Intelligent Classification and Grading Method for Construction Engineering Data Assets

Abstract

The present invention discloses an intelligent classification and grading method for construction project data assets, belonging to the field of construction project data asset management, including: collecting relevant data assets of construction projects, sorting out the data assets, and constructing a hierarchical classification directory for construction project data assets; analyzing the content of the data assets and making document classification marks; classifying the data assets into core assets, important assets and general assets according to the importance level, marking the importance levels of several data assets, and automatically calculating the importance levels of other data assets. This method can automatically calculate the importance levels of other data assets by using the correlation formula and asset score formula between component data assets, realize the accurate classification and reasonable grading of construction project data assets, and improve the management efficiency and use value of data assets.

Description

Technical Field

[0001] The present invention relates to an intelligent classification and grading method for building engineering data assets, belonging to the field of building engineering data asset management. Background Art

[0002] With the rapid development of the construction industry in China, the number of building engineering projects is increasing day by day, the project scale is constantly expanding, and a large number of building engineering data assets have been generated. These data assets include design drawings, construction plans, bill of quantities, on-site photos, acceptance reports, etc., covering all stages such as the pre-project stage, construction process, and final accounts settlement. It is difficult to efficiently manage and utilize these data assets in the construction industry.

[0003] The prerequisite for the management and utilization of building engineering data assets is to classify and grade these data assets. Currently, it mainly relies on manual methods for processing. Due to the lack of unified standards and specifications, the classification of data assets is chaotic, making it inconvenient to search and use; in the vast data assets, searching for specific information requires a large amount of time and manpower; the existing data asset management methods are difficult to discover the importance differences of data and cannot fully utilize the value of data. Summary of the Invention

[0004] Aiming at the problems existing in the current management of building engineering data assets, the present invention provides an intelligent classification and grading method for building engineering data assets, which can accurately classify and reasonably grade building engineering data assets, and improve the management efficiency and utilization value of data assets.

[0005] To solve the above technical problems, the present invention includes the following technical solutions:

[0006] An intelligent classification and grading method for building engineering data assets, comprising:

[0007] Step 1: Collect relevant data assets of building engineering, sort out the data assets, and construct a hierarchical classification directory of building engineering data assets;

[0008] Step 2: Analyze the content of the data assets and make document classification marks; among them, for text data assets, use a large language model for text analysis and automatically make document classification marks; for picture data assets, use computer vision technology to identify and analyze engineering pictures, automatically classify the engineering elements in the pictures, and make document classification marks;

[0009] Step 3: Divide the data assets into core assets, important assets, and general assets according to the importance level, mark the importance levels of several data assets, and automatically calculate the importance levels of other data assets, including:

[0010] Step 3.1: Construct any two data assets of building engineeringP , Q Relevance score D Calculation formula to determine P , Q Whether it is relevant, where

[0011] ;

[0012] In the formula L 、 k Are the total number of layers and the k th layer in the hierarchical classification directory respectively; I k Is a 0-1 function. If P and Q Belong to the same k th layer, then I k = 1, otherwise I k = 0; m Is the number of matches of the P and Q document classification tags extracted in step 2, σ Is the adjustment coefficient;

[0013] Set the threshold D 0. If D > D 0, then it is determined that the data assets < P , Q > are relevant, otherwise, it is determined that the data assets P , Q are not relevant;

[0014] Step 3.2: Determine the relevance between the data assets with unmarked importance levels and the data assets with marked importance levels; calculate the asset score S , where S = 100 a + 10 b + c , in the formula, a 、 b 、 c Are the quantities of the unmarked important assets associated with the core assets, important assets, and general assets respectively;

[0015] Step 3.3: Sort in descending order according to the asset score S and mark the importance levels of the data assets according to the set rules.

[0016] Furthermore, in step 3.3, control the importance levels of the data assets according to the ratio. After sorting the asset score S in descending order, the top C1% of the scored data assets are marked as core assets, and the top C 1% to C 2% of the data assets are marked as important assets, C and the data assets after 2% are marked as general assets, C 1. C 2. C 3 is the preset value, and 0 < C 1 < C 2 < 100.

[0017] Furthermore, C 1 = 5, C 2 = 20.

[0018] Furthermore, the hierarchical classification catalog of construction engineering data assets includes six layers, specifically:

[0019] The first layer: Classify according to the data generation stage;

[0020] The second layer: Classify according to the data source;

[0021] The third layer: Classify according to the data usage;

[0022] The fourth layer: Classify according to the data nature;

[0023] The fifth layer: Classify according to the application scenario;

[0024] The sixth layer: Classify according to the business requirements.

[0025] Furthermore, use the hierarchical clustering algorithm to intelligently allocate the data assets collected in step one into the hierarchical classification catalog of construction engineering data assets.

[0026] Furthermore, σ = 2.

[0027] Furthermore, in step 3.1, D 0 is obtained in the following way: The total number of data assets is N. Randomly pair all data assets in pairs, calculate the correlation scores, sort all the correlation scores, and statistically take the 90th percentile as the threshold D 0.

[0028] Due to the above technical solutions, the present invention has the following advantages and positive effects compared with the prior art: The intelligent classification and grading method of construction project data assets provided by the present invention first constructs a hierarchical classification directory of data assets according to the collected construction project-related data assets, distributes the data assets into the hierarchical classification directory of data assets, and also makes document classification marks after analyzing the content of the data assets. After marking the importance levels of some data assets, using the correlation formula and asset score formula between data assets, the importance levels of other data assets can be automatically calculated, realizing the precise classification and reasonable grading of construction project data assets, and improving the management efficiency and use value of data assets. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 It is a flowchart of the intelligent classification and grading method of construction project data assets in an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0030] The following further details the intelligent classification and grading method of construction project data assets provided by the present invention in conjunction with the accompanying drawings and specific embodiments. In combination with the following description, the advantages and features of the present invention will be clearer. It should be noted that the drawings are all in a very simplified form and use non-precise scales, only for the purpose of facilitating and clearly assisting in explaining the purpose of the embodiments of the present invention.

[0031] This embodiment provides an intelligent classification and grading method of construction project data assets, which specifically includes the following steps:

[0032] Step 1: Collect construction project-related data assets, organize the data assets, and construct a hierarchical classification directory of construction project data assets.

[0033] The collected construction project-related data assets include design drawings, construction plans, bill of quantities, on-site photos, etc. The collected data assets are preprocessed, such as cleaning, de-duplication, and format unification.

[0034] The hierarchical classification directory of data assets can be set in advance, and the data assets are divided into L layers, and each layer is further divided into several categories. For example, the first layer includes 3 categories, the second layer includes 4 categories, and the 4 categories of the second layer are obtained by further dividing each of the 3 categories of the first layer into 4 categories.

[0035] It is necessary to distribute the collected data assets into the hierarchical classification directory of data assets, and an intelligent distribution can be carried out by using a hierarchical clustering algorithm.

[0036] In a specific embodiment, the hierarchical classification directory of construction project data assets includes six layers, specifically as follows:

[0037] The first layer: Classify according to the data generation stage. For example, it can be divided into: planning stage, design stage, bidding stage, construction stage, completion stage.

[0038] The second layer: Classify according to the data source. For example, it can be divided into: provided by the employer, provided by the design unit, provided by the construction unit, provided by the supervision unit, provided by the third-party testing agency.

[0039] The third layer: Classify according to the data usage. For example, it can be divided into: design documents, construction planning, cost control, quality and safety management, project site management.

[0040] The fourth layer: Classify according to the data nature. For example, it can be divided into: text, drawings, images, databases.

[0041] The fifth layer: Classify according to the application scenario. For example, it can be divided into: design scenario, construction scenario, supervision scenario, employer scenario.

[0042] The sixth layer: Classify according to the business requirements. For example, it can be divided into: engineering entity requirements, safety requirements, environmental protection requirements.

[0043] It should be noted that the specific number of layers of the building engineering data assets and the specific categories of each layer can be set as needed.

[0044] Step 2: Analyze the content of the data assets and make document classification marks.

[0045] The document classification mark is to mark a specific data asset based on its content, that is, to set several labels according to the identified content. These labels are equal in status and do not involve the upper and lower hierarchical relationships. The main function is to determine the number of matching marks between two data assets in Step 3. mFor example, for text - type data assets, large - language models can be used to perform text analysis on engineering documents and automatically classify and mark them; for picture - type data assets, computer vision technology can be used to identify and analyze engineering pictures and automatically classify engineering elements in the pictures. As an example, mark as text, image, drawing, database, etc. according to the content and format of the document; perform content recognition on text data and mark the stage classification according to the construction stage involved in the document content, such as the planning stage, design stage, bidding stage, construction stage, completion stage, etc.; mark the special - topic classification according to the professional field of the document content, and the special - topic classification is marked as structural engineering, mechanical and electrical installation, decoration, landscape architecture, municipal engineering, etc.; through natural language processing technology, extract keywords in the document, and the keyword extraction is, for example, concrete, steel structure, energy - saving sub - project, etc. Use computer vision technology to identify elements such as buildings, equipment, and materials in the picture, and mark them as buildings, tower cranes, building materials (steel bars), etc. according to the engineering elements; mark the construction status classification as foundation construction, main - body construction, decoration construction, etc. according to the recognition results; mark the quality - problem classification as cracks, leaks, deformations, etc. according to the recognition results; mark the scene classification as construction site, material yard, office area, etc. according to the recognition results. The marking types can be the same as the names of the hierarchical classifications or have a corresponding relationship with the names of the hierarchical classifications, so as to facilitate the allocation of data assets to the hierarchical classification table.

[0046] Step 3: Classify data assets into core assets, important assets, and general assets according to the importance level, mark the importance levels of several data assets, and automatically calculate the importance levels of other data assets.

[0047] Mark all these data assets with marked importance levels as classified, and mark other data assets as unclassified. Specifically: according to the characteristics of construction projects, formulate discrimination rules for important assets, such as key processes, important parts, high - value assets, etc.; mark several data assets for each importance level.

[0048] Automatically calculate the importance levels of other data assets, including:

[0049] Step 3.1: Construct the relevance score P , Q formula for two data assets of a construction project D to determine P , Q whether they are related, where

[0050] ;

[0051] In the formula, L 、 k are the total number of layers and the k th layer in the hierarchical classification directory respectively;I k is a 0-1 function. If P and Q both belong to the k layer, then I k = 1; otherwise I k = 0; m is the number of matches between the P and Q document classification tags extracted in step 2. σ is a regulation coefficient, and preferably σ = 2.

[0052] As an example, P and Q are classified the same in the first to fourth layers, but different in the fifth layer. Naturally, they are also different in the sixth layer, and P and Q have 11 identical tags and are matched. Thus:

[0053] D = 1×2 1 + 1×2 2 + 1×2 3 + 1×2 4 + 0×2 5 + 0×2 6 + 2×11 = 52.

[0054] Set a threshold D 0. If D > D 0, then it is determined that the two assets are associated. D 0 can be preset according to experience. In a specific embodiment, D 0 is obtained in the following way: The total number of data assets is N. All data assets are randomly paired in twos, and the correlation scores are calculated. All the correlation scores are sorted, and the 90th percentile is statistically taken as the threshold D 0.

[0055] Step 3.2: Determine the association between the data assets with unmarked importance levels and the data assets with marked importance levels; calculate the asset score S , where S = 100 a + 10 b + c , in the formula, a , b , c are the quantities of the unmarked importance assets associated with the core assets, important assets, and general assets respectively.

[0056] Step 3.3: According to the asset score SSort from largest to smallest, and set the top 5% of the scores as core assets, 5% to 20% as important assets, and those after 20% as general assets.

[0057] If a new data asset enters the classification system, set the asset as "unclassified" and determine the importance level of the data asset according to steps 3.2 and 3.3.

[0058] In this embodiment, the importance levels of data assets are classified by controlling the percentage of quantity. Through steps 3.2 and 3.3, the unclassified data assets enter the classified ones. At this time, the total quantity of the classified data assets changes, and the asset scores of all data assets can be recalculated and then re-sorted to obtain the importance levels of each data asset again.

[0059] The technical features of the above-described embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope described in this specification.

[0060] The above-described embodiments only represent several implementation manners of the present invention, and their descriptions are relatively specific and detailed, but they should not be construed as limiting the scope of the invention. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present invention, several modifications and improvements can still be made, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the present invention should be subject to the appended claims.

Claims

1. An intelligent classification and grading method for building engineering data assets, characterized in that Including: Step 1: Collect relevant data assets of construction projects, organize the data assets, and construct a hierarchical classification directory for construction project data assets; Step 2: Analyze the content of the data assets and make document classification marks; among them, for text-based data assets, use large language models for text analysis and automatically make document classification marks; for picture-based data assets, use computer vision technology to identify and analyze engineering pictures, automatically classify engineering elements in the pictures, and make document classification marks; Step 3: Classify the data assets into core assets, important assets, and general assets according to the importance level, mark the importance levels of several data assets, and automatically calculate the importance levels of other data assets, including: Step 3.1: Construct the relevance scores of any two data assets in a construction project P , Q Relevance score D Calculation formula to determine P , Q Whether they are related, where ； Wherein, L and k are the total number of layers and the k -th layer in the hierarchical classification directory respectively; I k takes a value of 0 or 1. If P and Q both belong to the k -th layer, then I k = 1; otherwise I k = 0; m is the number of matches of the P and Q document classification tags extracted in the second step, and σ is the adjustment coefficient; Set a threshold D 0, if D > D 0, then it is determined that the data asset P , Q is associated, otherwise, it is determined that the data asset P , Q is not associated; Step 3.2: Determine the relevance between data assets with unmarked importance levels and data assets with marked importance levels; calculate the asset score S , where S = 100 a + 10 b + c , in the formula a , b , c are the quantities of unmarked importance assets associated with core assets, important assets, and general assets respectively; Step 3.3: Sort according to the asset scores S in descending order, and mark the importance levels of the data assets according to the set rules.

2. The intelligent classification and grading method for construction project data assets according to claim 1, characterized in that In step 3.3, control the importance level of data assets according to the ratio, and sort the asset scores S from large to small. After that, mark the data assets with the top C 1% of the scores as core assets, and mark the data assets with the scores from C 1% to C 2% as important assets. C Mark the data assets after 2% as general assets. C 1. C 2. C 3 are preset values, and 0 < C 1 < C 2 < 100.

3. The intelligent classification and grading method for construction project data assets according to claim 2, characterized in that C 1=5， C 2=20。 4. The intelligent classification and grading method for construction project data assets according to claim 1, characterized in that The hierarchical classification directory of construction project data assets includes six layers, specifically: The first layer: Classify according to the data generation stage; The second layer: Classify according to the data source; The third layer: Classify according to the data usage; The fourth layer: Classify according to the data nature; The fifth layer: Classify according to the application scenario; The sixth layer: Classify according to the business requirements.

5. The intelligent classification and grading method for construction project data assets according to claim 4, characterized in that Adopt a hierarchical clustering algorithm to intelligently allocate the data assets collected in Step 1 into the hierarchical classification directory of construction project data assets.

6. The intelligent classification and grading method for construction project data assets according to claim 1, characterized in that σ = 2.

7. The intelligent classification and grading method for construction project data assets according to claim 1, characterized in that In step 3.1, D 0 is obtained in the following way: The total number of data assets is N. All data assets are randomly paired in twos, and the correlation scores are calculated. All the correlation scores are sorted, and the 90th percentile is statistically taken as the threshold D 0.

Images