Data classification processing method and device based on bid invitation theme classification

Through the data classification processing method based on the bidding topic, the bidding data is identified and analyzed, and the problem of low data processing efficiency in large-scale bidding is solved, efficient data classification and review are achieved, and processing efficiency and accuracy are improved.

CN119046737AInactive Publication Date: 2024-11-29NINGXIA INTERNATIONAL TENDERING CONSULTING GROUP CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411195137.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-29
Publication Date
2024-11-29
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

In large-scale bidding, it is difficult for the existing technology to efficiently process a large amount of bidding data, resulting in low processing efficiency, high human resources consumption and easy data processing exceptions.

Method used

Adopt data classification processing methods based on bidding themes, and obtain the bidding company's corporate bidding documents and review standard information, identify demand information and condition information, conduct content division and large language model analysis, improve the accuracy and comprehensiveness of the review, and avoid manual review high cost and inefficiency.

Benefits of technology

It has improved the classification processing efficiency and audit efficiency of bidding data, improved the accuracy and comprehensiveness of audits, and comprehensively improved the processing efficiency of batch bidding data.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119046737A_ABST
    Figure CN119046737A_ABST
Patent Text Reader

Abstract

The invention relates to a data classification processing method and device based on bid invitation theme classification. The method comprises the steps of obtaining enterprise bid invitation files of each bid invitation enterprise and bid invitation auditing standard information corresponding to each bid invitation theme, and identifying bid invitation demand information and bid invitation condition information of each bid invitation theme based on each bid invitation auditing standard information; according to each bid invitation theme, classifying each enterprise bid invitation file to obtain each bid invitation file group, and identifying bid invitation detail information of each enterprise bid invitation file so as to divide the bid invitation file into sub-bid invitation contents of each bid invitation category of each enterprise bid invitation file; and based on the sub-bid invitation content of each bid invitation category of each enterprise bid invitation file and each bid invitation condition information, analyzing a bid invitation evaluation value of each enterprise bid invitation file, each enterprise bid invitation file and bid invitation missing information through a big language model, and obtaining a file processing result of each enterprise bid invitation file. By adopting the method, the data processing efficiency of the batch bid invitation data can be improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the technical field of data classification processing of batch data, and in particular to a data classification processing method and device based on bidding subject classification. Background Art

[0002] When a tenderer conducts large-scale bidding, the number of bidding companies will be particularly large. Therefore, when classifying and processing data, the processing is difficult, the processing time is slow, and data processing anomalies are prone to occur. Therefore, how to improve the efficiency of classifying and processing large quantities of bidding data is the current research focus.

[0003] The traditional way to improve the efficiency of batch data classification processing is to process each batch of data through a multi-manpower, multi-time, and review processing method. However, this method requires a lot of human resources and is slow, resulting in low data processing efficiency for batch bidding data. Summary of the invention

[0004] Based on this, it is necessary to provide a data classification processing method and device based on bidding subject classification to address the above technical issues.

[0005] In a first aspect, the present application provides a data classification processing method based on bidding subject classification, comprising: Obtaining the enterprise bidding documents of each bidding enterprise and the bidding review standard information corresponding to each bidding theme, and identifying the bidding requirement information of each bidding theme and the bidding condition information corresponding to each bidding requirement information based on each bidding review standard information; According to each of the bidding themes, each enterprise bidding document is classified and processed to obtain a bidding document group corresponding to each bidding theme, and for each bidding document group, the bidding detailed information of each enterprise bidding document in the bidding document group is identified; Based on each of the bidding details and the bidding demand information of the bidding subject corresponding to the bidding document group, each of the bidding details is divided into sub-bidding contents of each bidding category of each enterprise bidding document through a content division strategy; Based on the sub-tendering contents of each tendering category in each enterprise tendering document and the tendering condition information of the tendering subject corresponding to the tendering document group, the tendering evaluation value of each enterprise tendering document and the tendering missing information of each enterprise tendering document are analyzed through a large language model, and the tendering evaluation value of each enterprise tendering document and the tendering missing information of each enterprise tendering document are used as the file processing results of each enterprise tendering document.

[0006] Optionally, the identifying, based on each bidding review standard information, the bidding requirement information of each bidding subject and the bidding condition information corresponding to each bidding requirement information includes: For each bidding review standard information, the bidding review standard information is split into content review information of each bidding category and condition review information of each bidding category; Based on the content review information of each bidding category, identifying the content format of each bidding category, the content type of each bidding category, and the content restriction information of each bidding category, and using the content format of each bidding category, the content type of each bidding category, and the content restriction information of each bidding category as the bidding demand information of the bidding subject corresponding to the bidding review standard information; Based on the condition review information of each bidding category, the category review standards of each bidding category and the indicator evaluation information of each bidding category are identified, and the category review standards of each bidding category and the indicator evaluation information of each bidding category are used as the bidding condition information of the bidding subject corresponding to the bidding review standard information.

[0007] Optionally, the bidding documents of each enterprise are classified and processed according to each bidding theme to obtain a bidding document group corresponding to each bidding theme, including: Extract the file identification information of each enterprise's bidding document, and query the bidding subject corresponding to each file identification information in the bidding database to obtain the bidding subject of each enterprise's bidding document; According to the bidding theme of each enterprise bidding document, clustering processing is performed on the bidding documents of each enterprise to obtain the bidding documents of each enterprise corresponding to each bidding theme, and the bidding documents of each enterprise corresponding to each bidding theme are used as the bidding document group corresponding to each bidding theme.

[0008] Optionally, the identifying the bidding details of each enterprise bidding document in the bidding document group includes: For each bidding document group, extract the file directory information of each enterprise bidding document in the bidding document group, and based on the file directory information of each enterprise bidding document, perform file splitting processing on the file content of each enterprise bidding document according to each file directory information to obtain the sub-file content of each directory title of each file directory information; The sub-file contents of each directory title of each file directory information are used as the bidding details information of each enterprise bidding document.

[0009] Optionally, based on each of the tender details and the tender demand information of the tender subject corresponding to the tender document group, each of the tender details is divided into sub-tender contents of each tender category of each enterprise tender document through a content division strategy, including: In the bidding database, query the subordinate information between each bidding category and each catalog title, and for each bidding detail information, split the bidding detail information into the file content group corresponding to each bidding category based on the subordinate information; The sub-file contents of each directory title in each file content group are distributed and arranged according to the arrangement order of each directory title to obtain the initial sub-tender content corresponding to each tender category; Based on the bidding demand information corresponding to each bidding category, in the initial sub-bidding content corresponding to each bidding category, marking the qualified bidding content of each sub-file content, the unqualified bidding content of each sub-file content, and the missing bidding content of each sub-file content; The marked initial sub-tendering content of each tendering category shall be used as the sub-tendering content of each tendering category.

[0010] Optionally, the sub-tendering contents of each tender category of each enterprise tender document and the tendering condition information of the tender subject corresponding to the tender document group are analyzed through a large language model to analyze the tendering evaluation value of each enterprise tender document and the tendering missing information of each enterprise tender document, including: For each enterprise bidding document, using a large language model, identifying content semantic information of sub-bidding contents of each bidding category of the enterprise bidding document and content feature information of each content semantic information; Based on the large language model, the standard feature information corresponding to the category review standard of each bidding category is identified, and the similarity between the standard feature information of each bidding category and the content feature information of each sub-bidding content of each bidding category is calculated by a similarity algorithm; For each standard feature information, the content feature information corresponding to the similarity greater than the similarity threshold is selected as the content feature information corresponding to the standard feature information, and the content feature information for which there is no corresponding standard feature information is selected as abnormal content feature information; The abnormal content feature information and the category review standard to which the standard feature information without corresponding content feature information belongs are used as the bidding missing information of the enterprise bidding document, and based on the marking information of each sub-file content, the number of content feature information with corresponding standard feature information, and the content semantic information of each sub-file content, the file evaluation value of each sub-file content is evaluated through the indicator evaluation information of each bidding category; The file evaluation values ​​of all sub-file contents are weighted and summed up by the preset weight value of the bidding category corresponding to each sub-file content to obtain the bidding evaluation value of the enterprise bidding document.

[0011] In a second aspect, the present application also provides a data classification processing device based on bidding subject classification, comprising: An acquisition module is used to acquire the enterprise bidding documents of each bidding enterprise and the bidding review standard information corresponding to each bidding theme, and based on each bidding review standard information, identify the bidding demand information of each bidding theme and the bidding condition information corresponding to each bidding demand information; An identification module is used to classify and process each enterprise's bidding documents according to each bidding theme, obtain a bidding document group corresponding to each bidding theme, and for each bidding document group, identify bidding details of each enterprise's bidding document in the bidding document group; A division module, for dividing each of the bidding details into sub-bidding contents of each bidding category of each enterprise bidding document through a content division strategy based on each of the bidding details and the bidding demand information of the bidding subject corresponding to the bidding document group; The determination module is used to analyze the tender evaluation value of each enterprise tender document and the tender missing information of each enterprise tender document through a large language model based on the sub-tender contents of each tender category of each enterprise tender document and the tender condition information of the tender subject corresponding to the tender document group, and use the tender evaluation value of each enterprise tender document and the tender missing information of each enterprise tender document as the file processing result of each enterprise tender document.

[0012] Optionally, the acquisition module is specifically used to: For each bidding review standard information, the bidding review standard information is split into content review information of each bidding category and condition review information of each bidding category; Based on the content review information of each bidding category, identifying the content format of each bidding category, the content type of each bidding category, and the content restriction information of each bidding category, and using the content format of each bidding category, the content type of each bidding category, and the content restriction information of each bidding category as the bidding demand information of the bidding subject corresponding to the bidding review standard information; Based on the condition review information of each bidding category, the category review standards of each bidding category and the indicator evaluation information of each bidding category are identified, and the category review standards of each bidding category and the indicator evaluation information of each bidding category are used as the bidding condition information of the bidding subject corresponding to the bidding review standard information.

[0013] Optionally, the identification module is specifically used to: Extract the file identification information of each enterprise's bidding document, and query the bidding subject corresponding to each file identification information in the bidding database to obtain the bidding subject of each enterprise's bidding document; According to the bidding theme of each enterprise bidding document, clustering processing is performed on the bidding documents of each enterprise to obtain the bidding documents of each enterprise corresponding to each bidding theme, and the bidding documents of each enterprise corresponding to each bidding theme are used as the bidding document group corresponding to each bidding theme.

[0014] Optionally, the identification module is specifically used to: For each bidding document group, extract the file directory information of each enterprise bidding document in the bidding document group, and based on the file directory information of each enterprise bidding document, perform file splitting processing on the file content of each enterprise bidding document according to each file directory information to obtain the sub-file content of each directory title of each file directory information; The sub-file contents of each directory title of each file directory information are used as the bidding details information of each enterprise bidding document.

[0015] Optionally, the division module is specifically used to: In the bidding database, query the subordinate information between each bidding category and each catalog title, and for each bidding detail information, split the bidding detail information into the file content group corresponding to each bidding category based on the subordinate information; The sub-file contents of each directory title in each file content group are distributed and arranged according to the arrangement order of each directory title to obtain the initial sub-tender content corresponding to each tender category; Based on the bidding demand information corresponding to each bidding category, in the initial sub-bidding content corresponding to each bidding category, marking the qualified bidding content of each sub-file content, the unqualified bidding content of each sub-file content, and the missing bidding content of each sub-file content; The marked initial sub-tendering content of each tendering category shall be used as the sub-tendering content of each tendering category.

[0016] Optionally, the determining module is specifically used to: For each enterprise bidding document, using a large language model, identifying content semantic information of sub-bidding contents of each bidding category of the enterprise bidding document and content feature information of each content semantic information; Based on the large language model, the standard feature information corresponding to the category review standard of each bidding category is identified, and the similarity between the standard feature information of each bidding category and the content feature information of each sub-bidding content of each bidding category is calculated by a similarity algorithm; For each standard feature information, the content feature information corresponding to the similarity greater than the similarity threshold is selected as the content feature information corresponding to the standard feature information, and the content feature information for which there is no corresponding standard feature information is selected as abnormal content feature information; The abnormal content feature information and the category review standard to which the standard feature information without corresponding content feature information belongs are used as the bidding missing information of the enterprise bidding document, and based on the marking information of each sub-file content, the number of content feature information with corresponding standard feature information, and the content semantic information of each sub-file content, the file evaluation value of each sub-file content is evaluated through the indicator evaluation information of each bidding category; The file evaluation values ​​of all sub-file contents are weighted and summed up by the preset weight value of the bidding category corresponding to each sub-file content to obtain the bidding evaluation value of the enterprise bidding document.

[0017] In a third aspect, the present application provides a computer device, wherein the computer device comprises a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of any one of the methods in the first aspect are implemented.

[0018] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the steps of any one of the methods in the first aspect are implemented.

[0019] In a fifth aspect, the present application provides a computer program product. The computer program product includes a computer program, and when the computer program is executed by a processor, the steps of any one of the methods in the first aspect are implemented.

[0020] The data classification processing method and device based on the classification of bidding topics obtain the enterprise bidding documents of each bidding enterprise and the bidding review standard information corresponding to each bidding topic, and identify the bidding demand information of each bidding topic and the bidding condition information corresponding to each bidding demand information based on each bidding review standard information; classify and process each enterprise bidding document according to each bidding topic to obtain the bidding document group corresponding to each bidding topic, and identify the bidding detailed information of each enterprise bidding document in the bidding document group for each bidding document group; based on each bidding detailed information and the bidding demand information of the bidding topic corresponding to the bidding document group, divide each bidding detailed information into sub-bidding content of each bidding category of each enterprise bidding document through a content division strategy; based on the sub-bidding content of each bidding category of each enterprise bidding document and the bidding condition information of the bidding topic corresponding to the bidding document group, analyze the bidding evaluation value of each enterprise bidding document and the bidding missing information of each enterprise bidding document through a large language model, and use the bidding evaluation value of each enterprise bidding document and the bidding missing information of each enterprise bidding document as the file processing result of each enterprise bidding document. This solution improves the accuracy and comprehensiveness of the review of each bidding theme by splitting the bidding documents of each enterprise into bidding document groups corresponding to each bidding theme, and then classifying and reviewing them according to the bidding review standard information corresponding to each bidding theme. Then, this solution can also use a large language model to perform intelligent bidding evaluation and intelligent bidding missing document analysis on each enterprise bidding document, avoiding the high cost, low timeliness, and frequent re-reviews of manual review. This can not only effectively improve the classification processing efficiency and review efficiency of each enterprise bidding document, but also improve the review accuracy and comprehensiveness, thereby comprehensively improving the data processing efficiency of batch bidding data. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0022] Figure 1 A schematic diagram of a flow chart of a data classification processing method based on bidding subject classification in one embodiment; Figure 2 A flowchart of an example of data classification processing based on bidding subject classification in one embodiment; Figure 3 It is a structural block diagram of a data classification processing device based on bidding subject classification in one embodiment; Figure 4 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0023] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0024] The data classification processing method based on the classification of tender themes provided in the embodiment of the present application can be applied to the application environment of data classification processing of batch tender documents. The method can be applied to a terminal, a server, or a system including a terminal and a server, and is realized by the interaction between the terminal and the server. Among them, the terminal can be, but is not limited to, various personal computers, laptops, smart phones, tablet computers, etc. Among them, the terminal divides the tender documents of each enterprise into tender document groups corresponding to each tender theme, and then classifies and audits the tender audit standard information corresponding to each tender theme, thereby improving the audit accuracy and comprehensiveness of each tender theme. Then, this scheme can also perform intelligent bidding evaluation and intelligent bidding missing file analysis on each enterprise tender document through a large language model, avoiding the high cost, low timeliness, and more review problems of manual audit, thereby effectively improving not only the classification processing efficiency and audit efficiency of each enterprise tender document, but also the audit accuracy and comprehensiveness, thereby comprehensively improving the data processing efficiency of batch tender data.

[0025] In an exemplary embodiment, Figure 1 As shown, a data classification processing method based on bidding theme classification is provided, and the method is applied to a terminal as an example for explanation, including the following steps S101 to S104. Among them: Step S101, obtaining the enterprise bidding documents of each bidding enterprise and the bidding review standard information corresponding to each bidding theme, and based on each bidding review standard information, identifying the bidding requirement information of each bidding theme and the bidding condition information corresponding to each bidding requirement information.

[0026] In this embodiment, the terminal receives the file content transmitted from each bidding enterprise terminal, obtains the enterprise bidding documents of each bidding enterprise, and obtains the bidding review standard information corresponding to each bidding theme in response to the information upload operation of the staff. Among them, the enterprise bidding document can be a bidding document for different bidding themes. For example, enterprise A has bidding requirements for multiple projects, each project corresponds to a bidding theme, and an enterprise formulates a bidding document for each project that wants to bid, and then obtains the enterprise bidding document corresponding to each bidding theme. Among them, the bidding review standard information is the review standard for each bidding theme, and the review standard includes content review information and condition review information, wherein the content review information is the review information for reviewing the content format of each bidding category, the content type of each bidding category, and the content restriction information of each bidding category. Among them, the content type includes the type of each data type, for example, text type, picture type, table type, video type, etc., the content format corresponds to the format requirement information of different data types, and the content restriction information includes the word limit, length limit, text language limit and other restriction information of each content. The condition review information includes the category review standard of each bidding category and the indicator evaluation information of each bidding category. Among them, each bidding category includes but is not limited to various bidding sub-items, such as company profile, company R&D capability introduction, company honor introduction, company and project-related personnel qualifications introduction, company project undertaking experience introduction, company cash flow summary report and other sub-items. Among them, the category review standard is the review standard information of each category, and the review standard information is used to review whether there is content information required for the category in each category. When there is no sub-file content corresponding to the review standard information, the review standard information is missing information, and the review process will be described in detail later. Finally, based on each bidding review standard information, the terminal identifies the bidding demand information of each bidding theme and the bidding condition information corresponding to each bidding demand information.

[0027] Step S102, classify and process each enterprise's bidding document according to each bidding theme, obtain a bidding document group corresponding to each bidding theme, and for each bidding document group, identify bidding detailed information of each enterprise's bidding document in the bidding document group.

[0028] In this embodiment, the terminal classifies and processes each enterprise bidding document according to each bidding theme, obtains a bidding document group corresponding to each bidding theme, and identifies the bidding details of each enterprise bidding document in the bidding document group for each bidding document group. The bidding details include the category review standards of each bidding category and the indicator evaluation information of each bidding category. The specific identification process will be described in detail later.

[0029] Step S103, based on each bidding detail information and the bidding requirement information of the bidding subject corresponding to the bidding document group, each bidding detail information is divided into sub-bidding contents of each bidding category of each enterprise bidding document through a content division strategy.

[0030] In this embodiment, based on each bidding detail information and the bidding demand information of the bidding theme corresponding to the bidding document group, the terminal divides each bidding detail information into sub-bidding contents of each bidding category of each enterprise bidding document through a content division strategy. Each sub-bidding content is the content of each sub-file obtained after dividing and marking each bidding detail information. The marked content includes qualified marked content, unqualified marked content, and missing marked content. The specific marking process will be described in detail later.

[0031] Step S104, based on the sub-tendering contents of each tender category of each enterprise tender document and the tendering condition information of the tender subject corresponding to the tender document group, the tender evaluation value of each enterprise tender document and the tender missing information of each enterprise tender document are analyzed through a large language model, and the tender evaluation value of each enterprise tender document and the tender missing information of each enterprise tender document are used as the file processing result of each enterprise tender document.

[0032] In this embodiment, the terminal analyzes the bidding evaluation value of each enterprise bidding document and the bidding missing information of each enterprise bidding document through a large language model based on the sub-bidding content of each bidding category of each enterprise bidding document and the bidding condition information of the bidding subject corresponding to the bidding document group, and uses the bidding evaluation value of each enterprise bidding document and the bidding missing information of each enterprise bidding document as the file processing result of each enterprise bidding document. The specific analysis process will be described in detail later.

[0033] Based on the above scheme, by splitting the bidding documents of each enterprise into bidding document groups corresponding to each bidding theme, and then classifying and reviewing them according to the bidding review standard information corresponding to each bidding theme, the review accuracy and comprehensiveness of each bidding theme are improved. Then, this scheme can also use a large language model to perform intelligent bidding evaluation and intelligent bidding missing document analysis on each enterprise bidding document, avoiding the high cost, low timeliness, and frequent re-review problems of manual review, thereby effectively improving not only the classification processing efficiency and review efficiency of each enterprise bidding document, but also the review accuracy and comprehensiveness, thereby comprehensively improving the data processing efficiency of batch bidding data.

[0034] Optionally, based on each tender review standard information, the tender demand information of each tender subject and the tender condition information corresponding to each tender demand information are identified, including: for each tender review standard information, the tender review standard information is split into content review information of each tender category and condition review information of each tender category; based on the content review information of each tender category, the content format of each tender category, the content type of each tender category, and the content restriction information of each tender category are identified, and the content format of each tender category, the content type of each tender category, and the content restriction information of each tender category are used as the tender demand information of the tender subject corresponding to the tender review standard information; based on the condition review information of each tender category, the category review standard of each tender category and the indicator evaluation information of each tender category are identified, and the category review standard of each tender category and the indicator evaluation information of each tender category are used as the tender condition information of the tender subject corresponding to the tender review standard information.

[0035] In this embodiment, the terminal divides the tender review standard information into content review information of each tender category and condition review information of each tender category for each tender review standard information. Then, based on the content review information of each tender category, the terminal identifies the content format of each tender category, the content type of each tender category, and the content restriction information of each tender category, and uses the content format of each tender category, the content type of each tender category, and the content restriction information of each tender category as the tender demand information of the tender subject corresponding to the tender review standard information. Among them, the identification method is that the staff adds the content format identifier, the content type identifier, and the content restriction information identifier to the content format, content type, and content restriction information of each tender category, and then the terminal identifies the content format, content type, and content restriction information of each tender category by identifying each identification information. Among them, the identification information can be but not limited to identification information such as color identification, text identification, and symbol identification.

[0036] Then, the terminal identifies the category review standard of each bidding category and the index evaluation information of each bidding category based on the condition review information of each bidding category, and uses the category review standard of each bidding category and the index evaluation information of each bidding category as the bidding condition information of the bidding subject corresponding to the bidding review standard information. Among them, the index evaluation information includes the evaluation index value corresponding to each evaluation content range of each evaluation type, and the terminal determines each evaluation index value by identifying the evaluation content range to which the evaluation content of each evaluation type belongs.

[0037] Based on the above solution, by splitting the bidding review standard information into content review information of each bidding category and condition review information of each bidding category, each bidding category can be analyzed and identified, thereby improving the comprehensiveness and granularity of the analysis.

[0038] Optionally, each enterprise bidding document is classified according to each bidding theme to obtain a bidding document group corresponding to each bidding theme, including: extracting the file identification information of each enterprise bidding document, and querying the bidding theme corresponding to each file identification information in the bidding database to obtain the bidding theme of each enterprise bidding document; clustering the enterprise bidding documents according to the bidding theme of each enterprise bidding document to obtain the enterprise bidding documents corresponding to each bidding theme, and using the enterprise bidding documents corresponding to each bidding theme as the bidding document group corresponding to each bidding theme.

[0039] In this embodiment, the terminal extracts the file identification information of each enterprise bidding document, and searches the bidding subject corresponding to each file identification information in the bidding database to obtain the bidding subject of each enterprise bidding document. Then, the terminal performs clustering processing on each enterprise bidding document according to the bidding subject of each enterprise bidding document to obtain each enterprise bidding document corresponding to each bidding subject, and uses each enterprise bidding document corresponding to each bidding subject as the bidding document group corresponding to each bidding subject.

[0040] Based on the above solution, by dividing the bidding documents of each enterprise into bidding themes, the processing accuracy and efficiency of the enterprise bidding documents corresponding to each bidding theme are improved.

[0041] Optionally, identifying detailed bidding information of each enterprise bidding document in the bidding document group includes: for each bidding document group, extracting the file directory information of each enterprise bidding document in the bidding document group, and based on the file directory information of each enterprise bidding document, performing file splitting processing on the file content of each enterprise bidding document according to each file directory information to obtain the sub-file content of each directory title of each file directory information; using the sub-file content of each directory title of each file directory information as the detailed bidding information of each enterprise bidding document.

[0042] In this embodiment, the terminal extracts the file directory information of each enterprise bidding document in the bidding document group for each bidding document group, and based on the file directory information of each enterprise bidding document, performs file splitting processing on the file content of each enterprise bidding document according to each file directory information, and obtains the sub-file content of each directory title of each file directory information. The directory title includes a large title, a middle title, and a small title, that is, as long as it is a title existing in the directory, it is a directory title in the directory information of the file, and there is a parallel relationship or a subordinate relationship between each directory title.

[0043] Finally, the terminal uses the sub-file contents of each directory title of each file directory information as the bidding details information of each enterprise bidding document.

[0044] Based on the above solution, the comprehensiveness and accuracy of the analysis of each company's bidding documents are improved by splitting the contents of each bidding document into directories.

[0045] Optionally, based on each bidding detail information and the bidding demand information of the bidding theme corresponding to the bidding document group, each bidding detail information is divided into sub-bidding content of each bidding category of each enterprise bidding document through a content division strategy, including: in the bidding database, querying the subordinate information between each bidding category and each directory title, and for each bidding detail information, based on the subordinate information, splitting the bidding detail information into the file content group corresponding to each bidding category; distributing and arranging the sub-file contents of each directory title in each file content group according to the arrangement order of each directory title to obtain the initial sub-bidding content corresponding to each bidding category; based on the bidding demand information corresponding to each bidding category, marking the qualified bidding content of each sub-file content, the unqualified bidding content of each sub-file content, and the missing bidding content of each sub-file content in the initial sub-bidding content corresponding to each bidding category; using the marked initial sub-bidding content of each bidding category as the sub-bidding content of each bidding category.

[0046] In this embodiment, the terminal searches for subordinate information between each bidding category and each catalog title in the bidding database. The bidding database stores a catalog title range corresponding to a bidding category, and then the terminal treats all first catalog titles belonging to the catalog title range of the bidding category and all second catalog titles subordinate to the first catalog title as catalog titles with subordinate information to the bidding category.

[0047] For each bidding detail information, the terminal splits the bidding detail information into file content groups corresponding to each bidding category based on subordinate information. Then, the terminal distributes and arranges the sub-file contents of each directory title in each file content group according to the arrangement order of each directory title to obtain the initial sub-bidding content corresponding to each bidding category.

[0048] Based on the bidding demand information corresponding to each bidding category, in the initial sub-bidding content corresponding to each bidding category, the qualified bidding content of each sub-file content, the unqualified bidding content of each sub-file content, and the missing bidding content of each sub-file content are marked. Among them, the marking process is that the terminal identifies whether the file content format of the file content in each sub-file content conforms to the content format of the bidding category, whether the file content type of the file content in each sub-file content belongs to the content type of the bidding category, and whether the content information of the file content in each sub-file content does not exceed the content restriction information of the bidding category through the content format of each bidding category, the content type of each bidding category, and the content restriction information of each bidding category. If there is a sub-file content that does not conform to the content format of the bidding category, does not belong to the content type of the bidding category, or exceeds the content restriction information of the bidding category, then the sub-file content is an unqualified bidding content. Only when the sub-file content meets the above information is a qualified bidding content, and when there is a missing file content in the sub-file content, then the sub-file content is a missing bidding content. Finally, the terminal uses the marked initial sub-bidding content of each bidding category as the sub-bidding content of each bidding category.

[0049] Based on the above scheme, after the files of each directory title are divided, the content of the sub-files of each directory title is reviewed, thereby avoiding the inefficiency of manual review and classification and improving the efficiency and accuracy of the review.

[0050] Optionally, based on the sub-tender contents of each tender category of each enterprise tender document and the tender condition information of the tender subject corresponding to the tender document group, the tender evaluation value of each enterprise tender document and the tender missing information of each enterprise tender document are analyzed through a large language model, including: for each enterprise tender document, identifying the content semantic information of the sub-tender contents of each tender category of the enterprise tender document and the content feature information of each content semantic information through a large language model; identifying the standard feature information corresponding to the category review standard of each tender category based on the large language model, and calculating the similarity between the standard feature information of each tender category and the content feature information of each sub-tender content of each tender category through a similarity algorithm; for each standard feature information, screening The content feature information corresponding to the similarity is used as the content feature information corresponding to the standard feature information, and the content feature information for which there is no corresponding standard feature information is used as abnormal content feature information; the category review standard to which the abnormal content feature information and the standard feature information for which there is no corresponding content feature information belong is used as the bidding missing information of the enterprise bidding document, and based on the marking information of each sub-file content, the number of content feature information for which there is corresponding standard feature information, and the content semantic information of each sub-file content, the file evaluation value of each sub-file content is evaluated through the indicator evaluation information of each bidding category; the file evaluation values ​​of all sub-file contents are weighted and summed up through the preset weight value of the bidding category corresponding to each sub-file content to obtain the bidding evaluation value of the enterprise bidding document.

[0051] In this embodiment, the terminal uses a large language model to identify the content semantic information of the sub-tender content of each tender category of the enterprise tender document and the content feature information of each content semantic information for each enterprise tender document. The large language model is a neural network model based on natural language processing (NLP), which can identify the text semantic information of each text and extract the text feature information of the text content.

[0052] Then, based on the large language model, the terminal identifies the standard feature information corresponding to the category review standards of each bidding category, and calculates the similarity between the standard feature information of each bidding category and the content feature information of each sub-bidding content of each bidding category through a similarity algorithm, wherein the similarity algorithm is a cosine similarity algorithm.

[0053] For each standard feature information, the terminal screens the content feature information corresponding to the similarity greater than the similarity threshold preset in the terminal as the content feature information corresponding to the standard feature information, and the content feature information for which there is no corresponding standard feature information is taken as abnormal content feature information. Among them, the abnormal content feature information is the content feature information for which there is no corresponding standard feature information. Each standard feature information can correspond to one or more content feature information, and each content feature information corresponds to only one standard feature information. For example, the standard feature information is that the number of practicing professionals with ten years of experience is greater than 10 people, and the content feature information includes the basic introduction of each researcher. Then the content feature information of the researchers who meet the requirements of the practicing professionals with ten years of experience is the content feature information corresponding to the standard feature information. There is also a content feature information that is the statistical value of the number of practicing professionals with ten years of experience in the company. When the statistical value of the number is greater than 10 people, the content feature information corresponding to the statistical value of the number is the content feature information corresponding to the standard feature information. When the statistical value of the number is greater than 10 people, the content feature information corresponding to the statistical value of the number is not the content feature information corresponding to the standard feature information.

[0054] Then, the terminal uses the abnormal content feature information and the category review standard to which the standard feature information without corresponding content feature information belongs as the bidding missing information of the enterprise bidding document.

[0055] Then, based on the marking information of each sub-file content, the number of content feature information with corresponding standard feature information, and the content semantic information of each sub-file content, the terminal evaluates the file evaluation value of each sub-file content through the index evaluation information of each bidding category. The index evaluation information of the bidding category includes the sub-index evaluation information corresponding to the marking information, the sub-index evaluation information corresponding to the number of content feature information with corresponding standard feature information, and the sub-index evaluation information corresponding to the content semantic information of the sub-file content. The terminal evaluates each sub-index evaluation information separately to obtain the file evaluation value of the sub-file content, and then the terminal performs weighted summation processing on the file evaluation values ​​of all sub-file contents through the preset weight value of the bidding category corresponding to each sub-file content to obtain the bidding evaluation value of the enterprise bidding document.

[0056] Based on the above solution, a large language model is used to conduct intelligent bidding evaluation and intelligent bidding missing document analysis on each enterprise's bidding documents, avoiding the high cost, low timeliness, and frequent re-examination problems of manual review. This can not only effectively improve the classification processing efficiency and review efficiency of each enterprise's bidding documents, but also improve the review accuracy and comprehensiveness, thereby comprehensively improving the data processing efficiency of batch bidding data.

[0057] This application also provides an example of data classification processing based on bidding subject classification, such as Figure 2 As shown, the specific processing process includes the following steps: Step S201, obtaining the enterprise bidding documents of each bidding enterprise and the bidding review standard information corresponding to each bidding theme.

[0058] Step S202 , for each bidding review standard information, split the bidding review standard information into content review information of each bidding category and condition review information of each bidding category.

[0059] Step S203, based on the content review information of each bidding category, identify the content format of each bidding category, the content type of each bidding category, and the content restriction information of each bidding category, and use the content format of each bidding category, the content type of each bidding category, and the content restriction information of each bidding category as the bidding requirement information of the bidding subject corresponding to the bidding review standard information.

[0060] Step S204, based on the condition review information of each bidding category, identify the category review standards of each bidding category and the indicator evaluation information of each bidding category, and use the category review standards of each bidding category and the indicator evaluation information of each bidding category as the bidding condition information of the bidding subject corresponding to the bidding review standard information.

[0061] Step S205, extracting the file identification information of each enterprise's bidding document, and searching the bidding subject corresponding to each file identification information in the bidding database to obtain the bidding subject of each enterprise's bidding document.

[0062] Step S206, clustering the tender documents of each enterprise according to the tender subject of each enterprise tender document to obtain the tender documents of each enterprise corresponding to each tender subject, and taking the tender documents of each enterprise corresponding to each tender subject as the tender document group corresponding to each tender subject.

[0063] Step S207, for each bidding document group, extract the file directory information of each enterprise bidding document in the bidding document group, and based on the file directory information of each enterprise bidding document, split the file content of each enterprise bidding document according to each file directory information to obtain the sub-file content of each directory title of each file directory information.

[0064] Step S208, using the sub-file contents of each directory title of each file directory information as the bidding details information of each enterprise bidding document.

[0065] Step S209, querying the subordinate information between each bidding category and each catalog title in the bidding database, and for each bidding detail information, splitting the bidding detail information into file content groups corresponding to each bidding category based on the subordinate information.

[0066] Step S210: distribute and arrange the sub-file contents of each directory title in each file content group according to the arrangement order of each directory title to obtain initial sub-bidding contents corresponding to each bidding category.

[0067] Step S211, based on the bidding requirement information corresponding to each bidding category, in the initial sub-bidding content corresponding to each bidding category, marking the qualified bidding content of each sub-file content, the unqualified bidding content of each sub-file content, and the missing bidding content of each sub-file content.

[0068] Step S212: The marked initial sub-tendering content of each tendering category is used as the sub-tendering content of each tendering category.

[0069] Step S213: for each enterprise bidding document, the content semantic information of the sub-bidding contents of each bidding category of the enterprise bidding document and the content feature information of each content semantic information are identified by using a large language model.

[0070] Step S214, based on the large language model, identify the standard feature information corresponding to the category review standard of each bidding category, and calculate the similarity between the standard feature information of each bidding category and the content feature information of each sub-bidding content of each bidding category through a similarity algorithm.

[0071] Step S215 , for each standard feature information, the content feature information corresponding to the similarity greater than the similarity threshold is screened as the content feature information corresponding to the standard feature information, and the content feature information for which there is no corresponding standard feature information is taken as abnormal content feature information.

[0072] Step S216, uses the category review standards of abnormal content feature information and standard feature information without corresponding content feature information as the bidding missing information of the enterprise bidding document, and evaluates the file evaluation value of each sub-file content based on the marking information of each sub-file content, the number of content feature information with corresponding standard feature information, and the content semantic information of each sub-file content through the indicator evaluation information of each bidding category.

[0073] Step S217, the file evaluation values ​​of all sub-file contents are weighted and summed up by the preset weight value of the tender category corresponding to each sub-file content to obtain the tender evaluation value of the enterprise tender document.

[0074] Step S218, taking the bidding evaluation value of each enterprise bidding document and the bidding missing information of each enterprise bidding document as the file processing result of each enterprise bidding document.

[0075] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0076] Based on the same inventive concept, the embodiment of the present application also provides a data classification processing device based on tender subject classification for implementing the data classification processing method based on tender subject classification involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in the embodiments of one or more data classification processing devices based on tender subject classification provided below can refer to the limitations of the data classification processing method based on tender subject classification above, and will not be repeated here.

[0077] In an exemplary embodiment, Figure 3 As shown, a data classification processing device based on tender subject classification is provided, comprising: an acquisition module 310, an identification module 320, a division module 330 and a determination module 340, wherein: The acquisition module 310 is used to acquire the enterprise bidding documents of each bidding enterprise and the bidding review standard information corresponding to each bidding theme, and identify the bidding requirement information of each bidding theme and the bidding condition information corresponding to each bidding requirement information based on each bidding review standard information; The identification module 320 is used to classify the bidding documents of each enterprise according to the bidding themes, obtain the bidding document group corresponding to each bidding theme, and for each bidding document group, identify the bidding details of the bidding documents of each enterprise in the bidding document group; A division module 330 is used to divide each of the bidding details into sub-bidding contents of each bidding category of each enterprise bidding document through a content division strategy based on each of the bidding details and the bidding demand information of the bidding subject corresponding to the bidding document group; Determination module 340 is used to analyze the bidding evaluation value of each enterprise bidding document and the bidding missing information of each enterprise bidding document based on the sub-bidding content of each bidding category of each enterprise bidding document and the bidding condition information of the bidding subject corresponding to the bidding document group through a large language model, and use the bidding evaluation value of each enterprise bidding document and the bidding missing information of each enterprise bidding document as the file processing result of each enterprise bidding document.

[0078] Optionally, the acquisition module 310 is specifically configured to: For each bidding review standard information, the bidding review standard information is split into content review information of each bidding category and condition review information of each bidding category; Based on the content review information of each bidding category, identifying the content format of each bidding category, the content type of each bidding category, and the content restriction information of each bidding category, and using the content format of each bidding category, the content type of each bidding category, and the content restriction information of each bidding category as the bidding demand information of the bidding subject corresponding to the bidding review standard information; Based on the condition review information of each bidding category, the category review standards of each bidding category and the indicator evaluation information of each bidding category are identified, and the category review standards of each bidding category and the indicator evaluation information of each bidding category are used as the bidding condition information of the bidding subject corresponding to the bidding review standard information.

[0079] Optionally, the identification module 320 is specifically configured to: Extract the file identification information of each enterprise's bidding document, and query the bidding subject corresponding to each file identification information in the bidding database to obtain the bidding subject of each enterprise's bidding document; According to the bidding theme of each enterprise bidding document, clustering processing is performed on the bidding documents of each enterprise to obtain the bidding documents of each enterprise corresponding to each bidding theme, and the bidding documents of each enterprise corresponding to each bidding theme are used as the bidding document group corresponding to each bidding theme.

[0080] Optionally, the identification module 320 is specifically configured to: For each bidding document group, extract the file directory information of each enterprise bidding document in the bidding document group, and based on the file directory information of each enterprise bidding document, perform file splitting processing on the file content of each enterprise bidding document according to each file directory information to obtain the sub-file content of each directory title of each file directory information; The sub-file contents of each directory title of each file directory information are used as the bidding details information of each enterprise bidding document.

[0081] Optionally, the division module 330 is specifically configured to: In the bidding database, query the subordinate information between each bidding category and each catalog title, and for each bidding detail information, split the bidding detail information into the file content group corresponding to each bidding category based on the subordinate information; The sub-file contents of each directory title in each file content group are distributed and arranged according to the arrangement order of each directory title to obtain the initial sub-tender content corresponding to each tender category; Based on the bidding demand information corresponding to each bidding category, in the initial sub-bidding content corresponding to each bidding category, marking the qualified bidding content of each sub-file content, the unqualified bidding content of each sub-file content, and the missing bidding content of each sub-file content; The marked initial sub-tendering content of each tendering category shall be used as the sub-tendering content of each tendering category.

[0082] Optionally, the determining module 340 is specifically configured to: For each enterprise bidding document, using a large language model, identifying content semantic information of sub-bidding contents of each bidding category of the enterprise bidding document and content feature information of each content semantic information; Based on the large language model, the standard feature information corresponding to the category review standard of each bidding category is identified, and the similarity between the standard feature information of each bidding category and the content feature information of each sub-bidding content of each bidding category is calculated by a similarity algorithm; For each standard feature information, the content feature information corresponding to the similarity greater than the similarity threshold is selected as the content feature information corresponding to the standard feature information, and the content feature information for which there is no corresponding standard feature information is selected as abnormal content feature information; The abnormal content feature information and the category review standard to which the standard feature information without corresponding content feature information belongs are used as the bidding missing information of the enterprise bidding document, and based on the marking information of each sub-file content, the number of content feature information with corresponding standard feature information, and the content semantic information of each sub-file content, the file evaluation value of each sub-file content is evaluated through the indicator evaluation information of each bidding category; The file evaluation values ​​of all sub-file contents are weighted and summed up by the preset weight value of the bidding category corresponding to each sub-file content to obtain the bidding evaluation value of the enterprise bidding document.

[0083] Each module in the above-mentioned data classification processing device based on tender subject classification can be implemented in whole or in part by software, hardware and their combination. Each of the above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the corresponding operations of each of the above modules.

[0084] In an exemplary embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as shown in FIG. Figure 4 As shown. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit and an input device. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface, the display unit and the input device are connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be realized through WIFI, a mobile cellular network, near field communication (NFC) or other technologies. When the computer program is executed by the processor, a data classification processing method based on the classification of bidding topics is realized. The display unit of the computer device is used to form a visually visible picture, which can be a display screen, a projection device or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad set on the computer device shell, or an external keyboard, touchpad or mouse.

[0085] Those skilled in the art will understand that Figure 4 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0086] In an exemplary embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the steps of any one of the methods in the first aspect are implemented.

[0087] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of any one of the methods in the first aspect are implemented.

[0088] In one embodiment, a computer program product is provided, comprising a computer program, which, when executed by a processor, implements the steps of any one of the methods in the first aspect.

[0089] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0090] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., but are not limited to this.

[0091] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, 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, they should be considered to be within the scope of this application.

[0092] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A data classification processing method based on bidding subject classification, characterized in that: The method comprises: Obtaining the enterprise bidding documents of each bidding enterprise and the bidding review standard information corresponding to each bidding theme, and identifying the bidding requirement information of each bidding theme and the bidding condition information corresponding to each bidding requirement information based on each bidding review standard information; According to each of the bidding themes, each enterprise bidding document is classified and processed to obtain a bidding document group corresponding to each bidding theme, and for each bidding document group, the bidding detailed information of each enterprise bidding document in the bidding document group is identified; Based on each of the bidding details and the bidding demand information of the bidding subject corresponding to the bidding document group, each of the bidding details is divided into sub-bidding contents of each bidding category of each enterprise bidding document through a content division strategy; Based on the sub-tendering contents of each tendering category in each enterprise tendering document and the tendering condition information of the tendering subject corresponding to the tendering document group, the tendering evaluation value of each enterprise tendering document and the tendering missing information of each enterprise tendering document are analyzed through a large language model, and the tendering evaluation value of each enterprise tendering document and the tendering missing information of each enterprise tendering document are used as the file processing results of each enterprise tendering document.

2. The method according to claim 1, characterized in that: The step of identifying the bidding requirement information of each bidding subject and the bidding condition information corresponding to each bidding requirement information based on each bidding review standard information includes: For each bidding review standard information, the bidding review standard information is split into content review information of each bidding category and condition review information of each bidding category; Based on the content review information of each bidding category, identifying the content format of each bidding category, the content type of each bidding category, and the content restriction information of each bidding category, and using the content format of each bidding category, the content type of each bidding category, and the content restriction information of each bidding category as the bidding demand information of the bidding subject corresponding to the bidding review standard information; Based on the condition review information of each bidding category, the category review standards of each bidding category and the indicator evaluation information of each bidding category are identified, and the category review standards of each bidding category and the indicator evaluation information of each bidding category are used as the bidding condition information of the bidding subject corresponding to the bidding review standard information.

3. The method according to claim 2, characterized in that According to the bidding themes, each enterprise bidding document is classified and processed to obtain a bidding document group corresponding to each bidding theme, including: Extract the file identification information of each enterprise's bidding document, and query the bidding subject corresponding to each file identification information in the bidding database to obtain the bidding subject of each enterprise's bidding document; According to the bidding theme of each enterprise bidding document, clustering processing is performed on the bidding documents of each enterprise to obtain the bidding documents of each enterprise corresponding to each bidding theme, and the bidding documents of each enterprise corresponding to each bidding theme are used as the bidding document group corresponding to each bidding theme.

4. The method according to claim 2, characterized in that: The identification of the bidding details of each enterprise bidding document in the bidding document group includes: For each bidding document group, extract the file directory information of each enterprise bidding document in the bidding document group, and based on the file directory information of each enterprise bidding document, perform file splitting processing on the file content of each enterprise bidding document according to each file directory information to obtain the sub-file content of each directory title of each file directory information; The sub-file contents of each directory title of each file directory information are used as the bidding details information of each enterprise bidding document.

5. The method according to claim 2, characterized in that: Based on the bidding detailed information and the bidding demand information of the bidding subject corresponding to the bidding document group, the bidding detailed information is divided into sub-bidding contents of each bidding category of each enterprise bidding document through a content division strategy, including: In the bidding database, query the subordinate information between each bidding category and each catalog title, and for each bidding detail information, split the bidding detail information into the file content group corresponding to each bidding category based on the subordinate information; The sub-file contents of each directory title in each file content group are distributed and arranged according to the arrangement order of each directory title to obtain the initial sub-tender content corresponding to each tender category; Based on the bidding demand information corresponding to each bidding category, in the initial sub-bidding content corresponding to each bidding category, marking the qualified bidding content of each sub-file content, the unqualified bidding content of each sub-file content, and the missing bidding content of each sub-file content; The marked initial sub-tendering content of each tendering category shall be used as the sub-tendering content of each tendering category.

6. The method according to claim 2, characterized in that The sub-tendering contents of each tender category based on each enterprise tender document and the tendering condition information of the tender subject corresponding to the tender document group are analyzed through a large language model to analyze the tendering evaluation value of each enterprise tender document and the tendering missing information of each enterprise tender document, including: For each enterprise bidding document, using a large language model, identifying content semantic information of sub-bidding contents of each bidding category of the enterprise bidding document and content feature information of each content semantic information; Based on the large language model, the standard feature information corresponding to the category review standard of each bidding category is identified, and the similarity between the standard feature information of each bidding category and the content feature information of each sub-bidding content of each bidding category is calculated by a similarity algorithm; For each standard feature information, the content feature information corresponding to the similarity greater than the similarity threshold is selected as the content feature information corresponding to the standard feature information, and the content feature information for which there is no corresponding standard feature information is selected as abnormal content feature information; The abnormal content feature information and the category review standard to which the standard feature information without corresponding content feature information belongs are used as the bidding missing information of the enterprise bidding document, and based on the marking information of each sub-file content, the number of content feature information with corresponding standard feature information, and the content semantic information of each sub-file content, the file evaluation value of each sub-file content is evaluated through the indicator evaluation information of each bidding category; The file evaluation values ​​of all sub-file contents are weighted and summed up by the preset weight value of the bidding category corresponding to each sub-file content to obtain the bidding evaluation value of the enterprise bidding document.

7. A data classification processing device based on bidding subject classification, characterized in that: The device comprises: An acquisition module is used to acquire the enterprise bidding documents of each bidding enterprise and the bidding review standard information corresponding to each bidding theme, and based on each bidding review standard information, identify the bidding demand information of each bidding theme and the bidding condition information corresponding to each bidding demand information; An identification module is used to classify and process each enterprise's bidding documents according to each bidding theme, obtain a bidding document group corresponding to each bidding theme, and for each bidding document group, identify bidding details of each enterprise's bidding document in the bidding document group; A division module, for dividing each of the bidding details into sub-bidding contents of each bidding category of each enterprise bidding document through a content division strategy based on each of the bidding details and the bidding demand information of the bidding subject corresponding to the bidding document group; A determination module is used to analyze the tender evaluation value of each enterprise tender document and the tender missing information of each enterprise tender document through a large language model based on the sub-tender contents of each tender category of each enterprise tender document and the tender condition information of the tender subject corresponding to the tender document group, and use the tender evaluation value of each enterprise tender document and the tender missing information of each enterprise tender document as the file processing result of each enterprise tender document.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.