Processing method and device for project features in bill of quantity, medium and equipment

Abstract

The invention discloses a processing method and device for project features in an engineering quantity list, a medium and electronic equipment, and the method comprises the steps: obtaining a to-be-processed target engineering quantity list, and determining a project feature group associated with the target engineering quantity list; wherein the project feature group is associated with a job type; determining a keyword template corresponding to the job type; wherein the keyword template is a data structure defined with keyword extraction rules under a specific job type; the keyword template comprises a field list and a format specification; performing keyword extraction on the project feature group based on a keyword template to obtain feature keywords corresponding to the project feature group; and determining a target cost item corresponding to the target project quantity list based on the project feature group and the feature keyword corresponding to the project feature group. According to the invention, key information extraction is carried out on the project features in the bill of quantity, parameter integrity, structure normalization and information accuracy can be ensured, and the accuracy of cost item determination can be improved.

Description

Technical Field

[0001] This application relates to the fields of construction project management and construction project data processing, and in particular to a method, apparatus, medium and electronic equipment for processing project characteristics in a bill of quantities. Background Technology

[0002] In the field of engineering cost management, the key to accurately generating cost items corresponding to the bill of quantities lies in the effective extraction of key information from project characteristics. Project characteristics, as the technical description carrier of bill of quantities items, directly determine the constituent elements and parameters of cost items such as labor, materials, and machinery.

[0003] Currently, the industry commonly uses information extraction methods based on traditional natural language processing techniques, such as entity extraction and syntactic analysis, to attempt to extract key information from the project characteristics of the bill of quantities. However, these information extraction methods suffer from problems such as missing parameter details, inconsistent formats, and insufficient information accuracy when processing bills of quantities, which severely restricts the accuracy and reliability of subsequent cost item determination. Summary of the Invention

[0004] This application provides a method, apparatus, medium, and electronic equipment for processing project characteristics in a bill of quantities, which can achieve the purpose of ensuring parameter integrity, structural standardization, and information accuracy, and improving the accuracy of cost item determination.

[0005] According to a first aspect of this application, a method for processing item characteristics in a bill of quantities is provided, the method comprising:

[0006] Obtain the target bill of quantities to be processed, and determine the project feature group associated with the target bill of quantities; wherein, the project feature group is associated with a work type;

[0007] Determine the keyword template corresponding to the job type; wherein, the keyword template is a data structure that defines keyword extraction rules for a specific job type; the keyword template includes a field list and format specifications;

[0008] Based on the keyword template, keywords are extracted from the project feature group to obtain the feature keywords corresponding to the project feature group;

[0009] Based on the project feature group and the feature keywords corresponding to the project feature group, the target cost item corresponding to the target bill of quantities is determined.

[0010] According to a second aspect of this application, an apparatus for processing item characteristics in a bill of quantities is provided, the apparatus comprising:

[0011] The job type determination module is used to obtain the target quantity list to be processed and determine the project feature group associated with the target quantity list; wherein, the project feature group is associated with the job type;

[0012] The keyword template determination module is used to determine the keyword template corresponding to the job type; wherein, the keyword template is a data structure that defines keyword extraction rules under a specific job type; the keyword template includes a field list and a format specification;

[0013] The keyword extraction module is used to extract keywords from the project feature group based on the keyword template, so as to obtain the feature keywords corresponding to the project feature group.

[0014] The cost item determination module is used to determine the target cost item corresponding to the target bill of quantities based on the project feature group and the feature keywords corresponding to the project feature group.

[0015] According to a third aspect of the present invention, embodiments of this application provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method for processing item features in a bill of quantities as described in embodiments of this application.

[0016] According to a fourth aspect of the present invention, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements a method for processing item features in a bill of quantities as described in the embodiments of the present application.

[0017] According to a fifth aspect of this application, an embodiment of this application provides a computer program product, including a computer program that, when executed by a processor, implements a method for processing item features in a bill of quantities as described in an embodiment of this application.

[0018] The technical solution of this application, through predefined keyword templates deeply associated with job types, utilizes the field list set in the keyword templates to forcibly cover key parameters of interest for specific job types, thus avoiding parameter omission issues from the source and ensuring that feature keywords can fully cover the underlying parameters required for cost determination, improving the parameter completeness of keyword extraction. The format specifications set in the keyword templates are used to format the extracted feature keywords, eliminating ambiguity and inconsistency caused by free text descriptions and improving the structural standardization of keyword extraction. A precise mapping from project feature groups to keyword templates is established based on job types, ensuring that the extraction and organization of feature keywords are always under control, guaranteeing the accuracy and business relevance of feature keywords. Combining complete, standardized, and accurate feature keywords with project feature groups to determine the target cost items corresponding to the target bill of quantities helps improve the accuracy of cost item determination.

[0019] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of this application, nor is it intended to limit the scope of this application. Other features of this application will become readily apparent from the following description. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 This is a flowchart of the method for processing item characteristics in the bill of quantities provided in Example 1;

[0022] Figure 2 This is a flowchart of the method for processing item characteristics in the bill of quantities provided in Example 2;

[0023] Figure 3 This is a schematic diagram of the structure of the processing device for item features in the bill of quantities provided in Embodiment 3 of this application;

[0024] Figure 4 This is a schematic diagram of the structure of an electronic device provided in Embodiment 4 of this application. Detailed Implementation

[0025] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0026] It should be noted that the terms "first," "second," "target," and "candidate," etc., used in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0027] Example 1

[0028] Figure 1 This is a flowchart of a method for processing project features in a bill of quantities according to Embodiment 1. This embodiment is applicable to extracting keywords from project features in a bill of quantities and using the extracted keywords to determine cost items. This method can be executed by a device for processing project features in a bill of quantities. The device for processing project features in a bill of quantities is implemented in hardware and / or software and can be integrated into an electronic device running this system.

[0029] like Figure 1 As shown, the method includes:

[0030] S110. Obtain the target quantity list to be processed and determine the project feature group associated with the target quantity list; wherein, the project feature group is associated with the job type.

[0031] S120. Determine the keyword template corresponding to the job type; wherein, the keyword template is a data structure that defines keyword extraction rules for a specific job type; the keyword template includes a field list and format specifications.

[0032] S130. Extract keywords from the project feature group based on the keyword template to obtain the feature keywords corresponding to the project feature group.

[0033] S140. Based on the project feature group and the feature keywords corresponding to the project feature group, determine the target cost item corresponding to the target bill of quantities.

[0034] The target bill of quantities includes project feature text, which contains at least two project features. Project feature groups are obtained by grouping the project features in the project feature text. A project feature group is a combination of features with complete semantics under a specific job type.

[0035] Among these, project characteristics are associated with work types. Work types belong to the classification of the construction engineering field. Work types can cover engineering sub-items, construction measures, material and process attributes, and auxiliary information such as bill of quantities descriptions involved in the construction engineering field. The specific content of work types is not limited here and is determined according to actual business needs. For example, work types include the following classifications: {Location, Transport Distance, Plain Cement, Block Materials, Formwork, Pile Foundation and Support, Pricing, Subbase, Concrete Engineering, Surface Materials, Ceiling Engineering, Reinforcing Steel Engineering, Base Cleaning, Work Content, Painting Engineering, Masonry Engineering, Unit Price Measures, Insulation, Scaffolding, Painting, Earthwork Engineering, Leveling, Drainage and Ventilation, Keel, Waterproofing, Skirting Board, Metal Structure Engineering}.

[0036] The job type is used to determine the keyword template required for keyword extraction from project feature groups. A keyword template is a data structure that defines keyword extraction rules for a specific job type; it includes a field list and a format specification. The field list specifies the key parameters to be extracted, and the format specification clarifies the data standardization requirements. The engineering description text in project feature groups is generally unstructured. Based on the keyword template, structured key parameters can be extracted from the unstructured engineering description of the project feature group.

[0037] Keyword templates are deeply tied to job types, exhibiting specificity across different job types. This is because different job types focus on entirely different key parameters, and their corresponding formatting specifications also differ.

[0038] Feature keywords are obtained by extracting keywords from project feature groups based on keyword templates. Feature keywords include key parameters that are of interest to specific job types and conform to the format specifications required by specific job types.

[0039] The specific content of the keyword template is determined based on actual business needs and is not limited here. For example, the target bill of quantities is associated with two project feature groups: Project Feature Group A "Roof 1 - Waterproofing membrane, Location: Non-accessible flat roof, 3.0mm thick self-adhesive polyester-modified bitumen waterproofing membrane" and Project Feature Group B "Ground 1 Insulation, 50mm thick flame-retardant extruded polystyrene board insulation, laid within 2m inwards along the exterior wall". The work type for Project Feature Group A is waterproofing, and the work type for Project Feature Group B is insulation. The keyword template for waterproofing could be "Work=?; Thickness=?; Material=?". The keyword template for insulation could be "Work=?; Thickness=?; Material=?; Location=?". Based on the keyword template, the feature keywords extracted from Project Feature Group A could be "Work=Waterproofing membrane; Thickness=3.0; Material=Self-adhesive polyester-modified bitumen". Based on the keyword template, the feature keywords extracted from Project Feature Group B could be "Work=Insulation; Thickness=50; Material=Extruded polystyrene board; Location=Ground 1". Understandably, the keyword templates for waterproofing and thermal insulation can also constrain the unit of thickness.

[0040] The feature keywords corresponding to the project feature group serve as the basis for determining cost items. They can be used together with the project feature group to determine the target cost items corresponding to the target bill of quantities. This approach retains the efficiency advantages of direct feature comparison while using keyword matching to compensate for the identification blind spots caused by differences in feature descriptions, thereby improving the success rate of cost item determination.

[0041] The expense items translate the work items in the bill of quantities into actual costs and prices. The specific content of the expense items is determined based on actual business needs and is not limited here. For example, expense items may include material costs and labor subcontracting fees.

[0042] The technical solution of this application, through predefined keyword templates deeply associated with job types, utilizes the field list set in the keyword templates to forcibly cover key parameters of interest for specific job types, thus avoiding parameter omission issues from the source and ensuring that feature keywords can fully cover the underlying parameters required for cost determination, improving the parameter completeness of keyword extraction. The format specifications set in the keyword templates are used to format the extracted feature keywords, eliminating ambiguity and inconsistency caused by free text descriptions and improving the structural standardization of keyword extraction. A precise mapping from project feature groups to keyword templates is established based on job types, ensuring that the extraction and organization of feature keywords are always under control, guaranteeing the accuracy and business relevance of feature keywords. Combining complete, standardized, and accurate feature keywords with project feature groups to determine the target cost items corresponding to the target bill of quantities helps improve the accuracy of cost item determination.

[0043] In an optional embodiment, determining the target cost item corresponding to the target bill of quantities based on the project feature group and the feature keywords corresponding to the project feature group includes: obtaining a first correspondence between preset project features and candidate cost items, and the feature keywords corresponding to the preset project features; performing similarity matching between the project feature group and the preset project features to obtain a first matching result; if the first matching result is a matching failure, performing similarity matching between the feature keywords corresponding to the project feature group and the feature keywords corresponding to the preset project features to obtain a second matching result; if the second matching result is a matching success, determining the preset project feature that matches the project feature group as the target project feature; and determining the target cost item corresponding to the project feature group from the candidate cost items based on the first correspondence and the target project feature.

[0044] The first correspondence record records the association between project characteristics and candidate cost items. The first correspondence is predefined according to business requirements, and this application supports the expansion and adjustment of the first correspondence.

[0045] The first matching result is determined based on the text similarity between the project feature group and the preset project features. Text similarity can adapt to natural language variations in feature descriptions and can handle new or complex project feature expressions. Optionally, the project feature group and the preset project features are vectorized separately to obtain a first feature vector and a second feature vector. The vector similarity between the first feature vector and the second feature vector is calculated as the text similarity between the project feature group and the preset project features.

[0046] Optionally, the text similarity is compared with a text similarity threshold. If the text similarity is greater than the threshold, the first matching result is determined to be a successful match; if the text similarity is less than or equal to the threshold, the second matching result is determined to be a failed match. Optionally, the preset project features that were successfully matched in the first matching result are used as target project features. Based on the first correspondence, a target cost item corresponding to the target project feature can be determined from the candidate cost items. The target cost item is then determined as the target cost item corresponding to the project feature group.

[0047] If the first matching result fails, it means that the target cost item cannot be determined directly using the project feature group. It is necessary to combine the feature keywords corresponding to the project feature group and the feature keywords corresponding to the preset project features to determine it.

[0048] The second matching result is determined based on the feature similarity between the feature keywords corresponding to the project feature group and the feature keywords corresponding to the preset project features. Optionally, the feature similarity is compared with a feature similarity threshold. If the feature similarity is greater than the feature similarity threshold, the second matching result is determined to be a successful match; if the feature similarity is less than or equal to the feature similarity threshold, the second matching result is determined to be a failed match.

[0049] Optionally, the preset project features that are successfully matched in the second matching result are used as the target project features. Based on the first correspondence, the target cost item corresponding to the target project feature can be determined from the candidate cost items. The target cost item is then determined as the target cost item corresponding to the project feature group.

[0050] The above technical solution establishes a primary correspondence between preset project features and candidate expense items, and introduces feature keywords as auxiliary matching criteria, constructing a two-layer intelligent matching mechanism. When a project feature group fails to directly match the preset features, it automatically switches to secondary matching at the feature keyword level, significantly improving the accuracy of expense item identification in complex business scenarios. It retains the efficiency advantages of direct feature comparison while compensating for recognition blind spots caused by differences in feature descriptions through keyword semantic matching, demonstrating strong robustness, especially when dealing with variations in industry terminology, multilingual scenarios, or non-standardized feature descriptions. While ensuring matching accuracy, it also exhibits good adaptability to changes in business requirements.

[0051] In an optional embodiment, determining the project feature group associated with the target bill of quantities includes: splitting the project feature text of the target bill of quantities to obtain at least two candidate project features to be grouped; determining the work type and importance level corresponding to the candidate project features; wherein the importance level is determined based on whether the candidate project features can independently explain the cost items involved in the target bill of quantities; and grouping the candidate project features based on the work type and the importance level to obtain the project feature group corresponding to the target bill of quantities.

[0052] The target bill of quantities refers to the bill of quantities that needs to be broken down into project features. The target bill of quantities includes project feature text. Optionally, project feature text can be extracted from the bill of quantities to be broken down based on content identifiers. The content identifier corresponding to the project feature text in the target bill of quantities can be "project feature". Candidate project features are obtained by breaking down the project feature text. Candidate project features are the objects of feature grouping, and job type and importance level are important bases for grouping candidate project features. Job type belongs to the classification of the construction engineering field. Job type can cover engineering sub-items, construction measures, material and process attributes, and auxiliary information in the bill of quantities description, etc., involved in the construction engineering field. Job type is used to group candidate project features; optionally, candidate project features belonging to the same job type are grouped into the same feature group to ensure the semantic integrity of the project feature group under a specific job type.

[0053] The specific content of the work type is not limited here, and will be determined according to actual business needs. For example, the work types include the following categories: {Location, Transport Distance, Plain Cement, Block Materials, Formwork, Pile Foundation and Support, Pricing, Subbase, Concrete Engineering, Surface Materials, Ceiling Engineering, Reinforcing Steel Engineering, Base Cleaning, Work Content, Painting Engineering, Masonry Engineering, Unit Price Measures, Insulation, Scaffolding, Painting, Earthwork Engineering, Leveling, Drainage and Ventilation, Keel, Waterproofing, Skirting Board, Metal Structure Engineering}.

[0054] The importance level is determined based on whether the candidate project characteristics can independently describe the cost items involved in the target bill of quantities, and is used to quantify the independence of project characteristic groups in the cost item descriptions. Optionally, the importance levels include a first level and a second level. The first level is higher than the second level. Candidate project characteristics of the first level can independently describe the cost items involved in the target bill of quantities; candidate project characteristics of the second level require a combination of descriptions to describe the cost items involved in the target bill of quantities.

[0055] Optionally, candidate project features at the first level can be used as independent feature groups. Candidate project features at the second level need to be combined to form combined feature groups. The job type guides how to combine candidate project features. Each project feature group must include at least basic project features. Basic project features refer to candidate project features whose job type belongs to a basic type. Job types related to the bill of quantities description auxiliary information can be identified as basic types. Optionally, basic types include locations. The bill of quantities description auxiliary information essentially describes non-physical, non-process-related auxiliary information in the bill of quantities. Optionally, basic project features are added to both the independent feature groups and the combined feature groups to obtain project feature groups.

[0056] The above technical solution, by splitting the project feature text in the target bill of quantities, achieves item-by-item parsing of the project feature text, transforming the originally mixed project feature text into candidate project features with appropriate granularity, laying the foundation for subsequent analysis. Considering the work type when grouping candidate project features ensures the semantic integrity of project feature groups under specific work types. Considering the importance level when grouping candidate project features, by judging whether the candidate feature can independently explain the cost item corresponding to the bill of quantities to be split, the grouping of candidate project features is closely linked to the cost item determination requirements. Through the collaborative grouping mechanism of work type and importance level, a logically rigorous project feature group is constructed, which not only conforms to the professional classification habits in the construction engineering field but also clearly presents the decision-making path for cost item determination. Using project feature groups to determine the project feature group corresponding to the target bill of quantities helps improve the accuracy of cost item determination.

[0057] Example 2

[0058] Figure 2 This is a flowchart illustrating the processing method for project characteristics in the bill of quantities provided in Example 2. This example further optimizes the above examples.

[0059] like Figure 2 As shown, the method includes:

[0060] S210. Obtain the target quantity list to be processed and determine the project feature group associated with the target quantity list; wherein, the project feature group is associated with a work type.

[0061] S220. Determine the keyword template corresponding to the job type; wherein, the keyword template is a data structure that defines keyword extraction rules for a specific job type; the keyword template includes a field list and format specifications.

[0062] S230. Based on the project feature group and the keyword template corresponding to the project feature group, generate keyword extraction prompts for the keyword extraction model.

[0063] The keyword extraction prompts guide the keyword extraction model to extract feature keywords from the project feature group according to the keyword template. Optionally, the keyword extraction prompts include the project feature group and the corresponding keyword template.

[0064] The keyword extraction model is obtained by fine-tuning the large language model based on the labeled first sample; the first sample is the item feature group, and the sample label of the first sample is the feature keyword extracted from the first sample based on the keyword template.

[0065] Large Language Models (LLMs) are deep learning models trained on massive amounts of text data that can generate natural language text or understand the meaning of language text. LLMs can handle various natural language tasks, such as text classification, question answering, and dialogue. Through training, LLMs capture knowledge from large amounts of labeled and unlabeled data and store this knowledge in a vast number of parameters, which can reach tens or hundreds of billions.

[0066] The first sample and its label are determined based on actual business requirements and are not limited here. For example, the first sample could be “thermal insulation wall surface; waterproof concrete roll waterproof exterior wall: 30mm thick XPS foam board protective layer bonded with construction adhesive”, and the first sample label could be “Operation = thermal insulation; thickness = 30mm; material = XPS foam board; location = exterior wall”.

[0067] S240. Using the keyword extraction model and the keyword extraction prompts, keywords are extracted from the project feature group to obtain the feature keywords corresponding to the project feature group.

[0068] Input the keyword extraction prompts into the keyword extraction model. Based on the keyword templates provided in the keyword extraction prompts, the keyword extraction model extracts keywords from the project feature groups and outputs the feature keywords corresponding to the project feature groups.

[0069] Keyword templates in keyword extraction prompts can constrain the keyword extraction model and prevent terminology deviations caused by the model's free interpretation.

[0070] S250. Based on the project feature group and the feature keywords corresponding to the project feature group, determine the target cost item corresponding to the target bill of quantities.

[0071] The technical solution of this application utilizes predefined keyword templates as extraction standards, providing a clear semantic framework and constraints for the fine-tuned large language model. This effectively avoids deviations or redundancy caused by free model generation, ensuring that the extracted feature keywords closely match business needs and maintain a consistent format. Simultaneously, the large language model is fine-tuned based on labeled sample pairs (project feature groups) and feature keywords extracted according to the keyword templates. This allows the model to deeply internalize the keyword extraction rules and expression habits of specific domains or projects, significantly improving the model's semantic understanding of complex feature descriptions and the accuracy of key information location, while reducing reliance on manual rule writing or post-processing. This application, through a combination of structured template guidance and targeted model fine-tuning, significantly improves the accuracy, consistency, and efficiency of keyword extraction from project feature groups. It also reduces the cost of manual intervention and increases processing speed.

[0072] In an optional embodiment, generating keyword extraction prompts for the keyword extraction model based on the project feature group and the keyword template corresponding to the project feature group includes: obtaining keyword records; wherein the keyword records include at least two sample feature groups and feature keywords extracted from the sample feature groups based on the keyword template; determining the text similarity between the project feature group and the sample feature group, and determining a target feature group from the sample feature group based on the text similarity; generating auxiliary prompts based on the target feature group and the feature keywords corresponding to the target feature group; generating requirement prompts based on the project feature group and the keyword template corresponding to the project feature group; and generating keyword extraction prompts for the keyword extraction model based on the auxiliary prompts and the requirement prompts.

[0073] The keyword record refers to a pre-stored structured example library, determined based on the keyword extraction history of the keyword extraction model. Each keyword record contains multiple sets of example feature groups and corresponding feature keywords. The feature keywords corresponding to the example feature groups are extracted from the example feature groups using the keyword extraction model based on keyword templates.

[0074] The target feature group is selected from the keyword records by calculating text similarity to find the sample feature group with the highest similarity to the current project feature group. By calculating text similarity, the sample feature group most relevant to the current project features is retrieved from the keyword records, ensuring the domain relevance of the auxiliary prompts and avoiding excessive deviation between the examples and the current task.

[0075] The auxiliary prompts are generated based on the target feature group and the corresponding feature keywords. The auxiliary prompts are used to provide a reference paradigm for the keyword extraction model, clarify the keyword format and extraction logic, and help reduce model ambiguity.

[0076] The requirement prompts are based on project feature groups and corresponding keyword templates, and are used to help the keyword extraction model clarify the content to be processed and the keyword extraction requirements.

[0077] Keyword extraction prompts are composed of auxiliary prompts and requirement prompts. Examples of auxiliary prompts are provided, and requirement prompts correspond to the current task.

[0078] The above technical solution, by calculating the text similarity between project feature groups and example feature groups, can automatically select the most relevant example feature groups as references, ensuring that the examples provided by the auxiliary prompts are highly adapted to the current task. This effectively solves the model understanding bias problem caused by the blind selection of examples in traditional methods. Simultaneously, by providing business rule templates through auxiliary prompts, and clearly defining the current task with requirement prompts, the keyword extraction model can quickly understand the extraction logic and format requirements of specific keyword templates without retraining, significantly reducing the reliance on repeated fine-tuning. This not only enhances the generalization ability of the keyword extraction model to complex business scenarios and enables logical transfer through similar examples, but also significantly reduces the cost of manually customizing rules or correcting outputs.

[0079] In an optional embodiment, before determining the target cost item corresponding to the target bill of quantities based on the project feature group and the feature keywords corresponding to the project feature group, the method further includes: obtaining a correction request for the feature keywords; correcting the feature keywords corresponding to the project feature group based on the correction request and / or the job type corresponding to the feature keywords to obtain new keywords; and updating the feature keywords corresponding to the project feature group using the new keywords.

[0080] Since feature keywords are extracted from project feature groups, free textual expressions within some job types' project feature groups can lead to non-standardized expressions in the feature keywords for these job types. This mainly manifests as multiple different expressions being used for the same entity or process. To eliminate these expression differences, after determining the feature keywords corresponding to the project feature groups, we determine whether there are non-standardized expressions in the feature keywords based on the job types corresponding to the project feature groups, and correct any feature keywords with non-standardized expressions.

[0081] The feature keyword correction request is used to request corrections to specific keywords within the feature keywords. The correction request is related to the actual business logic, and its specific content is not limited here.

[0082] The new keywords may be obtained by revising specific keywords within the feature keywords based on a revision request, thus ensuring the new keywords align with the actual business logic. Alternatively, they may be obtained by revising feature keywords corresponding to certain job types, resulting in standardized feature keywords. Optionally, specific keywords within the feature keywords may be replaced with the new keywords.

[0083] The above technical solution provides a practical keyword correction scheme, which not only supports the correction of feature keywords corresponding to specific job types, but also supports the correction of specific keywords in feature keywords based on actual business logic through correction requests, thereby improving the accuracy of feature keywords.

[0084] In an optional embodiment, based on the job type corresponding to the feature keyword, the feature keyword corresponding to the project feature group is modified to obtain a new keyword, including: if the job type corresponding to the feature keyword belongs to a preset type, then determining whether the feature keyword is a non-standard keyword based on the keyword set corresponding to the preset type; if the feature keyword is a non-standard keyword, then inputting the project feature group corresponding to the non-standard keyword into the keyword modification model corresponding to the preset type; outputting the standardized keyword corresponding to the project feature group through the keyword modification model, and using the standardized keyword as the new keyword; wherein, the keyword modification model is a multi-classification model pre-trained based on a labeled second sample; the second sample is the project feature group corresponding to the preset type, and the sample label of the second sample is the standardized keyword corresponding to the second sample.

[0085] The free textual expressions present in the project feature groups corresponding to the preset types mainly manifest as using multiple different expressions for the same entity or the same process. The feature keywords corresponding to the preset types may contain non-standardized expressions.

[0086] The keyword set corresponding to a preset type records the standardized expressions for that preset type. Based on whether the feature keyword corresponding to a preset type belongs to the keyword set corresponding to that preset type, it is determined whether the feature keyword corresponding to the preset type is a non-standard keyword. If the feature keyword corresponding to a preset type belongs to the keyword set corresponding to that preset type, it is determined to be a standard keyword. If the feature keyword corresponding to a preset type does not belong to the keyword set corresponding to that preset type, it is determined to be a non-standard keyword. Here, non-standard keywords are feature keywords expressed using non-standardized expressions.

[0087] Optionally, the default type is "location." In actual engineering scenarios, the project feature group containing "location" as the work type contains a large number of free text descriptions. Multiple different descriptions may all refer to the same location. For example, "interior wall" and "within thermal bridge" are different descriptions of the same location. The keyword set corresponding to the location is determined according to actual business needs and is not limited here. For example, the keyword set corresponding to the location can be {floor, roof, ceiling, interior wall, exterior wall, underground, above ground, wall surface}.

[0088] Since "inside the thermal bridge" does not belong to the keyword set corresponding to the location, it can be determined that "inside the thermal bridge" is a non-standard keyword, while "inner wall" is a standard keyword.

[0089] If the feature keyword corresponding to the preset type is determined to be a non-standard keyword, the keyword correction model corresponding to the preset type is used to correct the non-standard keyword, and the keyword correction model corrects the non-standard keyword to a standard keyword. Continuing with the previous example, the keyword correction model can correct the non-standard keyword "heat bridge inside" to the standard keyword "inner wall".

[0090] The keyword correction model is associated with a preset type and is used to correct non-standard keywords corresponding to the preset type. The project feature groups corresponding to the non-standard keywords are the input data of the keyword correction model, and the standardized keywords are the output results of the keyword correction model. The standardized keywords are feature keywords expressed in a non-standardized manner.

[0091] The keyword correction model is a multi-class classification model pre-trained based on labeled second samples. The second sample is a feature group of an item corresponding to a preset type, and the sample label of the second sample is the canonical keyword corresponding to the second sample. Optionally, the canonical keywords used as sample labels are generated from the keyword set corresponding to the preset type. Optionally, the keyword correction model is a BERT model (Bidirectional Encoder Representations from Transformers).

[0092] For example, the second sample is "Interior wall 1 mortar interior wall surface or Interior wall 2 waterproof brick interior wall surface; used for exterior wall interior surfaces of materials such as shear walls, thermal bridges, slabs and columns, etc., which are thermal bridges". The sample label for the second sample is "Interior wall".

[0093] The above technical solution identifies non-standard keywords corresponding to preset types based on a set of keywords. Then, using a pre-trained keyword correction model, it takes the project feature groups corresponding to the non-standard keywords as input and outputs standardized keywords. This preserves the original engineering intent of the non-standard keywords while forcibly converting them into standardized expressions, thus avoiding cost item matching deviations caused by differences in expression at the source and improving the accuracy of cost item determination.

[0094] Example 3

[0095] Figure 3 This is a schematic diagram of the structure of the processing device for project features in the bill of quantities provided in Embodiment 3 of this application. This embodiment can be applied to extracting keywords from project features in the bill of quantities and using the extracted keywords to determine cost items. The device can be implemented by software and / or hardware and can be integrated into electronic devices such as smart terminals.

[0096] like Figure 3 As shown, the device may include:

[0097] The job type determination module 310 is used to obtain the target quantity list to be processed and determine the project feature group associated with the target quantity list; wherein, the project feature group is associated with a job type;

[0098] The keyword template determination module 320 is used to determine a keyword template corresponding to the job type; wherein, the keyword template is a data structure that defines keyword extraction rules under a specific job type; the keyword template includes a field list and a format specification;

[0099] The keyword extraction module 330 is used to extract keywords from the project feature group based on the keyword template to obtain the feature keywords corresponding to the project feature group.

[0100] The cost item determination module 340 is used to determine the target cost item corresponding to the target bill of quantities based on the project feature group and the feature keywords corresponding to the project feature group.

[0101] The technical solution of this application, through predefined keyword templates deeply associated with job types, utilizes the field list set in the keyword templates to forcibly cover key parameters of interest for specific job types, thus avoiding parameter omission issues from the source and ensuring that feature keywords can fully cover the underlying parameters required for cost determination, improving the parameter completeness of keyword extraction. The format specifications set in the keyword templates are used to format the extracted feature keywords, eliminating ambiguity and inconsistency caused by free text descriptions and improving the structural standardization of keyword extraction. A precise mapping from project feature groups to keyword templates is established based on job types, ensuring that the extraction and organization of feature keywords are always under control, guaranteeing the accuracy and business relevance of feature keywords. Combining complete, standardized, and accurate feature keywords with project feature groups to determine the target cost items corresponding to the target bill of quantities helps improve the accuracy of cost item determination.

[0102] Optionally, the keyword extraction module 330 includes: a prompt word generation submodule, used to generate keyword extraction prompt words for the keyword extraction model based on the project feature group and the keyword template corresponding to the project feature group; and a keyword extraction submodule, used to extract keywords from the project feature group using the keyword extraction prompt words through the keyword extraction model to obtain the feature keywords corresponding to the project feature group; wherein, the keyword extraction model is obtained by fine-tuning the large language model based on a labeled first sample; the first sample is the project feature group, and the sample label of the first sample is the feature keyword extracted from the first sample based on the keyword template.

[0103] Optionally, the prompt word generation submodule includes: a record acquisition unit, used to acquire keyword records; wherein the keyword records include at least two sample feature groups and feature keywords extracted from the sample feature groups based on the keyword template; a target feature group determination unit, used to determine the text similarity between the project feature group and the sample feature group, and determine the target feature group from the sample feature group based on the text similarity; an auxiliary prompt word generation unit, used to generate auxiliary prompt words based on the target feature group and the feature keywords corresponding to the target feature group; a requirement prompt word generation unit, used to generate requirement prompt words based on the project feature group and the keyword template corresponding to the project feature group; and a final prompt word generation unit, used to generate keyword extraction prompt words for the keyword extraction model based on the auxiliary prompt words and the requirement prompt words.

[0104] Optionally, the apparatus further includes: a correction request acquisition module, configured to acquire a correction request for the feature keywords before determining the target cost item corresponding to the target bill of quantities based on the project feature group and the feature keywords corresponding to the project feature group; a keyword correction module, configured to correct the feature keywords corresponding to the project feature group based on the correction request and / or the job type corresponding to the feature keywords to obtain new keywords; and a keyword update module, configured to update the feature keywords corresponding to the project feature group using the new keywords.

[0105] Optionally, the keyword correction module includes: a non-standard keyword determination submodule, used to determine whether the feature keyword is a non-standard keyword based on the keyword set corresponding to the preset type if the job type corresponding to the feature keyword belongs to a preset type; a non-standard keyword input submodule, used to input the project feature group corresponding to the non-standard keyword into the keyword correction model corresponding to the preset type if the feature keyword is a non-standard keyword; and a standardized keyword output submodule, used to output the standardized keyword corresponding to the project feature group through the keyword correction model, and use the standardized keyword as the new keyword; wherein, the keyword correction model is a multi-classification model pre-trained based on a labeled second sample; the second sample is the project feature group corresponding to the preset type, and the sample label of the second sample is the standardized keyword corresponding to the second sample.

[0106] Optionally, the cost item determination module 340 includes: a correspondence acquisition submodule, used to acquire a first correspondence between preset project features and candidate cost items, and feature keywords corresponding to the preset project features; a first matching result determination submodule, used to perform similarity matching between the project feature group and the preset project features to obtain a first matching result; a second matching result determination submodule, used to perform similarity matching between the feature keywords corresponding to the project feature group and the feature keywords corresponding to the preset project features if the first matching result is a matching failure, to obtain a second matching result; a target project feature determination submodule, used to determine the preset project feature that matches the project feature group as a target project feature if the second matching result is a matching success; and a target cost item determination submodule, used to determine the target cost item corresponding to the project feature group from the candidate cost items based on the first correspondence and the target project feature.

[0107] Optionally, the job type determination module 310 includes: a project feature splitting submodule, used to split the project feature text of the target bill of quantities to obtain at least two candidate project features to be grouped; a feature attribute determination submodule, used to determine the job type and importance level corresponding to the candidate project features; wherein the importance level is determined based on whether the candidate project features can independently explain the cost items involved in the target bill of quantities; and a project feature grouping submodule, used to group the candidate project features based on the job type and the importance level to obtain the project feature group corresponding to the target bill of quantities.

[0108] The apparatus for processing item features in a bill of quantities provided in the embodiments of the invention can execute the processing method for item features in a bill of quantities provided in any embodiment of the present application, and has the corresponding performance modules and beneficial effects for executing the processing method for item features in a bill of quantities.

[0109] Example 4

[0110] According to embodiments of this application, this application also provides an electronic device, a readable storage medium, and a computer program product.

[0111] Figure 4 A schematic diagram of an electronic device 410, which can be implemented using an embodiment, is shown. The electronic device 410 includes at least one processor 411 and a memory, such as a read-only memory (ROM) 412, a random access memory (RAM) 413, etc., communicatively connected to the at least one processor 411. The memory stores computer programs executable by the at least one processor. The processor 411 can perform various appropriate actions and processes based on the computer program stored in the ROM 412 or loaded from storage unit 418 into the RAM 413. The RAM 413 may also store various programs and data required for the operation of the electronic device 410. The processor 411, ROM 412, and RAM 413 are interconnected via a bus 414. An input / output (I / O) interface 415 is also connected to the bus 414.

[0112] Multiple components in electronic device 410 are connected to I / O interface 415, including: input unit 416, such as keyboard, mouse, etc.; output unit 417, such as various types of displays, speakers, etc.; storage unit 418, such as disk, optical disk, etc.; and communication unit 419, such as network card, modem, wireless transceiver, etc. Communication unit 419 allows electronic device 410 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0113] Processor 411 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 411 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 411 performs the various methods and processes described above, such as the processing methods for project features in a bill of quantities.

[0114] In some embodiments, the method for processing item features in the bill of quantities may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 418. In some embodiments, part or all of the computer program may be loaded and / or installed on electronic device 410 via ROM 412 and / or communication unit 419. When the computer program is loaded into RAM 413 and executed by processor 411, one or more steps of the method for processing item features in the bill of quantities described above may be performed. Alternatively, in other embodiments, processor 411 may be configured to perform the method for processing item features in the bill of quantities by any other suitable means (e.g., by means of firmware).

[0115] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations may include: implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0116] Computer programs used to implement the methods of this application may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other processing device for the features of an item in a programmable bill of quantities, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0117] In the context of this application, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium can be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0118] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0119] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., processing servers that act as features of items in a bill of quantities), or computing systems that include middleware components (e.g., application servers), or computing systems that include frontend components (e.g., user computers with a graphical user interface or web browser through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., a communication network). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0120] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0121] This application also discloses a computer program product, which includes a computer program that, when executed by a processor, implements the method for processing item features in the bill of quantities provided in any embodiment of this application. This program product and the method for processing item features in the bill of quantities disclosed in the embodiments of this application belong to the same inventive concept, and therefore will not be described in detail here.

[0122] It should be understood that the various forms of processes shown above can be used to rearrange, add, or delete steps. For example, the steps described in this application can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this application can be achieved, and this is not limited herein.

[0123] The specific embodiments described above do not constitute a limitation on the scope of protection of this application. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for processing item characteristics in a bill of quantities, characterized in that, The method includes: Obtain the target bill of quantities to be processed, and determine the project feature group associated with the target bill of quantities; wherein, the project feature group is associated with a work type; Determine the keyword template corresponding to the job type; wherein, the keyword template is a data structure that defines keyword extraction rules for a specific job type; the keyword template includes a field list and format specifications; Based on the keyword template, keywords are extracted from the project feature group to obtain the feature keywords corresponding to the project feature group; Based on the project feature group and the feature keywords corresponding to the project feature group, the target cost item corresponding to the target bill of quantities is determined.

2. The method according to claim 1, characterized in that, The step of extracting keywords from the project feature group based on the keyword template to obtain the feature keywords corresponding to the project feature group includes: Based on the project feature group and the keyword template corresponding to the project feature group, generate keyword extraction prompts for the keyword extraction model; The keyword extraction model is used to extract keywords from the project feature group using the keyword extraction prompts, thereby obtaining the feature keywords corresponding to the project feature group. The keyword extraction model is obtained by fine-tuning the large language model based on the labeled first sample; the first sample is the project feature group, and the sample label of the first sample is the feature keyword extracted from the first sample based on the keyword template.

3. The method according to claim 2, characterized in that, The step of generating keyword extraction prompts for the keyword extraction model based on the project feature group and the corresponding keyword template includes: Obtain keyword records; wherein, the keyword records include at least two sample feature groups and feature keywords extracted from the sample feature groups based on the keyword template; Determine the text similarity between the project feature group and the sample feature group, and determine the target feature group from the sample feature group based on the text similarity; Based on the target feature group and the feature keywords corresponding to the target feature group, auxiliary prompt words are generated; Based on the project feature group and the keyword template corresponding to the project feature group, generate requirement prompt words; Based on the auxiliary prompts and the requirement prompts, keyword extraction prompts are generated for the keyword extraction model.

4. The method according to claim 1, characterized in that, Before determining the target cost item corresponding to the target bill of quantities based on the project feature group and the feature keywords corresponding to the project feature group, the method further includes: Obtain a correction request for the aforementioned feature keywords; Based on the correction request and / or the job type corresponding to the feature keyword, the feature keyword corresponding to the project feature group is corrected to obtain a new keyword; The new keywords are used to update the feature keywords corresponding to the project feature group.

5. The method according to claim 4, characterized in that, Based on the job type corresponding to the aforementioned feature keywords, the feature keywords corresponding to the project feature group are modified to obtain new keywords, including: If the job type corresponding to the feature keyword belongs to a preset type, then it is determined whether the feature keyword is a non-standard keyword based on the keyword set corresponding to the preset type; If the feature keyword is the non-standard keyword, then the project feature group corresponding to the non-standard keyword is input into the keyword correction model corresponding to the preset type; The keyword correction model outputs the standardized keywords corresponding to the project feature group, and the standardized keywords are used as the new keywords. The keyword correction model is a multi-classification model pre-trained based on a labeled second sample; the second sample is the item feature group corresponding to the preset type, and the sample label of the second sample is the standard keyword corresponding to the second sample.

6. The method according to claim 1, characterized in that, The step of determining the target cost item corresponding to the target bill of quantities based on the project feature group and the corresponding feature keywords includes: Obtain the first correspondence between preset project features and candidate cost items, as well as the feature keywords corresponding to the preset project features; The project feature group is matched with the preset project features to obtain a first matching result; If the first matching result is a failure, then the feature keywords corresponding to the project feature group and the feature keywords corresponding to the preset project features are matched for similarity to obtain a second matching result; If the second matching result is a successful match, then the preset project feature that matches the project feature group is determined as the target project feature; Based on the first correspondence and the target project characteristics, a target cost item corresponding to the project characteristic group is determined from the candidate cost items.

7. The method according to claim 1, characterized in that, The process of determining the project feature group associated with the target bill of quantities includes: The project feature text of the target bill of quantities is split to obtain at least two candidate project features to be grouped; Determine the job type and importance level corresponding to the candidate project characteristics; wherein, the importance level is determined based on whether the candidate project characteristics can independently explain the cost items involved in the target bill of quantities; Based on the job type and the importance level, the candidate project features are grouped to obtain the project feature group corresponding to the target bill of quantities.

8. A processing device for item characteristics in a bill of quantities, characterized in that, The device includes: The job type determination module is used to obtain the target quantity list to be processed and determine the project feature group associated with the target quantity list; wherein, the project feature group is associated with the job type; The keyword template determination module is used to determine the keyword template corresponding to the job type; wherein, the keyword template is a data structure that defines keyword extraction rules under a specific job type; the keyword template includes a field list and a format specification; The keyword extraction module is used to extract keywords from the project feature group based on the keyword template, so as to obtain the feature keywords corresponding to the project feature group. The cost item determination module is used to determine the target cost item corresponding to the target bill of quantities based on the project feature group and the feature keywords corresponding to the project feature group.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by the processor, the program implements the processing method for the item characteristics in the bill of quantities as described in any one of claims 1-7.

10. An electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method for processing the project characteristics in the bill of quantities as described in any one of claims 1-7.

Images