A method and system for GIS automatic review based on structured natural language rules

Abstract

The present application relates to a kind of GIS automatic review method, system and readable medium based on structured natural language rule, comprising the following steps: parsing the syntax component of the SNL rule to be reviewed, the rule type of each SNL clause in SNL rule is analyzed;According to syntax component and rule type, query is carried out in the geographic database of the GIS model to be reviewed imported, obtains the target entity set and attribute information corresponding to rule type;The target entity set of each rule type is carried out logical operation, obtains the entity set participating in spatial geometric calculation;Through spatial topological constraint condition, the entity set participating in spatial geometric calculation is filtered, and the entity set after filtering is calculated and processed, obtains the review result of each SNL clause;According to the logical relationship of each SNL clause, the review result of each SNL clause is carried out logical operation, obtains the review result of SNL rule.It solves the problem that the building specification review method of city space does not support custom review rule.

Description

Technical Field

[0001] This invention relates to an automatic review method, system, and readable medium for GIS (Geographic Information System) based on structured natural language rules, belonging to the field of computer automatic detection technology, and is particularly suitable for the review of building-related regulations in urban spaces. Background Technology

[0002] With the emergence of new concepts such as "smart cities" and "CIM (City Information Modeling) platforms" both domestically and internationally in recent years, the application of computer-aided building code review for urban spaces has gradually come into focus. Urban space building code review primarily focuses on urban planning review, covering aspects such as building spacing, building density, and building floor area ratio. Currently, research on building code review in such large-scale scenarios is limited. Patent CN114119938A proposes a batch processing method for verifying and assigning urban building data. This method uses configuration files to batch assign values ​​to urban building layers in a GIS model, while simultaneously checking for discrepancies between the building layer data and the requirements in the configuration file. Patent CN114170041A proposes a construction site layout optimization method and system based on BIM+GIS, which optimizes the layout of construction sites through multi-objective function optimization. However, these patents all have weak scalability and limited scope of inspection. Although there are existing building code review methods and systems for BIM models, such as the complex building code compliance detection device combining semantic and geometric calculations proposed in patent CN107341598B, this device is not suitable for building code review in larger scenarios.

[0003] There is also a need for automated code review of city-level building complexes, such as urban building planning reviews. However, current inspection methods for various building models in urban spaces are highly customized, focusing on the review of specific types of building codes within the urban space, such as building spacing or lighting. These methods have limitations: different tools are needed for reviewing building codes in different fields and professions, increasing the learning curve for reviewers and impacting the overall review process and efficiency.

[0004] Furthermore, urban building planning regulations generally adopt local standards, meaning that the requirements for the same indicator may not be consistent across different regions. This implies that the aforementioned customized review methods and devices need to be developed with corresponding inspection methods tailored to the different regional standards, which not only increases the workload of R&D personnel but also affects the overall efficiency of the review process. Summary of the Invention

[0005] To address the aforementioned issues, the present invention aims to provide a GIS automatic review method, system, and readable medium based on structured natural language rules. This invention successfully solves the problems of existing urban space building code review methods, such as lack of support for custom review rules, difficulty in achieving scalable automated review for GIS models, and slow data retrieval and complex spatial calculations due to the large scale of GIS model data.

[0006] To achieve the above objectives, this invention proposes the following technical solution: an automatic GIS review method based on structured natural language rules, comprising the following steps: parsing the syntactic components of the SNL rules to be reviewed, and analyzing the rule type of each SNL clause in the SNL rules; querying the geographic database of the GIS model to be reviewed according to the syntactic components and rule types to obtain the target entity set and attribute information corresponding to the rule type; performing logical operations on the target entity sets of each rule type to obtain the entity set participating in spatial geometric calculations; filtering the entity set participating in spatial geometric calculations through spatial topological constraints, and performing calculation processing on the filtered entity set to obtain the review results of each SNL clause; and performing logical operations on the review results of each SNL clause according to the logical relationship of each SNL clause to obtain the review results of the SNL rules.

[0007] Furthermore, the grammatical components include entity names and constraints; the rule types include concluding clauses and non-rule clauses.

[0008] Furthermore, queries are performed in the geographic database using query statements, which are generated based on syntactic components and rule types.

[0009] Furthermore, the method for generating query statements based on syntactic components and rule types is as follows: For a single semantic query rule, the corresponding WHERE clause constraint is generated through structural mapping; for multiple semantic query rules, the two types of semantic query rules are distinguished by determining whether the SNL clause is in the conclusion clause, and based on the logical relationship between the corresponding SNL clauses of each semantic query rule, multiple constraint conditions of the same entity name in the same type of semantic query rule are connected using logical operators, and the constraint conditions corresponding to the entity name that have undergone logical operations are used as the WHERE clause to generate the query statement of the corresponding entity.

[0010] Furthermore, the method for obtaining the set of entities participating in spatial geometric calculations is as follows: by performing syntactic parsing and semantic analysis on the query statement, a set of semantic query rules and a set of semantic query rules are obtained; the set of semantic query rules is processed to obtain semantically constrained entities in the conclusion clause and semantically constrained entities in the non-conclusion clause; the set of computational rules is classified into computational rules to obtain a set of computational rules in the conclusion clause and a set of computational rules in the non-conclusion clause.

[0011] Furthermore, the semantic constraint entities of the conclusion clause and the set of computational rules of the conclusion clause are separated into the result set of the conclusion clause, the semantic constraint entities of the non-conclusion clause and the set of computational rules of the non-conclusion clause are separated into the result set of computational rules of the non-conclusion clause, and logical operations are performed on the result set of the conclusion clause and the result set of the non-conclusion clause to obtain the result set of computational rules.

[0012] Furthermore, the method for filtering the entity set participating in spatial geometric calculation is as follows: the computational rules in the result set of computational rules are parsed to generate topological relations, constraints, and rule subjects. The rule subjects and the entity set participating in the calculation generate the entity set participating in the spatial query. Based on the topological relations, constraints, and the entity set participating in the spatial query, a spatial query is performed to obtain the entity set that satisfies the topological relations. Logical operations are performed on the entity set that satisfies the topological relations, constraints, and the entity set participating in the calculation to generate the entity set participating in the calculation that satisfies the spatial constraints.

[0013] Furthermore, errors appearing in the review results of SNL rules are extracted, the errors are categorized, the review results are displayed according to different error types, and error information is supplemented according to the error type.

[0014] This invention also discloses an automatic GIS review system based on structured natural language rules (SNL), comprising: an SNL rule parsing module for parsing the syntactic components of the SNL rules to be reviewed and analyzing the rule type of each SNL clause; a query module for querying the geographic database of the GIS model to be reviewed based on the syntactic components and rule type to obtain the target entity set and attribute information corresponding to the rule type; a semantic query and spatial computation combined processing module for performing logical operations on the target entity sets of each rule type to obtain the entity set participating in spatial geometric computation; a spatial computation optimization module based on spatial query for filtering the entity set participating in spatial geometric computation through spatial topological constraints, performing computation processing on the filtered entity set to obtain the review result of each SNL clause; and a result output module for performing logical operations on the review result of each SNL clause according to the logical relationship of each SNL clause to obtain the review result of the SNL rule.

[0015] The present invention also discloses a computer-readable storage medium storing a computer program, which is executed by a processor to implement any of the above-mentioned GIS automatic review methods based on structured natural language rules.

[0016] The present invention has the following advantages due to the adoption of the above technical solutions:

[0017] 1. This invention applies SNL rules to define rules for GIS model inspection, and proposes a query statement generation method based on SNL rules, a two-stage processing method combining semantic query and spatial computation, and a computational rule optimization method based on spatial query.

[0018] 2. This invention solves the problem of lack of scalability in existing urban space building code review tools, making it convenient for users to use custom rules to describe and check building codes.

[0019] 3. This invention solves the problems of high query and calculation complexity and low review efficiency caused by the large scale of urban spatial data through various optimization methods, and reduces the review time of relevant specifications. Attached Figure Description

[0020] Figure 1 This is a schematic diagram of an SQL (Structured Query Language) query statement based on entity name clustering in one embodiment of the present invention;

[0021] Figure 2 This is a flowchart of a method for obtaining a set of entities participating in spatial geometric calculations according to an embodiment of the present invention;

[0022] Figure 3 This is a flowchart of a method for filtering a set of entities participating in spatial geometric calculations according to an embodiment of the present invention;

[0023] Figure 4 This is a schematic diagram of the structure of a GIS model inspection device based on SNL in one embodiment of the present invention. Detailed Implementation

[0024] To enable those skilled in the art to better understand the technical solutions of the present invention, the present invention is described in detail through specific embodiments. However, it should be understood that the specific embodiments are provided only for a better understanding of the present invention and should not be construed as limiting the present invention. In the description of the present invention, it should be understood that the terminology used is for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0025] To address the shortcomings of existing urban spatial building code review methods, such as lack of support for custom review rules, difficulty in achieving scalable automated review for GIS models, and slow data retrieval and complex spatial calculations due to the large scale of GIS model data, this invention discloses a GIS automatic review method, system, and readable medium based on Structured Natural Language (SNL) rules. It generates SQL query statements through SNL rules, performs queries in a geographic database using these SQL statements, and employs a two-stage processing method for SNL rules containing computational rules: first processing semantic query rules, then processing computational rules. The computational rules based on spatial queries optimize the entity set participating in spatial geometric calculations. This invention solves the problem of insufficient scalability in existing urban spatial building code review tools, allowing users to describe and check building codes using custom rules. It addresses the high query and computational complexity and low review efficiency caused by the large scale of urban spatial data through various optimization methods, reducing the review time for relevant codes. The invention is further illustrated below with reference to the accompanying drawings and embodiments.

[0026] Example 1

[0027] This implementation discloses an automatic GIS review method based on Structured Natural Language (SNL) rules. Addressing the issue that existing urban spatial building code review methods and systems do not support custom review rules, this embodiment applies SNL-defined inspection rules. SNL-defined inspection rules are engineering domain-oriented rules, describing domain-specific language. They define a clear set of rule description syntax, enabling structured descriptions of relevant engineering domain specifications. The structured description of specifications makes the logic clearer, easier for computers to read and understand, and also facilitates conversion to other query languages ​​(such as SPARQL and SQL). The basic syntax structure of SNL is shown in Table 1.

[0028] The GIS automatic review method in this embodiment includes the following steps:

[0029] S1 users input the GIS model to be reviewed and SNL rules into the device, and import the GIS model data into the geographic database for storage and management.

[0030] S2 uses the SNL parsing framework to parse the syntactic components of the SNL rules to be examined, analyzes the rule type of each SNL clause in the SNL rules, and adjusts the constraints of each SNL clause based on semantic information.

[0031] Syntactic components include entity names and constraints; rule types include concluding clauses and non-rule clauses.

[0032] S3 queries the geographic database of the imported GIS model based on the grammatical components and rule types to obtain the target entity set and attribute information corresponding to the rule types.

[0033] Queries are performed in the geographic database using query statements, which are generated based on syntactic components and rule types. This embodiment uses an SQL query statement as an example, but it is not limited to this; other query statements can also be used.

[0034] The method for generating query statements based on syntactic components and rule types is as follows:

[0035] For a single semantic query rule, each syntactic component in the SNL clause is mapped into the SQL SELECT query statement through structural mapping. The specific mapping relationship is shown in Table 2, generating the corresponding constraints of the WHERE clause.

[0036] Two types of semantic query rules are distinguished by whether they are in the conclusion clause, and multiple constraints with the same entity name in the same type of semantic query rule are connected by logical operators (OR, AND) according to the logical relationship between the corresponding SNL clauses of each semantic query rule.

[0037] The SQL query statement for the corresponding entity is generated by using the logically operated constraints corresponding to the entity name as the WHERE clause and the name part of the entity name as the table name in the SELECT statement. When there are multiple semantic query rules for the same entity in the SNL rules, this SQL generation method based on entity name clustering can reduce the number of SQL queries for the same table, and also reduce the time and performance overhead of finding the intersection or union of multiple result sets. In other words, this method can improve query efficiency. Figure 1 As shown in the figure, this diagram illustrates the generated SQL statements and mapping relationships for the SNL statements in the diagram. SQL queries are then used in the geodatabase to retrieve the target entity set and corresponding attribute information for semantic query rules in both conclusion and non-conclusion clauses.

[0038] Table 1. Basic Syntax Structure of SNL

[0039]

[0040] S4 distinguishes between two types of computational rules based on whether they are in the conclusion clause, and performs logical operations on the target entity sets of each rule type according to their logical relationship with the semantic query rules to obtain the entity sets participating in spatial geometric calculations.

[0041] like Figure 2As shown, the method for obtaining the set of entities participating in spatial geometric calculations is as follows: A set of semantic query rules and a set of non-conclusion clause semantic constraint entities are obtained by parsing and semantically analyzing the query statement. The set of semantic query rules is then processed to obtain the set of computational rules for the conclusion clause and the set of non-conclusion clause semantic constraint entities. The set of computational rules is then categorized to obtain the set of computational rules for the conclusion clause and the set of computational rules for the non-conclusion clause. The set of semantic constraint entities for the conclusion clause and the set of computational rules for the conclusion clause are then grouped into the result set of the conclusion clause, while the set of semantic constraint entities for the non-conclusion clause and the set of computational rules for the non-conclusion clause are grouped into the result set of computational rules for the non-conclusion clause. Logical operations are then performed on the result set of the conclusion clause and the set of computational rules for the non-conclusion clause to obtain the result set of computational rules. This processing flow avoids the complexity of processing logic caused by repeatedly switching between semantic information query and spatial calculation processes, thus optimizing the processing flow and improving processing efficiency.

[0042] Table 2. Correspondence between SNL and SQL

[0043]

[0044] S5 filters the set of entities participating in spatial geometric calculations through spatial topological constraints. Using a defined general geometric calculation method, it calculates and processes various relative positional relationships and corresponding attributes (such as building spacing) of the filtered set of entities to obtain the review results of each SNL clause. The spatial topological constraints are obtained through parsing of the SNL syntax parsing framework.

[0045] like Figure 3 As shown, the method for filtering the entity set participating in spatial geometric calculations is as follows: The computational rules in the result set of computational rules are parsed to generate topological relationships, constraints, and rule subjects. The rule subjects and the set of entities participating in the calculation generate the entity set participating in the spatial query. A spatial query is performed based on the topological relationships, constraints, and the entity set participating in the spatial query to obtain the entity set satisfying the topological relationships. Logical operations are then performed on the entity set satisfying the topological relationships, constraints, and the entity set participating in the calculation to generate the entity set satisfying the spatial constraints. This processing flow uses spatial queries to filter out the entity set participating in the calculation, reducing the time and performance consumption caused by a large amount of spatial computation. This module obtains the semantic constraints of the computational rules through SNL syntax parsing and then filters out the entity set satisfying these semantic constraints through spatial queries. Due to the existence of spatial indexes in the geographic database, the above method can effectively reduce the number of entities participating in spatial geometric calculations with less time and computational resource consumption, thereby optimizing the processing efficiency of computational rules.

[0046] S6 performs logical operations on the review results of each SNL clause based on the logical relationship between the clauses, and the resulting set of calculation results is the review result of the SNL rule.

[0047] Errors found in the review results of SNL rules are extracted, categorized, and the review results are displayed according to different error types. Error information is then supplemented based on the error type. For example, for integrity errors such as missing attributes, the entity number of the missing attribute and its name will be added; for compliance errors such as building spacing not conforming to rule constraints, the entity number of the error and its corresponding attribute value will be added.

[0048] This invention, based on the SNL syntax parsing framework, parses review rules and then uses SQL generation methods to generate corresponding semantic query statements. While combining spatial computation for compliance review, it employs spatial query optimization calculation methods to achieve automated review of GIS models based on SNL rules. This addresses the lack of scalability in existing urban spatial building code review tools, allowing users to easily describe and inspect building codes using custom rules.

[0049] Example 2

[0050] Based on the same inventive concept, this embodiment discloses a GIS automatic review system based on structured natural language rules, such as... Figure 4 As shown, it includes:

[0051] The SNL rule parsing module is used to parse the syntactic components of the SNL rules to be reviewed and analyze the rule type of each SNL clause in the SNL rule.

[0052] The query module is used to query the geographic database into which the GIS model to be reviewed is imported, based on grammatical components and rule types, to obtain the set of target entities and attribute information corresponding to the rule types.

[0053] The semantic query and spatial computation combined processing module is used to perform logical operations on the target entity sets of various rule types to obtain the entity sets participating in spatial geometric calculations;

[0054] The spatial computation optimization module based on spatial query is used to filter the set of entities participating in spatial geometric calculations through spatial topological constraints, and then perform calculations on the filtered set of entities to obtain the review results of each SNL clause.

[0055] The results output module is used to perform logical operations on the review results of each SNL clause according to the logical relationship between each SNL clause, and obtain the review results of the SNL rules.

[0056] Example 2

[0057] Based on the same inventive concept, this embodiment discloses a computer-readable storage medium storing a computer program, which is executed by a processor to implement any of the above-mentioned GIS automatic review methods based on structured natural language rules.

[0058] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0059] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0060] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0061] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0062] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific embodiments of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the protection scope of the claims of the present invention. The above content is only a specific embodiment of this application, but the protection scope of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in this application should be covered within the protection scope of this application. Therefore, the protection scope of this application should be determined by the protection scope of the claims.

Claims

1. A method for automatic GIS review based on structured natural language rules, characterized in that, Includes the following steps: Parse the syntactic components of the SNL rule to be reviewed and analyze the rule type of each SNL clause in the SNL rule; Based on the grammatical components and rule types, a query is performed in the geographic database of the imported GIS model to be reviewed to obtain the target entity set and attribute information corresponding to the rule type; Perform logical operations on the target entity sets of each rule type to obtain the entity sets participating in spatial geometric calculations; The set of entities participating in the spatial geometry calculation is filtered by spatial topological constraints, and the filtered set of entities is processed to obtain the review results of each SNL clause. The method for filtering the set of entities participating in spatial geometric calculation is as follows: parsing the computational rules in the set of entities to generate topological relations, constraints and rule subjects; generating a set of entities participating in spatial query based on the rule subjects and the set of entities participating in the calculation; performing a spatial query based on the topological relations, constraints and the set of entities participating in the spatial query to obtain a set of entities that satisfy the topological relations; and performing logical operations on the set of entities that satisfy the topological relations, constraints and the set of entities participating in the calculation to generate a set of entities participating in the calculation that satisfy the spatial constraints. Based on the logical relationship between each SNL clause, logical operations are performed on the review results of each SNL clause to obtain the review results of the SNL rule; A query is performed in the geographic database using a query statement, which is generated using the syntactic components and rule types. The method for generating the query statement using the syntactic components and rule types is as follows: For a single semantic query rule, the entity name constraint condition for the corresponding WHERE clause is generated through structural mapping; For multiple semantic query rules, the two types of semantic query rules are distinguished by determining whether the SNL clause is in the conclusion clause. Based on the logical relationship between the corresponding SNL clauses of each semantic query rule, logical operators are used to connect multiple constraints with the same entity name in the same type of semantic query rules. The logically processed constraints corresponding to the entity name are used as the WHERE clause to generate the query statement for the corresponding entity.

2. The GIS automatic review method based on structured natural language rules as described in claim 1, characterized in that, The grammatical components include entity names and constraints; the rule types include conclusion clauses and non-rule clauses.

3. The GIS automatic review method based on structured natural language rules as described in claim 1, characterized in that, The method for obtaining the set of entities participating in spatial geometric calculations is as follows: by performing syntax parsing and semantic analysis on the query statement, a set of semantic query rules and a set of computational query rules are obtained; the set of semantic query rules is processed to obtain semantically constrained entities of the conclusion clause and semantically constrained entities of the non-conclusion clause; the set of computational rules is classified into computational rules to obtain a set of computational rules of the conclusion clause and a set of computational rules of the non-conclusion clause.

4. The GIS automatic review method based on structured natural language rules as described in claim 1, characterized in that, The semantic constraint entities of the conclusion clause and the set of computational rules of the conclusion clause are assigned to the result set of the conclusion clause. The semantic constraint entities of the non-conclusion clause and the set of computational rules of the non-conclusion clause are assigned to the result set of the non-conclusion clause. Logical operations are performed on the result set of the conclusion clause and the result set of the non-conclusion clause to obtain the result set of computational rules.

5. The GIS automatic review method based on structured natural language rules as described in any one of claims 1-4, characterized in that, Errors appearing in the review results of the SNL rules are extracted, the errors are classified, the review results are displayed according to different error types, and error information is supplemented according to the error types.

6. A GIS automatic review system based on structured natural language rules, characterized in that, include: The SNL rule parsing module is used to parse the syntactic components of the SNL rule to be reviewed and analyze the rule type of each SNL clause in the SNL rule. The query module is used to query the geographic database into which the GIS model to be reviewed is imported, based on the grammatical components and rule types, to obtain the target entity set and attribute information corresponding to the rule types; The semantic query and spatial computation combined processing module is used to perform logical operations on the target entity sets of various rule types to obtain the entity sets participating in spatial geometric calculations; The spatial computation optimization module based on spatial query is used to filter the set of entities participating in spatial geometric computation through spatial topological constraints, and to perform computational processing on the filtered set of entities to obtain the review results of each SNL clause. The method for filtering the set of entities participating in spatial geometric calculation is as follows: parsing the computational rules in the set of entities to generate topological relations, constraints and rule subjects; generating a set of entities participating in spatial query based on the rule subjects and the set of entities participating in the calculation; performing a spatial query based on the topological relations, constraints and the set of entities participating in the spatial query to obtain a set of entities that satisfy the topological relations; and performing logical operations on the set of entities that satisfy the topological relations, constraints and the set of entities participating in the calculation to generate a set of entities participating in the calculation that satisfy the spatial constraints. The result output module is used to perform logical operations on the review results of each SNL clause according to the logical relationship of each SNL clause, and obtain the review results of the SNL rule. A query is performed in the geographic database using a query statement, which is generated using the syntactic components and rule types. The method for generating the query statement using the syntactic components and rule types is as follows: For a single semantic query rule, the entity name constraint condition for the corresponding WHERE clause is generated through structural mapping; For multiple semantic query rules, the two types of semantic query rules are distinguished by determining whether the SNL clause is in the conclusion clause. Based on the logical relationship between the corresponding SNL clauses of each semantic query rule, logical operators are used to connect multiple constraints with the same entity name in the same type of semantic query rules. The logically processed constraints corresponding to the entity name are used as the WHERE clause to generate the query statement for the corresponding entity.

7. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that is executed by a processor to implement the GIS automatic review method based on structured natural language rules as described in any one of claims 1-5.

Images