Engineering domain knowledge graph construction method and system

Abstract

The invention relates to an engineering field knowledge graph construction method and system. The engineering field knowledge graph construction method comprises the following steps: S1, obtaining an engineering specification document and a contract document; s2, extracting technical elements from the engineering specification document, and constructing a technical knowledge sub-graph which comprises mutually associated technical parameter nodes; s3, responsibility elements are extracted from the contract document, a contract knowledge sub-graph is constructed, and the contract knowledge sub-graph comprises mutually associated contract term nodes; according to the method, information islands are broken, three-layer knowledge graph fusion is carried out on the project specification document, the contract document and the project decomposition structure, three-dimensional penetration association of technical parameters-contract terms-project entities is established, automatic mapping of the technical parameters, the contract terms and the project entities is realized, and the project quality is improved. The efficiency of engineering technology compliance check and contract risk identification is greatly improved, and low efficiency and high error rate of traditional manual management are avoided.

Description

Technical Field

[0001] The present invention relates to the field of engineering construction technology, and in particular to a method and system for constructing a knowledge graph in the engineering field. Background Art

[0002] In the engineering field, the technical parameters and contract terms of the project are scattered in the engineering specification documents and contract documents respectively. The technical requirements of the sub-projects lack connection with the contract rights and responsibilities. When conducting project management, it is necessary to rely on manual review and manual experience to conduct linkage analysis. Not only is the technical compliance check time-consuming, but there is also the problem of delayed contract risk identification. Summary of the Invention

[0003] The purpose of the present invention is to overcome the defects of the existing technology and provide a method and system for constructing a knowledge graph in the engineering field to solve the problems of time-consuming engineering technology compliance inspection and delayed contract risk identification in the existing technology.

[0004] To achieve the above objectives, the present invention provides a method for constructing an engineering knowledge graph, comprising the following steps: S1. Obtain engineering specification documents and contract documents; S2. Extract technical elements from the engineering specification document and construct a technical knowledge subgraph, wherein the technical knowledge subgraph includes interrelated technical parameter nodes; S3. Extracting liability elements from the contract document and constructing a contract knowledge subgraph, wherein the contract knowledge subgraph includes mutually related contract clause nodes; S4. Constructing a project entity subgraph based on the project decomposition structure, wherein the project entity subgraph includes hierarchically organized project entity nodes; S5. Establish a cross-domain association relationship between the technical knowledge subgraph, the contract knowledge subgraph, and the engineering entity subgraph to form a complete engineering knowledge graph.

[0005] By adopting this technical solution, information silos are broken, and the three-layer knowledge graph of engineering specification documents, contract documents and engineering decomposition structures is integrated to establish a three-dimensional penetrating association of "technical parameters-contract terms-engineering entities". The automated mapping of technical parameters, contract terms and engineering entities is achieved, which greatly improves the efficiency of engineering technical compliance inspections and contract risk identification, and avoids the low efficiency and high error rate of traditional manual management.

[0006] Furthermore, the technical parameter nodes in step S2 include material parameter nodes, process parameter nodes and acceptance parameter nodes; wherein, The material parameter node represents the performance index of the engineering material; The process parameter node represents the quality standard of the construction process; The acceptance parameter node represents the inspection requirements for project acceptance.

[0007] Furthermore, the contract clause node in step S3 includes a liability clause node, a risk clause node, and a claim clause node; wherein, The liability clause node represents the rights and obligations of each party to the contract; The risk clause node represents the risk allocation rule; The claim clause node represents a dispute resolution mechanism.

[0008] Furthermore, the hierarchical organization of the engineering entity nodes in step S4 includes project-level nodes, unit engineering nodes, and sub-project nodes; wherein, The project-level node represents the overall engineering project; The unit project node represents an engineering unit that independently performs the project; The sub-project node represents the smallest manageable project unit.

[0009] Furthermore, the cross-domain association relationship in step S5 includes technical compliance association, responsibility binding association and causal triggering association; wherein, The technical compliance association connects the engineering entity node with the technical parameter nodes that meet its technical requirements; The liability binding association connects the project entity node and the contract clause node that constrains its execution liability; The causal trigger association connects the technical parameter node and the contract clause node affected by it.

[0010] Furthermore, it also includes: In response to engineering change information, dynamically updating the nodes and association relationships of the engineering knowledge graph; Based on the graph traversal algorithm, the impact path of changes in technical elements on contract elements is derived.

[0011] By adopting this technical solution, the timeliness and accuracy of the knowledge graph can be ensured, thereby enhancing the applicability of the knowledge graph.

[0012] The present invention also provides a system for implementing a method for constructing a knowledge graph in the engineering field, comprising a multi-source data acquisition module, a knowledge extraction module, an engineering modeling module, and a graph fusion module connected in sequence; wherein, The multi-source data acquisition module is used to collect engineering specification documents and contract documents; The knowledge extraction module is used to construct technical knowledge subgraph and contract knowledge subgraph; The engineering modeling module is used to construct engineering entity subgraphs; The graph fusion module is used to establish cross-domain association relationships.

[0013] Compared with the prior art, the present invention has the following beneficial effects: Break down information silos, integrate engineering specification documents, contract documents, and engineering breakdown structures into three-layer knowledge graphs, establish a three-dimensional penetrating association of "technical parameters-contract terms-engineering entities", and realize the automated mapping of technical parameters, contract terms, and engineering entities. This greatly improves the efficiency of engineering technical compliance checks and contract risk identification, and avoids the low efficiency and high error rate of traditional manual management. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 This is a flow chart of the method for constructing a knowledge graph in the engineering field in the present invention. DETAILED DESCRIPTION

[0015] The present invention will be further described below with reference to the accompanying drawings and specific embodiments.

[0016] Please see the attached Figure 1 , the present invention provides a method for constructing a knowledge graph in the engineering field, comprising the following steps: S1. Obtain engineering specification documents and contract documents; S2. Extract technical elements from engineering specification documents and construct a technical knowledge subgraph. The technical knowledge subgraph contains interrelated technical parameter nodes. Technical parameter nodes include material parameter nodes, process parameter nodes, and acceptance parameter nodes. Material parameter nodes represent the performance indicators of engineering materials; process parameter nodes represent the quality standards of construction processes; and acceptance parameter nodes represent the inspection requirements for project acceptance. S3. Extract liability elements from the contract document and construct a contract knowledge subgraph. The contract knowledge subgraph contains interrelated contract clause nodes. Contract clause nodes include liability clause nodes, risk clause nodes, and claim clause nodes. Liability clause nodes represent the rights and obligations of the contracting parties; risk clause nodes represent risk allocation rules; and claim clause nodes represent dispute resolution mechanisms. S4. Construct a project entity subgraph based on the project breakdown structure. The project entity subgraph contains hierarchically organized project entity nodes. The hierarchical organization of project entity nodes includes project-level nodes, unit project nodes, and sub-project nodes. Among them, the project-level node represents the entire project; the unit project node represents the independent project unit; and the sub-project node represents the smallest manageable project unit. S5. Establish cross-domain associations between the technical knowledge subgraph, the contract knowledge subgraph, and the engineering entity subgraph to form a complete engineering knowledge graph. Cross-domain associations include technical compliance associations, responsibility binding associations, and causal triggering associations. The technical compliance association connects engineering entity nodes with technical parameter nodes that meet their technical requirements; the responsibility binding association connects engineering entity nodes with the contract clause nodes that constrain their execution responsibilities; and the causal triggering association connects technical parameter nodes with the contract clause nodes affected by them. It also includes dynamically updating the nodes and association relationships of the engineering knowledge graph in response to engineering change information; and deducing the impact path of changes in technical elements on contract elements based on the graph traversal algorithm.

[0017] Furthermore, in step S5, the cross-domain association weights are optimized by introducing historical project data and using a graph neural network. Specifically, let the engineering entity be E, the technical parameter be S, the contract clause be F, the similarity function be bert_similarity(x, y), and the historical data weight function be historical_data_weight(x). Then, the cross-domain mapping confidence function is defined as: cross_domain_mapping(E,S,F)=0.4*bert_similarity(E,S)+0.4*bert_similarity(E,F)+0.2*historical_data_weight(E).

[0018] The present invention also provides a system for implementing a method for constructing a knowledge graph in the engineering field, comprising a multi-source data acquisition module, a knowledge extraction module, an engineering modeling module and a graph fusion module connected in sequence, wherein the multi-source data acquisition module is used to collect engineering specification documents and contract documents; the knowledge extraction module is used to construct a technical knowledge subgraph and a contract knowledge subgraph; the engineering modeling module is used to construct an engineering entity subgraph; and the graph fusion module is used to establish cross-domain association relationships.

[0019] In order to provide service interfaces such as risk engineering query, risk assessment, and solution recommendation, it should also include an application module connected to the graph fusion module. The application module is used to visually display practical pages such as sub-project query page, risk assessment display page, and solution recommendation display page.

[0020] In one embodiment, taking a pile foundation project and obtaining relevant SPEC technical parameters and FIDIC contract terms as an example: Extract technical elements from SPEC technical parameters and construct a technical knowledge subgraph. The technical knowledge subgraph contains interrelated technical parameter nodes, such as "pile length ≥ 30m" and "concrete strength C35". Extract liability elements from FIDIC contract clauses and construct a contract knowledge subgraph. The contract knowledge subgraph contains interrelated contract clause nodes, such as the related clauses 4.12 Unforeseeable material conditions and 17.3 Force majeure conditions. Construct a project entity subgraph based on the project breakdown structure. The project entity subgraph contains hierarchically organized project entity nodes, such as establishing a "pile foundation project - unit project - project project" hierarchy. Establish a three-dimensional penetrating association of "technical parameters-contract terms-pile foundation engineering" to realize the automated mapping of technical parameters, contract terms and pile foundation engineering, so as to form a complete engineering knowledge map and enable information interconnection in all aspects of engineering management.

[0021] For example, when the knowledge graph detects the SPEC technical parameter "pile integrity test pass rate <95%", the system automatically links it to the defect notification period in Article 11.3 of the FIDIC contract, triggering the contractor's rectification obligation; and calls the historical case library to recommend a "pile foundation repair plan for a certain project", which includes technical measures and claims strategies.

[0022] The knowledge graph is dynamically updated. For example, when the SPEC technical parameter "Environmental protection requirements for pile foundation construction" is updated, the system automatically recognizes the newly added parameter "Mud water pH value ≥ 6.5"; it is associated with the contractor's environmental protection obligations in Article 4.1 of the FIDIC contract terms and updates the "Pile Foundation Engineering" knowledge graph node.

[0023] The present invention has been described in detail above with reference to the embodiments of the accompanying drawings. A person skilled in the art can make various modifications to the present invention based on the above description. Therefore, certain details in the embodiments should not be construed as limiting the present invention. The scope of protection of the present invention shall be determined by the scope defined in the appended claims.

Claims

1. A method for constructing a knowledge graph in the engineering field, characterized in that: The steps include: S1. Obtain engineering specification documents and contract documents; S2. Extract technical elements from the engineering specification document and construct a technical knowledge subgraph, wherein the technical knowledge subgraph includes interrelated technical parameter nodes; S3. Extracting liability elements from the contract document and constructing a contract knowledge subgraph, wherein the contract knowledge subgraph includes mutually related contract clause nodes; S4. Constructing a project entity subgraph based on the project decomposition structure, wherein the project entity subgraph includes hierarchically organized project entity nodes; S5. Establish a cross-domain association relationship between the technical knowledge subgraph, the contract knowledge subgraph, and the engineering entity subgraph to form a complete engineering knowledge graph.

2. The method for constructing an engineering domain knowledge graph according to claim 1, characterized in that: The technical parameter nodes in step S2 include material parameter nodes, process parameter nodes and acceptance parameter nodes; wherein, The material parameter node represents the performance index of the engineering material; The process parameter node represents the quality standard of the construction process; The acceptance parameter node represents the inspection requirements for project acceptance.

3. The method for constructing an engineering domain knowledge graph according to claim 1, characterized in that: The contract clause nodes in step S3 include liability clause nodes, risk clause nodes, and claim clause nodes; wherein, The liability clause node represents the rights and obligations of each party to the contract; The risk clause node represents the risk allocation rule; The claim clause node represents a dispute resolution mechanism.

4. The method for constructing an engineering domain knowledge graph according to claim 1, wherein: The hierarchical organization of the engineering entity nodes in step S4 includes project-level nodes, unit engineering nodes, and sub-project nodes; wherein, The project-level node represents the overall engineering project; The unit project node represents an engineering unit that independently performs the project; The sub-project node represents the smallest manageable project unit.

5. The method for constructing an engineering domain knowledge graph according to claim 1, characterized in that: The cross-domain association relationship in step S5 includes technical compliance association, responsibility binding association and causal trigger association; among them, The technical compliance association connects the engineering entity node with the technical parameter nodes that meet its technical requirements; The liability binding association connects the project entity node and the contract clause node that constrains its execution liability; The causal trigger association connects the technical parameter node and the contract clause node affected by it.

6. The method for constructing an engineering domain knowledge graph according to claim 1, characterized in that: Also includes: In response to engineering change information, dynamically updating the nodes and association relationships of the engineering knowledge graph; Based on the graph traversal algorithm, the impact path of changes in technical elements on contract elements is derived.

7. A system for implementing the method for constructing an engineering domain knowledge graph according to any one of claims 1 to 6, characterized in that: It includes a multi-source data acquisition module, a knowledge extraction module, an engineering modeling module and a graph fusion module connected in sequence; among them, The multi-source data acquisition module is used to collect engineering specification documents and contract documents; The knowledge extraction module is used to construct technical knowledge subgraph and contract knowledge subgraph; The engineering modeling module is used to construct engineering entity subgraphs; The graph fusion module is used to establish cross-domain association relationships.

Images