data at different granularities, extracting features from each geographic entity data to obtain geometric information and attribute information, and then calculating geometric similarity and semantic similarity between geographic entity data based on the geometric information and attribute information, performing geographic entity matching based on the geometric similarity and semantic similarity to obtain matching relationships between geographic entities, and constructing a geographic entity association knowledge graph based on the matching relationships between geographic entities. The present application integrates geometric and semantic information to perform multi-granularity geographic entity matching, thereby achieving accurate matching of multi-granularity geographic entities, and is applicable to multi-granularity geographic entity matching, thereby improving adaptability.

Description

Technical Field

[0001] The present application relates to the field of geographic information technology, and in particular to a multi-granularity geographic entity matching method, device, equipment and medium. Background Art

[0002] In geographic information systems, accurate matching of geographic entities (such as farmland, buildings, roads, etc.) is the key to achieving spatial data integration, geospatial analysis, etc. However, due to the multi-scale characteristics of geographic entities, the diversity of data sources, and the difference in data quality, geographic entity matching has become a complex and challenging problem.

[0003] Liu Lingjia (2018) proposed a multi-scale face entity matching method based on the minimum bounding rectangle (MBR) combined optimization algorithm to solve the problem that the position deviation of entities with the same name in multi-scale matching is large and candidate matches cannot be directly obtained through the area overlap method. Zhao Yunpeng (2020) proposed a semantic similarity measurement method for geographic entities, which can reasonably calculate the semantic similarity of geographic entities and improve the efficiency of semantic matching of geographic entities. CN110807797B proposes a matching method for multi-source heterogeneous face entities and point entities that takes into account global optimization. It can be used to automatically match multi-source heterogeneous face and point geographic entities, avoid manual matching, and save workload.

[0004] The above methods only consider one of the factors, geometry or semantics, so the matching accuracy is low; the geometric matching research proposed by the above methods usually focuses on the single case of point-to-plane or face-to-face, and lacks the matching of entities of different granularities, such as the matching of point and volume data, and the matching of point and plane data, which still has a lot of room for improvement. Therefore, a multi-granularity geographic entity matching method that can improve matching accuracy and adaptability is urgently needed. Summary of the invention

[0005] The purpose of this application is to provide a multi-granularity geographic entity matching method, device, equipment and medium, which can integrate geometric and semantic information to perform multi-granularity geographic entity matching, achieve accurate matching of multi-granularity geographic entities, and improve adaptability.

[0006] To achieve the above objectives, this application provides the following solutions.

[0007] In a first aspect, the present application provides a multi-granularity geographic entity matching method, comprising the following steps.

[0008] Obtain the data to be matched in the study area; the data to be matched includes geographic entity data at different granularities; the geographic entity data is two-dimensional expression data or three-dimensional expression data; the granularity includes point expression form, line expression form, surface expression form and volume expression form.

[0009] Feature extraction is performed on each geographic entity data to obtain entity features corresponding to each geographic entity data; the entity features include geometric information and attribute information; the geometric information is the granularity of the geographic entity data.

[0010] The geometric similarity and semantic similarity between geographic entity data are calculated based on the entity features corresponding to the geographic entity data.

[0011] Geographic entity matching is performed based on geometric similarity and semantic similarity to obtain matching relationships between geographic entities.

[0012] Based on the matching relationship between geographic entities, a geographic entity association knowledge graph is constructed.

[0013] In a second aspect, the present application provides a multi-granularity geographic entity matching device, comprising the following modules.

[0014] The module for acquiring data to be matched is used to: acquire the data to be matched in the study area; the data to be matched includes geographic entity data at different granularities; the geographic entity data is two-dimensional expression data or three-dimensional expression data; the granularity includes point expression form, line expression form, surface expression form and volume expression form.

[0015] The feature extraction module is used to: extract features from each geographic entity data to obtain entity features corresponding to each geographic entity data; the entity features include geometric information and attribute information; the geometric information is the granularity of the geographic entity data.

[0016] The similarity calculation module is used to calculate the geometric similarity and semantic similarity between geographic entity data according to the entity features corresponding to the geographic entity data.

[0017] The geographic entity matching module is used to match geographic entities based on geometric similarity and semantic similarity to obtain matching relationships between geographic entities.

[0018] The knowledge graph construction module is used to build a geographic entity association knowledge graph based on the matching relationship between geographic entities.

[0019] In a third aspect, the present application provides a computer device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the above-mentioned multi-granularity geographic entity matching method.

[0020] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which implements the above-mentioned multi-granularity geographic entity matching method when executed by a processor.

[0021] According to the specific embodiments provided by the present application, the present application discloses the following technical effects: the present application provides a multi-granularity geographic entity matching method, apparatus, equipment and medium, which extracts features from each geographic entity data to obtain geometric information and attribute information, and then calculates the geometric similarity and semantic similarity between the geographic entity data based on the geometric information and attribute information, performs geographic entity matching based on the geometric similarity and semantic similarity, and obtains the matching relationship between the geographic entities. This solves the problem that the existing matching methods usually target single situations of point-to-plane or face-to-face, lack matching of entities of different granularities, and only consider one of the geometric or semantic factors, resulting in a low matching accuracy. The geometric and semantic information are integrated to perform multi-granularity geographic entity matching, thereby achieving accurate matching of multi-granularity geographic entities and improving adaptability. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0023] Figure 1 This is an application environment diagram of a multi-granularity geographic entity matching method in an embodiment of the present application;

[0024] Figure 2 A schematic diagram of a flow chart of a multi-granularity geographic entity matching method provided in one embodiment of the present application;

[0025] Figure 3 A schematic diagram of a specific implementation process of a multi-granularity geographic entity matching method provided in one embodiment of the present application;

[0026] Figure 4 A schematic diagram of a geographic entity matching model construction process provided in an embodiment of the present application;

[0027] Figure 5 A schematic diagram of a geographic entity association knowledge graph provided for an embodiment of the present application;

[0028] FIG6 (a) is a schematic diagram of a classification system A provided in an embodiment of the present application; FIG6 (b) is a schematic diagram of a classification system B provided in an embodiment of the present application;

[0029] Figure 7 A schematic diagram of functional modules of a multi-granularity geographic entity matching device provided in one embodiment of the present application;

[0030] Figure 8 A schematic diagram of the structure of a computer device provided in one embodiment of the present application.

[0031] Description of symbols: terminal—102, server—104. DETAILED DESCRIPTION

[0032] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0033] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0034] The multi-granularity geographic entity matching method provided in the embodiment of the present application can be applied to Figure 1 In the application environment shown. Among them, the terminal 102 communicates with the server 104 through the network. The data storage system can store the data that the server 104 needs to process. The data storage system can be set up separately, integrated on the server 104, or placed on the cloud or other servers. The terminal 102 can send the data to be matched to the server 104. After the server 104 receives the data to be matched, for the data to be matched, the server 104 extracts features of each geographic entity data in the matching data to obtain geometric information and attribute information, and then calculates the geometric similarity and semantic similarity between the geographic entity data based on the geometric similarity and the attribute information. The geographic entity matching is performed according to the geometric similarity and the semantic similarity to obtain the matching relationship between the geographic entities. Based on the matching relationship between the geographic entities, the geographic entity association knowledge graph is constructed. The server 104 can feedback the obtained geographic entity association knowledge graph to the terminal 102. In addition, in some embodiments, the multi-granularity geographic entity matching method can also be implemented independently by the server 104 or the terminal 102. For example, the terminal 102 can directly perform multi-granularity geographic entity matching on the data to be matched, or the server 104 can obtain the data to be matched from the data storage system and perform multi-granularity geographic entity matching on the data to be matched.

[0035] The terminal 102 may be, but is not limited to, various desktop computers, laptop computers, smart phones, and tablet computers. The server 104 may be implemented as an independent server or a server cluster consisting of multiple servers, or may be a cloud server.

[0036] In an exemplary embodiment, Figure 2 and Figure 3As shown, a multi-granularity geographic entity matching method is provided. The method is executed by a computer device, and can be executed by a computer device such as a terminal or a server alone, or by a terminal and a server together. In the embodiment of the present application, the method is applied to Figure 1 The server 104 in the example is used as an example to illustrate, including the following steps 201 to 205. Among them:

[0037] Step 201: Obtain the data to be matched in the study area; the data to be matched includes geographic entity data at different granularities; the geographic entity data is two-dimensional expression data or three-dimensional expression data; the granularity includes point expression form, line expression form, surface expression form and volume expression form.

[0038] Step 202: Extract features from each geographic entity data to obtain entity features corresponding to each geographic entity data; the entity features include geometric information and attribute information; the geometric information is the granularity of the geographic entity data.

[0039] Step 203: Calculate the geometric similarity and semantic similarity between the geographic entity data according to the entity features corresponding to the geographic entity data.

[0040] Step 204: Perform geographic entity matching based on geometric similarity and semantic similarity to obtain matching relationships between geographic entities.

[0041] Step 205: Based on the matching relationships between geographic entities, construct a geographic entity association knowledge graph.

[0042] By implementing the above-mentioned steps 201 to 205, geometric information and attribute information are obtained by extracting features from each geographic entity data, and then the geometric similarity and semantic similarity between the geographic entity data are calculated based on the geometric information and attribute information. The geographic entities are matched according to the geometric similarity and semantic similarity to obtain the matching relationship between the geographic entities. This solves the problem that the existing matching methods usually target single situations of point-to-plane or face-to-face, lack matching of entities of different granularities, and only consider one of the geometric or semantic factors, resulting in low matching accuracy. The geometric and semantic information are integrated to perform multi-granularity geographic entity matching, thereby achieving accurate matching of multi-granularity geographic entities and supporting matching of multi-granularity geographic entity data to improve adaptability.

[0043] In step 201, the acquisition of the study area data should cover multiple departments and multiple scales, including geographic entity data in the form of points, lines, surfaces, and volumes. In addition, the data in the study area should contain as much semantic information as possible to improve the matching accuracy of subsequent data.

[0044] Taking two-dimensional data as an example, the acquired data is mainly in Shapefile data format, that is, GIS data format stored in a file-based manner, including .shp\.dbf\.prj and other files, as shown in Table 1.

[0045] Table 1 Shapefile data format.

[0046]

[0047] In another exemplary embodiment of the present application, in order to better perform analysis and application, before the above step 202, the method may also include: preprocessing and structuring the geographic entity data; the preprocessing includes cleaning, deduplication and format conversion, etc., to clean, sort and organize the geographic information data.

[0048] A surface entity is actually connected by coordinate points. The X and Y coordinates of these coordinate points are stored in the .shp file mentioned in Table 1. For subsequent use, the information in the .shp file should be structured and converted into usable formats such as text or tables for easy display and storage. For format errors or garbled characters, data processing is required. For example, the coordinates of a surface entity in a txt file are shown in Table 2. The X and Y coordinates of each point are separated by a space, and two points are separated by a comma. Files in other data formats in Table 1 are processed in this way.

[0049] Table 2 Coordinate points of surface entities.

[0050]

[0051] In step 202, representative features are extracted from the geographic entity data, mainly including geometric information and attribute information, including but not limited to name, area, etc. These features will be used for subsequent matching calculations. Table 1 mentions that the attribute information of the geographic entity is stored in the .dbf file, so the .dbf file needs to be converted into a table format such as CSV format and then preprocessed. However, the content in the table is very rich, so the required information needs to be extracted. The extracted information is shown in Table 3. The extracted information includes land ownership certificate, land location, etc.

[0052] Table 3 Attribute information.

[0053]

[0054] In another exemplary embodiment of the present application, the above step 203 is replaced by the following steps 301 to 302.

[0055] Step 301: Calculate semantic similarity between geographic entity data according to attribute information of the geographic entity data; semantic similarity includes attribute similarity and category similarity.

[0056] Step 302: Based on the granularity of the geographic entity data and the geometric information of the geographic entity data, the geometric similarity between the geographic entity data is calculated; the geometric similarity includes spatial distance similarity, spatial topological relationship similarity and shape index similarity.

[0057] The calculation of semantic similarity includes attribute similarity and category similarity. Attribute similarity is calculated by the most commonly used attribute names, that is, by comparing the name attributes of each geographic entity in the two layers. Attribute similarity includes name similarity, encoding similarity and other attribute similarities. The calculation formula of name similarity is shown in the following formula (1).

[0058] (1).

[0059] in, N Indicates name similarity; Indicates i The similarity of a role label word in the name of a geographic entity. For example, the name text of a geographic entity is Jinjia Village No. 1 Courtyard South District Building No. 9, and its role labels are: Jinjia Village: proper name; No. 1 Courtyard: number word; South District: location word; No. 9 Building: number word; m Indicates the number of unions of two name text role labels.

[0060] In addition to name similarity, if the geographic entity data contains coding and other attribute information, other attribute similarities can also be calculated to increase matching accuracy.

[0061] When calculating the coding similarity, the coding of the geographic entity data is matched bit by bit, and the coding similarity is calculated based on the number of bits of the matching coding.

[0062] Calculation of category similarity: In order to more accurately match geographic entity categories from different sources, this application proposes a bottom-up category mapping method to enhance category constraints and thus improve matching accuracy. This is achieved by establishing a semantic mapping relationship between category nodes between two classification systems. For subcategories, by tracing the hierarchical structure tree, the mapping relationship of its parent category is assigned. For example, the third-level category of the reservoir entity has a corresponding second-level category of the reservoir entity, and the first-level category is the water conservancy and facilities category. The parent category entity of the reservoir entity is the reservoir entity, and the water conservancy and facilities category is the parent category of the reservoir entity. This application defines three mapping relationships: one is that the semantics of the category node are completely mapped, and the semantic distance is 0; the second is that the category node finds the mapping relationship through its parent node, and the semantic distance is divided into 1, 2, and 3; the third is that there is no mutual mapping relationship at all, and its semantic distance is +∞.

[0063] As shown in Figure 6 (a), in classification system A, the first-level category is entertainment, the second-level category includes restaurants, bars and cinemas, and the third-level category includes Chinese restaurants and Western restaurants; among them, restaurants are the parent node of (Chinese restaurants and Western restaurants). As shown in Figure 6 (b), in classification system B, the first-level category is entertainment and leisure, the second-level category includes catering and bars, and the third-level category includes Chinese restaurants, Western restaurants and cafes, among which catering is the parent node of (Chinese restaurants, Western restaurants and cafes).

[0064] In classification system A and classification system B, the bar in classification system A and the bar in classification system B represent exactly the same categories in the two classification systems, so their semantic distance is 0; the Chinese restaurant in classification system A and the Chinese restaurant in classification system B, the parent nodes of these two nodes are classification system A (restaurant) and classification system B (catering), respectively, and they both represent the "restaurant" category, so the semantic distances are 1, 2 and 3; similarly, the cinema in classification system A and the classification system B do not have any corresponding nodes, that is, there is a node in classification system A that cannot find any relevant mapping in classification system B (that is, there are no categories with the same or similar semantics), then their semantic distance is +∞.

[0065] The calculation of geometric similarity for multi-granularity geographic entity data includes six cases: point-to-point, point-to-plane, point-to-body, face-to-face, face-to-body, and body-to-body. Among them, line entities are considered to be a special type of surface.

[0066] Geometric similarity indicators include: spatial distance similarity, spatial topological relationship similarity and shape index similarity. The spatial distance indicator calculates the actual distance between two geographic entities and evaluates the degree of proximity between them. The spatial topological indicator uses spatial topological relationships to further verify the matching, such as topological relationships such as inclusion, intersection and connectivity. The shape indicator compares the shape features of two geographic entities, such as area, perimeter and compactness.

[0067] 1) Point entity and point entity: The geometric similarity between point entities mainly calculates the Euclidean distance between two points. Taking the WGS_1984 coordinate system as an example, the calculation formula is shown in the following formula (2).

[0068] (2).

[0069] In the formula, d is the plane distance between two points in the WGS_1984 coordinate system, ( , )and( , ) are the coordinates of the corresponding points of the first geographic entity and the second geographic entity respectively.

[0070] Set the distance threshold. If the distance between two point entities is calculated by formula (2) d If the calculated distance is less than the set distance threshold, the two point entities are considered to be the same geographic entity. d If the distance between the two points is not less than the set distance threshold, the two point entities are considered not to be the same geographic entity.

[0071] 2) The matching method between point entity and surface entity is as follows: 2D surface entity vector data is essentially n coordinate points, and the boundary is formed by connecting each coordinate point to form a surface entity. Therefore, the surface entity can be expressed as { ， ，... ... }, For the surface entity coordinate points, is the number of points on the surface entity.

[0072] The matching of point entities and surface entities requires judging the topological relationship between the points and the surfaces, which is achieved through coordinate calculation. Therefore, a unified coordinate system is required. The specific implementation method is to judge whether the point is within the circumscribed rectangle of the surface.

[0073] Taking the WGS_1984 coordinate system as an example, the calculation formula for the circumscribed rectangle is shown as follows.

[0074] (3).

[0075] (4).

[0076] (5).

[0077] (6).

[0078] (7).

[0079] In formulas (3) to (6), , are the maximum and minimum values ​​of the horizontal coordinates of each coordinate point of the surface entity, , are the maximum and minimum values ​​of the ordinate of each coordinate point of the surface entity. In formula (7), X and Y are the horizontal and vertical coordinates of the point entity, respectively. X , YIf they are respectively between the minimum and maximum values ​​of the horizontal and vertical coordinates of all coordinate points of the surface entity, it means that the point entity is within the circumscribed rectangle of the surface entity, that is, it is regarded as the same geographic entity; if not, the point entity and the surface entity are regarded as not being a single geographic entity.

[0080] If the point is within the circumscribed rectangle, the ray method is used to continue the judgment. The main implementation method of the ray method is to use the point to be matched as the origin and draw a ray horizontally to the left or right. If the intersection point with the surface is an odd number, the point is within the surface; if the intersection point is an even number, the point is outside the surface.

[0081] 3) The method for determining point entities and volume entities is to map the 3D data to a plane through orthographic projection and then match the point entity with the surface entity. Suppose the coordinates of the top of the original 3D model are [ xyz ], the projected coordinates are [ ]. The orthographic projection calculation formula (orthographic projection transformation matrix) is shown in the following formula (8).

[0082] (8).

[0083] but .

[0084] 4) The judgment between surface entities is carried out by overlay analysis method or geometric center method. Since the area and outline of geographical entities may change greatly at different stages, overlay analysis method and geometric center method are used as needed.

[0085] Overlay analysis method: Overlay analysis can use the overlay analysis function of ArcMap to generate the intersection of two layers, and then calculate the ratio of the area of ​​the intersection to the larger area of ​​the two original layers, set a threshold, and if the threshold is reached, it is considered to be the same geographic entity. If it is not reached, it is not considered to be the same geographic entity. The calculation formulas are shown in equations (9) and (10).

[0086] (9).

[0087] (10).

[0088] In the formula, , is the area of ​​the two original layers, S is the area of ​​the intersection, and in formula (9) P And in formula (10) P The areas of the intersections are and The proportion of the two original layers.

[0089] The geometric center method is to calculate the geometric center points of all surface entities in one layer, and then regard the geometric center points as point entities. The matching method of point entities and surface entities can be used for judgment. This step also requires a unified scale and coordinate system. The calculation formula of the geometric center point is shown in formula (11).

[0090] (11).

[0091] In the formula, , To constitute the first coordinate points; is the number of points on the surface entity; ( , ) is the geometric center point.

[0092] For example, the calculation result of the geometric center point of the data in Table 1 is the sum of all X and Y coordinates divided by the number. Then the geometric center point is used for topological matching.

[0093] 5) The matching method between the surface entity and the volume entity is as follows: the three-dimensional data of the volume entity is mapped to a plane through orthographic projection to obtain the surface entity corresponding to the volume entity, and then the overlay analysis method or the geometric center method is used to match the surface entity with the surface entity.

[0094] 6) The matching method between volume entities is to map the three-dimensional data of the volume entity to a plane through orthographic projection to obtain the surface entities corresponding to the two volume entities, and then use the overlay analysis method or the geometric center method to match the surface entities obtained by the orthographic projection mapping, while considering the height similarity.

[0095] Step 204 specifically includes: using a weight distribution method and an iterative optimization method to perform geographic entity matching according to geometric similarity and semantic similarity to obtain matching relationships between geographic entities.

[0096] Specifically, the weight distribution method and iterative optimization method are used to construct the geographic entity matching model, and the geometric and semantic similarities are integrated to perform geographic entity matching at multiple granularities. The establishment of the geographic entity matching model is divided into three levels.

[0097] like Figure 4 As shown in the figure, the first layer is attribute similarity judgment. If the geographic entity data value only contains the name attribute, it is matched only by name similarity; if the data also contains coding and other attribute information, multi-attribute fusion judgment is required to improve the accuracy. According to the importance of different attributes, each attribute needs to be assigned a weight. The initial weight is assigned according to the accuracy of the individual matching of various attributes, and then the weight is adjusted through multiple iterations to calculate the accuracy. When the calculated accuracy reaches the preset threshold, the iteration stops.

[0098] The second layer is matching based on category information. This step mainly verifies the matching results of the first step based on the category information in the semantics and supplements it.

[0099] The third layer matches according to geometric similarity, including six cases. Point-to-point matching requires setting a distance threshold, and face-to-face matching requires scaling small-scale data to the same scale as large-scale data, and then performing topological matching. By continuously adjusting each threshold and iterating repeatedly, a geographic entity matching model is obtained when the accuracy reaches the highest. The six cases of geometric similarity (point-to-point, point-to-face, point-to-body, face-to-face, face-to-body, and body-to-body) are eventually converted into point-to-point and face-to-face cases.

[0100] The final output results are divided into four cases: 1) Both geometric and semantic similarities match, which means that the name and location of the land have not changed much; 2) The semantic similarity matches but the geometric similarity does not match, which means that the area or location of the land has changed significantly; 3) The geometric similarity matches but the semantic similarity does not match, which means that the nature of the land has changed; 4) Both the geometric similarity and the semantic similarity do not match, which means it is not a piece of land.

[0101] In another exemplary embodiment of the present application, the above step 205 is replaced by the following steps 401 to 402.

[0102] Step 401: assign a code to each geographic entity based on the matching relationship between the geographic entities to obtain a triplet corresponding to each geographic entity.

[0103] Step 402: Construct a geographic entity associated knowledge graph based on the triples corresponding to all geographic entities.

[0104] Specifically, each geographic entity is assigned a unique code based on the matching relationship, and a geographic entity-related knowledge graph is constructed based on the triple structure through the unique code. The knowledge graph construction process includes relationship extraction, entity linking, graph construction, graph evaluation and optimization, among which relationship extraction mainly identifies the matching relationship of geographic entities. After the matching is completed, the information in Table 2 and Table 3 is uniformly imported into the semantic library. A certain geographic entity data eventually forms a geographic entity-related knowledge graph such as Figure 5 shown.

[0105] This application involves classification in geographic information systems (GIS), as well as cross-scale and cross-data source geographic entity matching and fusion technology. This application can comprehensively consider the multiple characteristics of geographic entities and the differences between different data sources (fusion of geometric and semantic information for geographic entity matching), achieve accurate matching of multi-granular geographic entities, and improve the accuracy and completeness of matching; flexibly handle the differences and multi-scale characteristics between different data sources, and achieve cross-data source and cross-scale geographic entity matching; propose a matching method suitable for expressing geographic entity data in all expression forms of points, lines, surfaces, and volumes, which increases applicability.

[0106] The present application also provides an application scenario, which applies the above-mentioned multi-granularity geographic entity matching method. Specifically: the multi-granularity geographic entity matching method provided in this embodiment can be applied in the construction scenario of geographic entity associated knowledge graph. The construction scenario of geographic entity associated knowledge graph includes a data acquisition link and a multi-granularity geographic entity matching link; the data to be matched enters the multi-granularity geographic entity matching link from the data acquisition link, and feature extraction is performed on each geographic entity data in the matching data to obtain geometric information and attribute information, and then based on the geometric information and attribute information, the geometric similarity and semantic similarity between the geographic entity data are calculated, and geographic entity matching is performed based on the geometric similarity and semantic similarity to obtain the matching relationship between the geographic entities, and the geographic entity associated knowledge graph is constructed based on the matching relationship between the geographic entities. The multi-granularity geographic entity matching method provided in this embodiment belongs to a multi-granularity geographic entity matching link. Specifically, in the multi-granularity geographic entity matching link process for the data to be matched, multi-granularity geographic entity matching can be performed on the matching data to obtain the matching relationship between the geographic entities.

[0107] Based on the same inventive concept, the embodiment of the present application also provides a multi-granularity geographic entity matching device for implementing the multi-granularity geographic entity matching method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more multi-granularity geographic entity matching device embodiments provided below can refer to the limitations of the multi-granularity geographic entity matching method above, and will not be repeated here.

[0108] In an exemplary embodiment, Figure 7 As shown, a multi-granularity geographic entity matching device is provided, including the following modules.

[0109] The module T1 for acquiring data to be matched is used to: acquire data to be matched in the study area; the data to be matched includes geographic entity data at different granularities; the geographic entity data is two-dimensional expression data or three-dimensional expression data; the granularity includes point expression form, line expression form, surface expression form and volume expression form.

[0110] The feature extraction module T2 is used to: extract features from each geographic entity data to obtain entity features corresponding to each geographic entity data; the entity features include geometric information and attribute information; the geometric information is the granularity of the geographic entity data.

[0111] The similarity calculation module T3 is used to calculate the geometric similarity and semantic similarity between geographic entity data according to the entity features corresponding to the geographic entity data.

[0112] The geographic entity matching module T4 is used to perform geographic entity matching based on geometric similarity and semantic similarity to obtain matching relationships between geographic entities.

[0113] The knowledge graph construction module T5 is used to: construct a geographic entity associated knowledge graph based on the matching relationship between geographic entities.

[0114] The multi-granularity geographic entity matching device also includes a preprocessing module; the preprocessing module is used to: preprocess the geographic entity data; the preprocessing includes cleaning, deduplication and format conversion.

[0115] As an optional implementation, the knowledge graph construction module T5 includes the following units.

[0116] The triplet determination unit is used to assign a code to each geographical entity based on the matching relationship between the geographical entities to obtain a triplet corresponding to each geographical entity.

[0117] The knowledge graph construction unit is used to: construct a geographic entity associated knowledge graph based on the triples corresponding to all geographic entities.

[0118] In an exemplary embodiment, a computer device is provided. The computer device may be a server or a terminal. The internal structure diagram thereof may be as follows: Figure 8As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store geographic entity matching data. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a multi-granularity geographic entity matching method is implemented.

[0119] Those skilled in the art will understand that Figure 8 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0120] In an exemplary embodiment, a computer device is further provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above-mentioned method embodiments when executing the computer program.

[0121] In an exemplary embodiment, a computer-readable storage medium is provided, storing a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0122] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0123] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM may be in various forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).

[0124] The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. The non-relational database may include a distributed database based on blockchain, etc., but is not limited thereto. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but is not limited thereto.

[0125] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0126] This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method and core ideas of this application. At the same time, for those skilled in the art, according to the ideas of this application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A multi-granularity geographic entity matching method, characterized in that: The multi-granularity geographic entity matching method comprises: Acquire the data to be matched in the study area; the data to be matched includes geographic entity data at different granularities; the geographic entity data is two-dimensional expression data or three-dimensional expression data; the granularity includes point expression form, line expression form, surface expression form and volume expression form; Extracting features from each geographic entity data to obtain entity features corresponding to each geographic entity data; the entity features include geometric information and attribute information; the geometric information is the granularity of the geographic entity data; Calculate the geometric similarity and semantic similarity between geographic entity data according to the entity features corresponding to the geographic entity data; Geographic entity matching is performed based on geometric similarity and semantic similarity to obtain matching relationships between geographic entities, specifically including: using weight allocation method and iterative optimization method to perform geographic entity matching based on geometric similarity and semantic similarity to obtain matching relationships between geographic entities; using weight allocation method and iterative optimization method to build a geographic entity matching model, integrating geometric and semantic similarity to perform geographic entity matching at multiple granularities; the establishment of the geographic entity matching model is divided into three levels, including: the first level is attribute similarity judgment, according to the importance of different attributes, initial weights are assigned to names, codes and other attributes respectively, and then the matching accuracy is calculated. When the matching accuracy does not meet the requirements, the weights are adjusted through multiple iterations. When the calculated accuracy reaches the preset threshold, the iteration is stopped. The second layer is matched according to the category information, and the matching results of the first layer are verified and supplemented according to the category information in the semantics; when the category matches, the third layer is entered; the third layer matches according to the geometric similarity, and the geometric similarity includes six cases: point-to-point, point-to-face, point-to-body, face-to-face, face-to-body, and body-to-body. When performing geometric matching calculations, they are all converted into point-to-point and point-to-face cases for matching calculations; the matching of points requires setting a distance threshold. When the distance between two point entities is less than the set distance threshold, they are the same geographic entity; by continuously adjusting the distance threshold and iterating repeatedly, when the matching accuracy reaches the highest, a geographic entity matching model is obtained; the geographic entity matching model is used to perform geographic entity matching at multiple granularities; Based on the matching relationship between geographic entities, a geographic entity association knowledge graph is constructed.

2. The multi-granularity geographic entity matching method according to claim 1, characterized in that: Before performing the feature extraction step on the geographic entity data, the multi-granularity geographic entity matching method further includes: The geographic entity data is preprocessed; the preprocessing includes cleaning, deduplication and format conversion.

3. The multi-granularity geographic entity matching method according to claim 1, characterized in that: The geometric similarity and semantic similarity between geographic entity data are calculated based on the entity features corresponding to the geographic entity data, including: Calculate the semantic similarity between geographic entity data based on the attribute information of geographic entity data; semantic similarity includes attribute similarity and category similarity; Based on the granularity of geographic entity data, the geometric similarity between geographic entity data is calculated according to the geometric information of the geographic entity data; the geometric similarity includes spatial distance similarity, spatial topological relationship similarity and shape index similarity.

4. The multi-granularity geographic entity matching method according to claim 1, characterized in that: The construction of a geographic entity associated knowledge graph based on the matching relationship between geographic entities specifically includes: Based on the matching relationship between geographic entities, each geographic entity is assigned a code to obtain a triple corresponding to each geographic entity; Construct a geographic entity association knowledge graph based on the triples corresponding to all geographic entities.

5. A multi-granularity geographic entity matching device, characterized in that: The multi-granularity geographic entity matching device comprises: The module for acquiring data to be matched is used to: acquire the data to be matched in the research area; the data to be matched includes geographic entity data at different granularities; the geographic entity data is two-dimensional expression data or three-dimensional expression data; the granularity includes point expression form, line expression form, surface expression form and volume expression form; The feature extraction module is used to: extract features from each geographic entity data to obtain entity features corresponding to each geographic entity data; the entity features include geometric information and attribute information; the geometric information is the granularity of the geographic entity data; The similarity calculation module is used to calculate the geometric similarity and semantic similarity between geographic entity data according to the entity features corresponding to the geographic entity data; The geographic entity matching module is used to: match geographic entities according to geometric similarity and semantic similarity to obtain matching relationships between geographic entities, specifically including: using weight allocation method and iterative optimization method to match geographic entities according to geometric similarity and semantic similarity to obtain matching relationships between geographic entities; using weight allocation method and iterative optimization method to build a geographic entity matching model, integrating geometric and semantic similarity to match geographic entities at multiple granularities; the establishment of the geographic entity matching model is divided into three levels, including: the first level is attribute similarity judgment, according to the importance of different attributes, initial weights are assigned to names, codes and other attributes respectively, and then the matching accuracy is calculated. When the matching accuracy does not meet the requirements, the weights are adjusted through multiple iterations. When the calculated accuracy reaches the preset threshold When the value is reached, the iteration process is stopped; the second layer is matched according to the category information, and the matching results of the first layer are verified and supplemented according to the category information in the semantics; when the category matches, the third layer is entered; the third layer matches according to geometric similarity, and the geometric similarity includes six cases: point-to-point, point-to-face, point-to-body, face-to-face, face-to-body, and body-to-body. When performing geometric matching calculations, they are all converted into point-to-point and point-to-face cases for matching calculations; the matching of points requires setting a distance threshold. When the distance between two point entities is less than the set distance threshold, they are the same geographic entity; by continuously adjusting the distance threshold and iterating repeatedly, when the matching accuracy reaches the highest, a geographic entity matching model is obtained; the geographic entity matching model is used to perform geographic entity matching at multiple granularities; The knowledge graph construction module is used to build a geographic entity association knowledge graph based on the matching relationship between geographic entities.

6. The multi-granularity geographic entity matching device according to claim 5, characterized in that: The multi-granularity geographic entity matching device also includes a pre-processing module; The preprocessing module is used to preprocess the geographic entity data; the preprocessing includes cleaning, deduplication and format conversion.

7. The multi-granularity geographic entity matching device according to claim 5, characterized in that: The knowledge graph construction module includes: A triplet determination unit is used to: assign a code to each geographical entity based on the matching relationship between the geographical entities, and obtain a triplet corresponding to each geographical entity; The knowledge graph construction unit is used to construct a geographic entity associated knowledge graph based on the triples corresponding to all geographic entities.

8. A computer device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the multi-granularity geographic entity matching method according to any one of claims 1 to 4.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the multi-granularity geographic entity matching method according to any one of claims 1 to 4 is implemented.

Images