A method and system for automatically constructing a building field knowledge base based on multi-source heterogeneous data, and a medium

Abstract

The application discloses a building field knowledge base automatic construction method and system based on multi-source heterogeneous data and a medium, and the method comprises the following steps: S1, intelligent document analysis, realizing document element layout analysis and structured reconstruction; S2, knowledge element extraction and calculation, based on the structured data stream obtained in S1, adopting a GlobalPointer global pointer recognition architecture to extract multi-level nested entities, combining a Transformer-CRF joint probability model and a comparative learning algorithm, and converting the extracted entities, attributes and logical correlations into standardized triples; S3, heterogeneous knowledge graph storage and fusion, based on the standardized knowledge elements obtained in S2, constructing a knowledge graph supporting implicit logical reasoning through a graph representation learning algorithm, and configuring a three-way hybrid index architecture of "vector + graph + full text"; S4, interactive application, completing high-confidence semantic retrieval and intelligent question answering through a RAG retrieval enhancement method; and S5, knowledge base dynamic evolution and management.

Description

Technical Field

[0001] This application belongs to the field of information technology in building engineering, and in particular relates to a method, system and medium for automatically constructing a knowledge base in the building field based on multi-source heterogeneous data. Background Technology

[0002] The construction engineering field is a knowledge-intensive industry, characterized by its wide range of knowledge resources, diverse formats, and complex structures. Currently, a large amount of professional knowledge in the construction field is scattered across various unstructured or semi-structured documents in multiple formats, such as PDF technical specifications, Word design instructions, Excel bills of quantities, CAD drawings, and scanned images. These documents carry key knowledge throughout the entire lifecycle of building design, construction, and operation and maintenance, including core information such as design parameters, construction techniques, material specifications, quality standards, and safety regulations.

[0003] Traditional architectural knowledge management mainly relies on manual organization and annotation, that is, professional technicians read documents one by one, manually extract key information, and manually enter it into the database or compile indexes. This method has the following significant drawbacks: (1) Inefficiency: Architectural documents are usually large in length and dense in professional terminology. Manually processing a single document often takes several hours or even days, making it difficult to meet the knowledge extraction needs of massive documents. (2) High cost: It requires a large number of technical personnel with architectural backgrounds. The labor cost increases linearly with the size of the documents, and the economic feasibility of building a large-scale knowledge base is poor. (3) Difficulty in ensuring consistency: Different people have different understandings and annotations of the same terms and parameters, resulting in inconsistent knowledge representation and affecting subsequent knowledge integration and sharing. (4) Lack of knowledge association: It is difficult to systematically identify the logical structure inside the document and the entity associations across documents in the manual method. The knowledge is in a fragmented state and cannot form a systematic knowledge network.

[0004] With the development of artificial intelligence technology, natural language processing (NLP), optical character recognition (OCR), knowledge graph and other technologies have made significant progress in general fields, but still face many challenges in knowledge extraction and structuring in the construction field: (1) Insufficient ability to parse multi-format documents: Existing document parsing tools are mostly optimized for single formats and lack the ability to uniformly parse common document formats in the construction field such as PDF, Word, Excel, and scanned images. In particular, for engineering documents containing complex layouts, nested tables, and mixed text and graphics, existing technologies are unable to accurately extract structured information. (2) Limited accuracy in recognizing construction terminology: The general named entity recognition (NER) model has a low accuracy in recognizing construction professional terms (such as component names, material models, construction techniques, etc.). There are a large number of professional abbreviations, symbol representations and polysemous terms in the construction field, and existing technologies lack targeted domain dictionaries and semantic disambiguation mechanisms. (3) Weak ability to understand document structure: Construction documents have a strict hierarchical structure (such as chapter-clause-subclause-parameter), and the structure of different document types varies significantly. Existing technologies focus primarily on pure text analysis, lacking sufficient ability to jointly model the logical structure and layout features of documents (such as heading levels, list indentation, table positions, etc.), making it difficult to achieve accurate document tree construction and content hierarchy extraction. (4) Entity relationship extraction is inadequate: Architectural knowledge involves complex entity relationships, such as "component-material" relationships, "process-parameter" relationships, and "standard-scope of application" relationships. Existing relationship extraction methods are mostly based on training with general corpora, and have limited ability to identify relationship types and implicit relationships specific to the architectural field. (5) The knowledge graph construction and maintenance mechanism is imperfect: Existing knowledge graph construction processes are mostly offline batch processing, lacking an incremental update mechanism for the dynamic evolution of knowledge in the architectural field. At the same time, quality control links such as knowledge conflict detection and consistency verification are weak, making it difficult to guarantee the reliability and timeliness of the knowledge base.

[0005] Existing technological bottlenecks prevent current methods from achieving large-scale, high-precision knowledge structuring and knowledge base construction in the construction field, severely restricting the development of the following application scenarios: Intelligent retrieval systems: unable to achieve accurate cross-document and cross-format knowledge retrieval based on semantic understanding; Intelligent question-answering systems: lacking structured knowledge support, making it difficult to answer professional questions involving multi-source information; Decision support systems: knowledge fragmentation prevents effective knowledge reasoning and solution recommendation; Knowledge reuse and collaboration: unstructured knowledge forms hinder the cross-project reuse of design experience and construction techniques.

[0006] Therefore, there is an urgent need for an automated construction method for a knowledge base in the construction field based on multi-source heterogeneous data. By integrating technologies such as document parsing, domain natural language processing, and knowledge graphs, this method can achieve efficient parsing, accurate extraction, structuring, and knowledge-based processing of construction documents, providing fundamental support for the digital transformation and intelligent upgrading of the construction industry. Summary of the Invention

[0007] The purpose of this application is to overcome the problems of existing technology by disclosing an automated construction method, system and medium for building domain knowledge base based on multi-source heterogeneous data, aiming to solve the inefficiency caused by knowledge fragmentation and unstructured nature in building engineering documents through multi-dimensional technology integration.

[0008] On the one hand, the objective of this application is achieved through the following technical solution: An automated method for constructing a knowledge base in the construction field based on multi-source heterogeneous data, the method comprising: S1: Intelligent document parsing. Based on a multimodal layout perception model that integrates visual topology and text sequence features, it intelligently parses heterogeneous documents such as PDF, Word, and scanned images, and completes visual topology feature extraction and text sequence feature extraction to achieve document element layout analysis and structured reconstruction. S2: Knowledge element extraction and computation. Based on the structured data stream obtained in S1, the GlobalPointer global pointer recognition architecture is used to extract multi-level nested entities. Combined with the Transformer-CRF joint probability model and contrastive learning algorithm, the extracted entities, attributes, and logical relationships are transformed into standardized triples. ,in, Represents a standardized triple. Indicates the head entity. Indicates a relationship. It represents tail entities or attribute values ​​and automatically maps them to a unified building ontology semantic space; S3: Heterogeneous knowledge graph storage and fusion. Based on the standardized knowledge elements obtained by S2, a knowledge graph supporting implicit logical reasoning is constructed through graph representation learning algorithms, and a three-way hybrid indexing architecture of "vector + graph + full text" is configured. S4: Interactive application, which uses the RAG retrieval enhancement method to achieve high-confidence semantic retrieval and intelligent question answering; S5: Dynamic evolution and management of the knowledge base, completing the dynamic incremental update and logical self-healing of the knowledge base obtained in step S3.

[0009] According to a preferred embodiment, step S1 includes: introducing two-dimensional position coding. Spatial modeling of document elements, where, This indicates layout-aware embedding. This indicates text semantic embedding. Represents a two-dimensional spatial location code; By using spatial coordinate constraints, we can accurately extract hierarchical directories, table parameters, and figure annotation information from building codes.

[0010] According to a preferred embodiment, step S2 further includes: A cosine similarity and edit distance fusion algorithm is introduced to map the extracted entities to a standard ontology library:

[0011] in, This represents the candidate entities that were selected. Represents a reference entity in the standard ontology library. These represent the candidate embedding vector and the reference embedding vector, respectively. These are text strings representing the candidate entity and the reference entity, respectively. This is the fusion weight coefficient.

[0012] According to a preferred implementation, in step S3, a knowledge base based on graph embedding is constructed: The TransE or RotatE algorithm is used to map knowledge triples to a low-dimensional continuous vector space, making... .

[0013] According to a preferred embodiment, in step S5, a consistency auditing mechanism based on first-order predicate logic and a semantic difference comparison algorithm are established to realize the dynamic incremental update and logical self-healing of the knowledge base obtained in step S3.

[0014] According to a preferred embodiment, step S5 includes: Establish semantic consistency constraint checks and define a set of logical rules. The logic inference engine is used to detect whether newly added knowledge conflicts with existing clauses.

[0015] in, Represents entity variables, Represents a set of entities. Represents a set of relations.

[0016] On the other hand, this application also discloses: An automated construction system for a knowledge base in the construction field based on multi-source heterogeneous data is provided, wherein the automated construction system for the knowledge base in the construction field is based on the aforementioned method.

[0017] According to a preferred embodiment, the automated construction system for the building domain knowledge base includes: The multimodal document intelligent parsing module is used to achieve intelligent document parsing; The knowledge element extraction and calculation module is used to implement knowledge element extraction and calculation. The heterogeneous knowledge graph storage and fusion module is used to realize the storage and fusion of heterogeneous knowledge graphs. The intelligent application service interface module connects to the building intelligent application terminal to enable interactive applications; The knowledge base dynamic evolution and management module is used to realize the homomorphic update of the knowledge base established by the heterogeneous knowledge graph storage and fusion module.

[0018] On the other hand, this application also discloses: A computer-readable storage medium for storing instructions that, when executed, cause the aforementioned method to be implemented.

[0019] The aforementioned main solution and its various further alternative solutions can be freely combined to form multiple solutions, all of which are solutions that can be adopted and are claimed in this application. Those skilled in the art, after understanding the solution of this application, will realize that there are many combinations based on the prior art and common general knowledge, all of which are technical solutions to be protected in this application, and will not be exhaustively listed here.

[0020] The beneficial effects of this application are: This application presents an automated construction method for a knowledge base in the construction field based on multi-source heterogeneous data. By integrating multi-dimensional technologies, it solves the problem of low efficiency caused by knowledge fragmentation and unstructured nature in construction engineering documents.

[0021] 1. Unified parsing and information extraction of multi-format documents: Supports automatic parsing of architectural documents in various heterogeneous formats such as PDF, Word, Excel, and scanned images, extracting text, tables, images, and structured data, providing a unified data interface for subsequent knowledge extraction.

[0022] 2. Entity Recognition and Terminology Standardization in the Architectural Field: Based on an architectural dictionary and named entity recognition model, architectural entities in text are identified and mapped and encoded with a standard terminology library to ensure terminology consistency.

[0023] 3. Document Structure Understanding and Relationship Extraction: Based on the document's logical structure and layout features, a document tree structure is constructed to extract and annotate relationships among multiple levels of content such as chapters, clauses, and parameters.

[0024] 4. Standardized construction of knowledge triples: The extracted entities, attributes and relations are transformed into triples in the form of (entity-relation-entity) or (entity-attribute-value), providing structured input for the construction of standard knowledge graphs.

[0025] 5. Knowledge Graph Construction and Dynamic Update Mechanism: A knowledge graph for the architectural domain is constructed based on triples, an index structure supporting multi-dimensional queries is established, and an incremental update and consistency verification mechanism is designed to support the continuous evolution of the knowledge base. Attached Figure Description

[0026] Figure 1 This is a schematic diagram of the system principle of this application. Detailed Implementation

[0027] The following specific examples illustrate the implementation of this application. Those skilled in the art can easily understand other advantages and effects of this application from the content disclosed in this specification. This application can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of this application. It should be noted that, unless otherwise specified, the following embodiments and features in the embodiments can be combined with each other.

[0028] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0029] In the description of this application, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, or the orientation or positional relationship commonly used when the product of this application is in use. They are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation on this application. In addition, the terms "first," "second," and "third," etc., are only used to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0030] Furthermore, terms such as "horizontal," "vertical," and "sag" do not imply that components must be absolutely horizontal or suspended, but rather that they can be slightly tilted. For example, "horizontal" simply means that its direction is more horizontal relative to "vertical," and does not mean that the structure must be completely horizontal, but can be slightly tilted.

[0031] In the description of this application, it should also be noted that, unless otherwise expressly specified and limited, the terms "set up," "install," "connect," and "link" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.

[0032] Furthermore, it should be noted that unless otherwise specified in this application, the specific structures, connections, positions, power sources, etc. involved are all things that a person skilled in the art can know without creative effort based on the prior art.

[0033] Example 1 refer to Figure 1 As shown, this application discloses an automated construction method for a knowledge base in the construction field based on multi-source heterogeneous data. The automated construction method for a knowledge base in the construction field based on multi-source heterogeneous data includes the following steps.

[0034] Step S1: Intelligent document parsing. Based on a multimodal layout perception model that integrates visual topology and text sequence features, the system performs intelligent parsing of heterogeneous documents such as PDF, Word, and scanned images, completing visual topology feature extraction and text sequence feature extraction to achieve document element layout analysis and structured reconstruction.

[0035] Preferably, step S1 includes: introducing two-dimensional position coding. Spatial modeling of document elements, where, This indicates layout-aware embedding. This indicates text semantic embedding. It represents a two-dimensional spatial location code; through spatial coordinate constraints, it accurately extracts hierarchical directories, table parameters, and figure annotation association information from building codes.

[0036] It supports automatic parsing of architectural documents in various heterogeneous formats such as PDF, Word, Excel, and scanned images, extracting text, tables, images, and structured data, providing a unified data interface for subsequent knowledge extraction.

[0037] Step S2: Knowledge element extraction and computation. Based on the structured data stream obtained in S1, the GlobalPointer global pointer recognition architecture is used to extract multi-level nested entities. Combined with the Transformer-CRF joint probability model and contrastive learning algorithm, the extracted entities, attributes, and logical relationships are transformed into standardized triples. ,in, Represents a standardized triple. Indicates the head entity. Indicates a relationship. It represents tail entities or attribute values ​​and automatically maps them to a unified building ontology semantic space.

[0038] Preferably, step S2 further includes: A cosine similarity and edit distance fusion algorithm is introduced to map the extracted entities to a standard ontology library:

[0039] in, This represents the candidate entities that were selected. Represents a reference entity in the standard ontology library. These represent the candidate embedding vector and the reference embedding vector, respectively. These are text strings representing the candidate entity and the reference entity, respectively. This is the fusion weighting coefficient.

[0040] Step S3: Heterogeneous knowledge graph storage and fusion. Based on the standardized knowledge elements obtained in S2, a knowledge graph supporting implicit logical reasoning is constructed through graph representation learning algorithms, and a three-way hybrid index architecture of "vector + graph + full text" is configured.

[0041] Preferably, in step S3, a knowledge base based on graph embedding is constructed: the knowledge triples are mapped to a low-dimensional continuous vector space using the TransE or RotatE algorithm, so that... .

[0042] Step S4: Interactive application, using the RAG retrieval enhancement method to complete high-confidence semantic retrieval and intelligent question answering.

[0043] Step S5: Dynamic evolution and management of the knowledge base, completing the dynamic incremental update and logical self-healing of the knowledge base obtained in Step S3.

[0044] Preferably, in step S5, a consistency auditing mechanism based on first-order predicate logic and a semantic difference comparison algorithm are established to realize the dynamic incremental update and logical self-healing of the knowledge base obtained in step S3.

[0045] Further, step S5 includes: Establish semantic consistency constraint checks and define a set of logical rules. The logic inference engine is used to detect whether newly added knowledge conflicts with existing clauses.

[0046] in, Represents entity variables, Represents a set of entities. Represents a set of relations.

[0047] This application presents an automated construction method for a knowledge base in the construction field based on multi-source heterogeneous data. By integrating multi-dimensional technologies, it solves the problem of low efficiency caused by knowledge fragmentation and unstructured nature in construction engineering documents.

[0048] Example 2 refer to Figure 1As shown in Example 1, this example discloses an automated construction system for a building domain knowledge base based on multi-source heterogeneous data. The automated construction system for a building domain knowledge base is based on the method described in Example 1 to automatically construct a building domain knowledge base.

[0049] Preferably, the automated knowledge base construction system in the construction field includes the following modules: a multimodal document intelligent parsing module for intelligent document parsing; a knowledge element extraction and calculation module for knowledge element extraction and calculation; a heterogeneous knowledge graph storage and fusion module for storing and fusing heterogeneous knowledge graphs; an intelligent application service interface module for interactive applications through connection with a building intelligent application terminal; and a knowledge base dynamic evolution and management module for homomorphic updates of the knowledge base established by the heterogeneous knowledge graph storage and fusion module.

[0050] Example 3 Based on Embodiment 1, this embodiment also discloses: a computer-readable storage medium for storing instructions that, when executed, cause the method described in Embodiment 1 to be implemented.

[0051] In some alternative embodiments, the present invention also provides that various aspects of the method for automatically constructing a building domain knowledge base based on multi-source heterogeneous data can also be implemented in the form of a program product, which includes program code that, when the program product is run on a device, causes the control device to perform the steps in the method for automatically constructing a building domain knowledge base based on multi-source heterogeneous data according to various exemplary embodiments of the present invention as described above.

[0052] It should be noted that although several units or sub-units of the apparatus have been mentioned in the detailed description above, this division is merely exemplary and not mandatory. In fact, according to embodiments of the invention, the features and functions of two or more units described above can be embodied in one unit. Conversely, the features and functions of one unit described above can be further divided and embodied by multiple units. Furthermore, although the operation of the method of the invention is described in a specific order in the drawings, this does not require or imply that these operations must be performed in that specific order, or that all the operations shown must be performed to achieve the desired result. Additionally or alternatively, certain steps may be omitted, multiple steps may be combined into one step, and / or one step may be broken down into multiple steps.

[0053] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can be implemented in one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROMs) containing computer-usable program code. The form of a computer program product implemented on ROM, optical memory, etc.

[0054] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. 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 server, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0055] Program code for performing the operations of this invention can be written using any combination of one or more programming languages, including object-oriented programming languages ​​such as Java and C++, as well as conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on the user's computing device, partially on the user's device, as a standalone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0056] In cases involving remote computing devices, the remote computing device can be connected to the user's computing device via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0057] 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.

[0058] 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.

Claims

1. A method for automatically constructing a knowledge base in the construction field based on multi-source heterogeneous data, characterized in that, The method for automatically constructing a knowledge base in the construction field based on multi-source heterogeneous data includes: S1: Intelligent document parsing. Based on a multimodal layout perception model that integrates visual topology and text sequence features, it intelligently parses heterogeneous documents such as PDF, Word, and scanned images, and completes visual topology feature extraction and text sequence feature extraction to achieve document element layout analysis and structured reconstruction. S2: Knowledge element extraction and computation. Based on the structured data stream obtained in S1, the GlobalPointer global pointer recognition architecture is used to extract multi-level nested entities. Combined with the Transformer-CRF joint probability model and contrastive learning algorithm, the extracted entities, attributes, and logical relationships are transformed into standardized triples. ,in, Represents a standardized triple. Indicates the head entity. Indicates a relationship. It represents tail entities or attribute values ​​and automatically maps them to a unified building ontology semantic space; S3: Heterogeneous knowledge graph storage and fusion. Based on the standardized knowledge elements obtained by S2, a knowledge graph supporting implicit logical reasoning is constructed through graph representation learning algorithms, and a three-way hybrid indexing architecture of "vector + graph + full text" is configured. S4: Interactive application, which uses the RAG retrieval enhancement method to achieve high-confidence semantic retrieval and intelligent question answering; S5: Dynamic evolution and management of the knowledge base, completing the dynamic incremental update and logical self-healing of the knowledge base obtained in step S3.

2. The automated construction method for a knowledge base in the construction field based on multi-source heterogeneous data as described in claim 1, characterized in that, Step S1 includes: introducing two-dimensional position coding Spatial modeling of document elements, where, This indicates layout-aware embedding. This indicates text semantic embedding. Represents a two-dimensional spatial location code; By using spatial coordinate constraints, we can accurately extract hierarchical directories, table parameters, and figure annotation information from building codes.

3. The method for automatically constructing a knowledge base in the construction field based on multi-source heterogeneous data as described in claim 1, characterized in that, Step S2 also includes: A cosine similarity and edit distance fusion algorithm is introduced to map the extracted entities to a standard ontology library: in, This represents the candidate entities that were selected. Represents a reference entity in the standard ontology library. These represent the candidate embedding vector and the reference embedding vector, respectively. These are text strings representing the candidate entity and the reference entity, respectively. This is the fusion weight coefficient.

4. The method for automatically constructing a knowledge base in the construction field based on multi-source heterogeneous data as described in claim 1, characterized in that, In step S3, a knowledge base for learning Graph Embedding based on graph representation is constructed: The TransE or RotatE algorithm is used to map knowledge triples to a low-dimensional continuous vector space, making... .

5. The method for automatically constructing a knowledge base in the construction field based on multi-source heterogeneous data as described in claim 4, characterized in that, In step S5, by establishing a consistency auditing mechanism based on first-order predicate logic and a semantic difference comparison algorithm, the dynamic incremental update and logical self-healing of the knowledge base obtained in step S3 are realized.

6. The method for automatically constructing a knowledge base in the construction field based on multi-source heterogeneous data as described in claim 5, characterized in that, Step S5 includes: Establish semantic consistency constraint checks and define a set of logical rules. The logic inference engine is used to detect whether newly added knowledge conflicts with existing clauses. in, Represents entity variables, Represents a set of entities. Represents a set of relations.

7. An automated construction system for a knowledge base in the construction field based on multi-source heterogeneous data, characterized in that, The automated construction system for the building domain knowledge base is based on the method described in any one of claims 1 to 6 to automatically construct the building domain knowledge base.

8. The automated construction system for a knowledge base in the construction field based on multi-source heterogeneous data as described in claim 7, characterized in that, The automated knowledge base construction system for the construction field includes: The multimodal document intelligent parsing module is used to achieve intelligent document parsing; The knowledge element extraction and calculation module is used to implement knowledge element extraction and calculation. The heterogeneous knowledge graph storage and fusion module is used to realize the storage and fusion of heterogeneous knowledge graphs. The intelligent application service interface module connects to the building intelligent application terminal to enable interactive applications; The knowledge base dynamic evolution and management module is used to realize homomorphic updates of the knowledge base established by the heterogeneous knowledge graph storage and fusion module.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium is used to store instructions that, when executed, cause the method as described in any one of claims 1 to 8 to be implemented.

Images