Intelligent bidding document correction method and correction system based on AI large model

Abstract

The invention relates to the technical field of artificial intelligence, in particular to an intelligent bidding document correction method and system based on an AI large model, and the method comprises the steps: obtaining a plurality of pieces of standard bidding document information, and carrying out the AI large model training according to the plurality of pieces of standard bidding document information to obtain standard library information; comparing the standard library information with the obtained new bidding document information to obtain a bidding document difference information set; and according to the bidding document difference information, judging whether the new bidding document information accords with a predetermined bidding. According to the method, the standard bidding documents of one batch are obtained, the information which exists most frequently in the same type is extracted to establish the bidding document database, the bidding documents input subsequently are compared with the bidding document database to generate the comparison result, and if the bidding documents meet requirements, the bidding documents can be directly applied to a bidding link; if the requirements are not met, labeling and automatic modification are carried out according to auditing provided by AI intelligentization, and the missing part of the bidding document can be actively modified through AI intelligentization prompt words.

Description

Technical Field

[0001] The present invention relates to the technical field of artificial intelligence technology, and in particular to a bid document correction method and correction system based on intelligent AI large model. Background Art

[0002] The bid document is the most important part of the bidding process. The quality of the bid document directly determines whether the project is awarded or rejected. The complexity of the bid document and the errors and omissions in the bid document are one of the reasons for the rejection of the bid.

[0003] In order to avoid the situation of bid rejection, multiple inspections are required, but manual inspections take a long time and lack accuracy, making it difficult to effectively avoid bid defects. Summary of the Invention

[0004] The technical problem to be solved by the present invention is: to overcome the problem that manual inspection of bid documents in the existing technology cannot overcome the defect of bid rejection, and to provide a bid document correction method and correction system based on AI large model intelligence.

[0005] The technical solution adopted by the present invention to solve the technical problem is: a bid revision method based on intelligent AI large model, comprising: Obtain a certain amount of standard tender information, and train the AI model based on this information to obtain standard library information; Comparing the standard library information with the acquired new bid information to obtain a bid difference information set; Determining whether the new bid information meets the predetermined standards based on the bid difference information; If it meets the predetermined standards, the new bid information is judged to be qualified; If the predetermined standards are not met, the bid will be marked and corrected based on the difference information in the bid; Obtain the correction content, and perform an AI large model judgment based on the correction content to determine whether it meets the predetermined standards.

[0006] Preferably, the step of training the AI large model based on a number of standard bidding document information to obtain standard library information includes: The classification analysis is performed based on a number of standard bidding document information to obtain homogeneous data groups and non-homogeneous data groups; An AI training model is obtained according to the homogeneous data acquisition process, and standard library information is obtained based on the AI training model.

[0007] Preferably, the step of performing classification analysis based on a number of standard tender information to obtain homogeneous data groups and heterogeneous data groups includes: Classify a certain amount of standard bidding information to obtain business information, technical information, quotation information and basic document information; Analyze the number of times business information, technical information, quotation information, and basic document information of different specifications appear in several standard bidding documents, and calculate the proportion of each information of different specifications in the total information; According to the size of the proportion value, a homogeneous data group and a non-homogeneous data group are divided; An AI training model is performed according to the acquisition process of the homogeneous data group, and standard library information is obtained based on the AI training model.

[0008] Preferably, the step of marking and correcting based on the bid difference information includes: Acquiring a plurality of bid document difference information to obtain a plurality of defect locations, performing color processing according to the plurality of defect locations, and automatically marking based on the color processing sections; Identify the automatic annotation information and classify it into manual modification requirement suggestions and automatic modification requirement suggestions, and automatically correct based on the automatic modification requirement suggestions and manually modify according to the manual modification requirements.

[0009] Preferably, the step of using the AI large model to determine whether the correction content meets the predetermined standard includes: Perform AI large model judgment based on the correction content to obtain feedback information; The feedback information is compared with the predetermined standard frame by frame to determine whether the feedback information meets the requirements.

[0010] Preferably, the present invention further provides an intelligent tender document revision system based on an AI large model, comprising: The extraction module obtains a certain amount of standard bidding document information, and trains the AI large model based on the information to obtain the standard library information; A first judgment module compares the standard library information with the acquired new bid information to obtain a bid difference information set; A second judgment module, judging whether the new bid information meets the predetermined standard according to the bid difference information; If it meets the predetermined standards, the new bid information is judged to be qualified; If the predetermined standards are not met, the bid will be marked and corrected based on the difference information in the bid; The correction module obtains the correction content and uses the AI large model to determine whether the correction content meets the predetermined standards.

[0011] The present invention also provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and wherein the processor implements the steps of the method according to any one of claims 1 to 5 when executing the computer program.

[0012] The present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the control method for automatic inspection of PCB appearance defects.

[0013] The beneficial effect of the present invention is that the present invention obtains a batch of standard bids, extracts the information of the same type that exists the most times, establishes a bid database, and compares the subsequently entered bids with the bid database to produce a comparison result. If the bid meets the requirements, it can be directly applied to the bidding process. If it does not meet the requirements, it is marked and automatically modified according to the review provided by AI intelligence. The missing parts of the bid can be actively modified through AI intelligent prompts, so that the bid meets the requirements based on reference. In addition, the present application is also provided with verification measures. The revised bid can be checked again to see if it meets the standards, so as to avoid bid defects through multiple efficient cycle modes and avoid the occurrence of bid cancellation. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] The present invention will be further described below with reference to the accompanying drawings and examples.

[0015] Figure 1 Schematic diagram of the method of the present invention.

[0016] Figure 2 Schematic diagram of the system structure of the present invention.

[0017] Figure 3 This is a schematic diagram of the internal structure of the computer device of this application. DETAILED DESCRIPTION

[0018] It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0019] like Figure 1-Figure 3 As shown, this application provides a bid revision method based on AI big model intelligence, including: S1. Obtain a certain amount of standard bidding document information, and train an AI large model based on the information to obtain standard library information; S2. Compare the standard library information with the acquired new bid information to obtain a bid difference information set; S3, judging whether the new bid information meets the predetermined standard based on the bid difference information; If it meets the predetermined standards, the new bid information is judged to be qualified; If the predetermined standards are not met, the bid will be marked and corrected based on the difference information in the bid; S4. Obtain the correction content, and use the AI big model to determine whether the correction content meets the predetermined standards.

[0020] As described in steps S1-S4 above, the standards of bid documents have continued to improve with the verification in recent years. This application obtains a certain amount of standard bid document information by referring to the recent bid document standards that comply with the new policy to avoid old bid document standards not meeting the new regulations. The acquisition method can be through crawling and manual input. The bid document acquisition method can then be based on the AI large model training to form a memory and continuously updated to form a standard library. After the actual bid document is completed, it can be compared and verified based on the standard library. The differences in the comparison will be displayed as data for reference. Then, multiple differences form a difference collection, such as font size, text format, paragraph format, and typo status. The intelligent AI function will judge the difference collection based on the previous training results. If it meets the standards, the bid is qualified and the work is completed. If it does not meet the standards, the intelligent AI function will automatically repair the formal defects and then display them together with the defects that cannot be automatically repaired. The staff can manually correct them based on this. The correction content can be sent back for re-verification until the standard defects are eliminated. This reduces the consumption of manual effort, ensures the qualification rate and quality of the bid, and greatly reduces the occurrence of rejected bids.

[0021] In one embodiment, the step S1 of training the AI large model based on a number of standard bidding documents to obtain standard library information includes: S11, performing classification analysis based on a certain amount of standard bidding information to obtain homogeneous data groups and heterogeneous data groups; S12. Obtain an AI training model according to the homogeneous data acquisition process, and obtain standard library information based on the AI training model.

[0022] As described in the above steps S11-S12, in this application, the multiple bid document information obtained has multiple versions. For example, the line spacing between different bid documents is different. Six out of ten documents have the same line spacing, which are defined as homogeneous data groups. The other four documents have different or partially identical line spacing, which are defined as non-homogeneous data groups. Excluding the non-homogeneous data group part, the algorithm based on the AI training model accumulates the information of the homogeneous data group into standard library information, so that the content of the bid has a relatively absolute reference standard.

[0023] In one embodiment, the step S11 of performing classification analysis based on a number of standard bid information to obtain homogeneous data groups and heterogeneous data groups includes: S111. Classify a certain amount of standard bidding document information to obtain business information, technical information, quotation information, and basic document information; S112. Analyze the number of times business information, technical information, quotation information, and basic document information of different specifications appear in several standard bidding documents, and calculate the proportion of each information of different specifications in the total information; S113, dividing the data into homogeneous data groups and heterogeneous data groups according to the values of the proportions; S114: Perform an AI training model according to the acquisition process of the homogeneous data group, and obtain standard library information based on the AI training model.

[0024] As described in the above steps S111-S114, the bid document in this application has multiple components such as business information, technical information, quotation information and basic file information. The text format, paragraphs, whether to stamp or not, and whether to sign or not of each part are different. The bid document in this application is roughly divided into four sections. The frequency of occurrence of the same type and standard data in each section is analyzed, and the low-frequency data is discarded. The high-frequency data is obtained based on the calculation of the proportion of the low-frequency data and the total frequency. Based on the proportion data and the calculation process, homogeneous data groups and non-homogeneous data groups are obtained, and the non-homogeneous data groups are discarded. The homogeneous data groups are used as annotations to establish a standard library to meet the needs of subsequent work.

[0025] In one embodiment, the step S3 of marking and correcting based on the bid difference information includes: S31, obtaining a plurality of bid document difference information to obtain a plurality of defect locations, performing color processing according to the plurality of defect locations, and automatically marking based on the color processing sections; S32: Identify and classify the automatically marked information into manual modification requirement suggestions and automatic modification requirement suggestions, and automatically correct based on the automatic modification requirement suggestions and manually modify according to the manual modification requirement.

[0026] As described in the above steps S31-S32, the bid document is a text, and different text defects will be displayed in different paragraphs. Under the conditions of the AI intelligent body, the defective colored parts of each paragraph will be automatically marked. The automatic marking content is error prompts and modification suggestions. Then, the AI intelligent body can automatically modify the bid document content that meets the automatic modification conditions and form a modification log for manual adjustment and tracking records. At the same time, the bid document maker can modify the content that the AI intelligent body cannot modify based on the marking, such as signatures and seals, which are not suitable for automatic generation.

[0027] In one embodiment, step S4 of performing the AI large model to determine whether the correction content meets the predetermined standard includes: S41, performing AI large model judgment based on the correction content to obtain feedback information; S42: Compare the feedback information with the predetermined standard frame by frame to determine whether the feedback information meets the requirements.

[0028] As shown in steps S41-S42 above, whether the defects of the bid document corrected by the intelligent agent and manual correction in this application meet the bidding standards, the revised text needs to be checked twice, and the inspection location is every category and paragraph of the revised bid document to avoid the situation where the correct content is modified into incorrect content due to negligence.

[0029] In one embodiment, the present invention further provides an intelligent tender document revision system based on an AI large model, comprising: The extraction module obtains a certain amount of standard bidding document information, and trains the AI large model based on the information to obtain the standard library information; A first judgment module compares the standard library information with the acquired new bid information to obtain a bid difference information set; A second judgment module, judging whether the new bid information meets the predetermined standard according to the bid difference information; If it meets the predetermined standards, the new bid information is judged to be qualified; If the predetermined standards are not met, the bid will be marked and corrected based on the difference information in the bid; The correction module obtains the correction content and uses the AI large model to determine whether the correction content meets the predetermined standards.

[0030] The above modules are the necessary foundation for realizing intelligent bid revision methods based on AI big models.

[0031] The present invention also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of the control method for automatic inspection of PCB appearance defects are implemented.

[0032] The present invention also provides a computer-readable storage medium storing a computer program, wherein the computer program is based on an intelligent bid revision method and revision system of an AI large model and the steps of the system.

[0033] Those skilled in the art will understand that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing the relevant hardware through a computer program. 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 memory, storage, database or other media provided in this application and used in the embodiments may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct RAM bus dynamic RAM (DRDRAM), and RAM bus dynamic RAM (RDRAM).

[0034] With the above-described preferred embodiments of the present invention as a guide, and with reference to the above description, relevant personnel are fully capable of making various changes and modifications without departing from the technical scope of this invention. The technical scope of this invention is not limited to the contents of the specification and must be determined according to the scope of the claims.

Claims

1. A method for revising tender documents based on intelligent AI large model, characterized in that: include: Obtain a certain amount of standard tender information, and train the AI model based on this information to obtain standard library information; Comparing the standard library information with the acquired new bid information to obtain a bid difference information set; Determining whether the new bid information meets the predetermined standards based on the bid difference information; If it meets the predetermined standards, the new bid information is judged to be qualified; If the predetermined standards are not met, the bid will be marked and corrected based on the difference information in the bid; Obtain the correction content, and perform an AI large model judgment based on the correction content to determine whether it meets the predetermined standards.

2. The method for revising a tender document based on intelligent AI large model according to claim 1 is characterized in that: The step of training the AI large model based on a certain amount of standard bidding document information to obtain standard library information includes: The classification analysis is performed based on a number of standard bidding document information to obtain homogeneous data groups and non-homogeneous data groups; An AI training model is obtained according to the homogeneous data acquisition process, and standard library information is obtained based on the AI training model.

3. The method for revising a tender document based on an intelligent AI large model according to claim 1, characterized in that: The step of performing classification analysis based on a number of standard tender information to obtain homogeneous data groups and heterogeneous data groups includes: Classify a certain amount of standard bidding information to obtain business information, technical information, quotation information and basic document information; Analyze the number of times business information, technical information, quotation information, and basic document information of different specifications appear in several standard bidding documents, and calculate the proportion of each information of different specifications in the total information; According to the size of the proportion value, a homogeneous data group and a non-homogeneous data group are divided; An AI training model is performed according to the acquisition process of the homogeneous data group, and standard library information is obtained based on the AI training model.

4. The method for revising a tender document based on an intelligent AI large model according to claim 1, characterized in that: The steps of marking and correcting based on the bid difference information include: Acquiring a plurality of bid document difference information to obtain a plurality of defect locations, performing color processing according to the plurality of defect locations, and automatically marking based on the color processing sections; Identify the automatic annotation information and classify it into manual modification requirement suggestions and automatic modification requirement suggestions, and automatically correct based on the automatic modification requirement suggestions and manually modify according to the manual modification requirements.

5. The method for revising a tender document based on intelligent AI large model according to claim 1, characterized in that: The step of using the AI big model to determine whether the correction content meets the predetermined standards includes: Perform AI big model judgment based on the correction content to obtain feedback information; The feedback information is compared with the predetermined standard frame by frame to determine whether the feedback information meets the requirements.

6. The intelligent tender document revision system based on AI large model according to claim 1 is characterized in that: include: The extraction module obtains a certain amount of standard bidding document information, and trains the AI large model based on the information to obtain the standard library information; A first judgment module compares the standard library information with the acquired new bid information to obtain a bid difference information set; A second judgment module judges whether the new bid information meets the predetermined standard based on the bid difference information; If it meets the predetermined standards, the new bid information is judged to be qualified; If the predetermined standards are not met, the bid will be marked and corrected based on the difference information in the bid; The correction module obtains the correction content and uses the AI large model to determine whether the correction content meets the predetermined standards.

7. 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 7 are implemented.

8. 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 5 are implemented.

Images