Technical standard intelligent assistant based on large model and recommendation method

Through the big model's Q&A data analysis and pre-processing methods, the problem of low efficiency in technical standards recommendation is solved, efficient technical standards analysis and recommendation are realized, and differentiated analysis processing is adapted to different needs, improving recommendation efficiency and reducing storage space requirements.

CN120297693AActive Publication Date: 2025-07-11ZHEJIANG ZHIKE ZHONGLIAN TECHNOLOGY CO LTD
View PDF 9 Cites 0 Cited by

Patent Information

Application Number
CN202510772116.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-06-11
Publication Date
2025-07-11
Estimated Expiration
2045-06-11

AI Technical Summary

Technical Problem

When using large models to recommend technical standards, there is a deviation in the number of technical standards related to different requirements, resulting in low recommendation processing efficiency and inability to effectively analyze and handle the problem of a large number of information items required by users.

Method used

Through the analysis of the question-and-answer data of the big model, the complex analysis of technical requirements is determined, and combined with the changes in the information items of the engineering personnel's demand, the pre-processing method is used to improve the recommendation efficiency, including the analysis method determination module, the analysis processing module and the recommendation processing module, which are used to determine and process the information items of the technical standards respectively.

Benefits of technology

It realizes efficient retrieval matching in the needs of complex analytical technology, reduces the storage space requirements of large models, improves the efficiency of recommendation processing, and adapts to differentiated analytical processing of different needs.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN120297693A_ABST
    Figure CN120297693A_ABST
Patent Text Reader

Abstract

The invention provides a technical standard intelligent assistant based on a large model and a recommendation method, and belongs to the technical field of data processing, and the recommendation method specifically comprises the following steps: when determining that the correlation degree between a technical standard and a complex analysis technical requirement meets a requirement according to calling analysis data, calling analysis data in different technical requirements according to the technical standard; the method comprises the following steps: determining associated technical requirements of a technical standard, so as to determine a pre-processing method of the technical standard according to a deviation condition of requirement data of information items of the technical standard in different associated technical requirements and composition data of the technical standard of a preset analysis mode, and when the technical standard is called and processed by the technical requirements of a large model, determining the pre-processing method of the technical standard. Recommendation processing is carried out by a technical standard preprocessing method, so that the forecasting processing efficiency of the regional water resource quantity is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the technical field of data processing, and particularly relates to a technical standard intelligent assistant and a recommendation method based on a large model. Background Art

[0002] In the process of designing and processing kitchen electric appliances, engineering R & D personnel often need to refer to technical standards in multiple dimensions to ensure that the design results can meet the requirements. Specifically, in the invention patent application CN202410292995.7, "A Software Development Method, Device, Equipment and Medium for Enterprise Management", the target code is optimized according to preset technical standards and the technical specifications of the target enterprise and / or customized and adjusted according to user customization requirements, so as to realize the intelligence and high efficiency of software development, thus greatly reducing the workload and time cost of developers and ensuring the reliability of the development results on this basis. However, there are the following technical problems: When using a large model for the recommendation processing of technical standards, there are deviations in the number of technical standards associated with different requirements. This leads to the situation that for requirements with a large number of associated technical standards and a large number of information items of user requirements in the technical standards, if the pre - analysis processing of technical standards cannot be carried out, the efficiency of the recommendation processing may be reduced. Therefore, how to generate a differentiated pre - analysis processing strategy for technical standards according to the association situation of technical standards in different requirements and the difficulty of parsing technical standards in different requirements, so as to improve the efficiency of the recommendation processing has become an urgent technical problem to be solved.

[0003] To solve the above - mentioned technical problems, the present application provides a technical standard intelligent assistant and a recommendation method based on a large model. Summary of the Invention

[0004] To achieve the purpose of the present invention, the present invention adopts the following technical solutions: Specifically, the present application provides a recommendation method, which specifically includes: S1 Determine the technical requirements of different engineering personnel based on the analysis results of the Q&A data of the large model. According to the number of technical standards associated with different technical requirements and in combination with the change situation of the information items of the engineering personnel's requirements in different technical standards, determine the complex parsing technical requirements in the technical requirements; S2 Obtain the call - parsing data of the technical standards in different complex parsing technical requirements. When it is determined that the association degree between the technical standards and the complex parsing technical requirements meets the requirements, proceed to the next step; S3 Determine the associated technical requirements of the technical standards according to the call - parsing data of the technical standards in different technical requirements; S4 determines the preprocessing method of the technical standard based on the deviation of the demand data of the information items of the technical standard in different related technical requirements and the composition data of the technical standard in the preset parsing mode. When the technical standard is called and processed by the technical requirements of the large model, the recommendation process is carried out using the preprocessing method of the technical standard.

[0005] The beneficial effects of the present invention are as follows: According to the call parsing data of the technical standard in different complex parsing technical requirements, it is determined whether the association degree between the technical standard and the complex parsing technical requirements meets the requirements, thereby realizing the call information items and call times in the complex parsing technical requirements, and evaluating the retrieval matching degree of the technical standard in the complex parsing technical requirements. Thus, it avoids the occurrence of the technical problem that the processing efficiency of complex parsing technical requirements is low due to the lack of preprocessing of the technical standard with a high retrieval matching degree, and improves the efficiency of the recommendation process.

[0006] Based on the deviation of the demand data of the information items of the technical standard in different related technical requirements and the composition data of the technical standard in the preset parsing mode, the preprocessing method of the technical standard is determined. It not only takes into account the changes in the demand data of the information items, but also considers the differences in the parsing processing difficulty of related technical requirements caused by the differences in the composition data of the technical standard in the preset parsing mode. It realizes the determination of technical standards with a high preprocessing requirement from multiple perspectives, and then realizes the differential preprocessing of different technical standards, improves the efficiency of the recommendation process, and at the same time reduces the storage space requirements of the large model.

[0007] A further technical solution is that the technical requirement is determined according to the matching result between the user's question-and-answer data and the preset keywords of different preset technical requirements. Specifically, the preset technical requirement with the largest number of matches is used as the technical requirement.

[0008] A further technical solution is that the technical standard associated with the technical requirement is determined according to the preset corresponding relationship between the technical requirement and the technical standard.

[0009] A further technical solution is that the change situation of the demand information items of the engineering personnel is determined according to the deviation situation of the demand information items of different engineering personnel under the technical requirement.

[0010] A further technical solution is that the method for determining the complex parsing technical requirement in the technical requirement is as follows: Based on the technical standard associated with the technical requirement, determine the number of associated technical standards of the technical requirement, and use it as the associated standard number; Based on the changes in the requirement information items of different engineers according to different associated technical standards, determine the technical standards for which the requirement information items of different engineers are all inconsistent, and use it as the requirement change technical standard; Based on the number of the associated standards and the number of the requirement change technical standards, determine whether the technical requirement is a complex analysis technical requirement.

[0011] A further technical solution is that when the number of the associated standards of the technical requirement is greater than the preset associated standard number threshold or the number of the requirement change technical standards is greater than the change standard number threshold, it is determined that the technical requirement is a complex analysis technical requirement.

[0012] A further technical solution is to perform recommendation processing with the preprocessing method of the technical standard, which specifically includes: Determine the matching requirement information items of the engineer with the technical requirement matched by the engineer's search keyword; According to the preset processing method of the technical standard, obtain the matching requirement information items of the technical standard; Based on the matching requirement information items of the technical standard, perform recommendation processing.

[0013] A further technical solution is that when the preset processing method of the technical standard is not required to perform preprocessing identification, when there are matching requirement information items, perform real-time parsing processing of the technical standard to obtain the matching requirement information items of the technical standard.

[0014] A further technical solution is that the matching requirement information items are the information items whose matching call times of the information items under the technical requirement are greater than the preset matching call times threshold.

[0015] In a second aspect, the present invention provides a technical standard intelligent assistant based on a large model, which specifically includes: A parsing method determination module, a parsing processing module, and a recommendation processing module; Wherein the parsing method determination module is responsible for determining the preprocessing methods of different technical standards; The parsing processing module is responsible for processing the information items of different technical standards based on the preprocessing method; The recommendation processing module is responsible for performing recommendation processing based on the information items of the technical standard.

[0016] Other features and advantages will be described in the subsequent description, and the purpose and other advantages of the present invention are realized and obtained by the structure specifically pointed out in the description and the drawings.

[0017] To make the above - mentioned objects, features and advantages of the present invention more obvious and understandable, the following specifically gives preferred embodiments and, in conjunction with the accompanying drawings, detailed descriptions are as follows. Description of the Drawings

[0018] By referring to the accompanying drawings and describing in detail its exemplary embodiments, the above - mentioned and other features and advantages of the present invention will become more obvious.

[0019] Figure 1 is a flowchart of a recommendation method; Figure 2 is a flowchart of a method for determining complex parsing technical requirements in technical requirements; Figure 3 is a flowchart for determining that the degree of association between technical standards and complex parsing technical requirements meets the requirements; Figure 4 is a flowchart of a method for determining a pre - processing method of technical standards; Figure 5 is a framework diagram of a technical standard intelligent assistant based on a large - model. Detailed Embodiments

[0020] To enable those skilled in the art of this technology to better understand the technical solutions in this specification, the following will, in conjunction with the accompanying drawings in the embodiments of this specification, clearly and completely describe the technical solutions in the embodiments of this specification. Obviously, the described embodiments are only a part of the embodiments of this specification, rather than all of the embodiments. Based on the embodiments of this specification, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the scope of protection of this specification.

[0021] When performing the recommendation processing of technical standards, according to the data of the technical requirements with relatively large parsing processing difficulty among the technical requirements associated with the technical standards, a differentiated pre - processing method is generated, thereby improving the recommendation processing efficiency of technical standards when users perform interactive processing with large - models.

[0022] When the number of associated technical standards is more than 5, or when there are more than 3 associated technical standards and the demand information items of different engineers are all different, then the technical requirements are determined as complex parsing technical requirements.

[0023] When the total number of matching calls in different complex parsing technical requirements is more than 200 times, then it is determined that the degree of association between technical standards and complex parsing technical requirements does not meet the requirements.

[0024] The technical requirements with the number of matching calls within the preset call range are used as associated technical requirements.

[0025] Determine the parsing influence factors under different associated technical requirements based on the average value of the deviation quantity of demand information items between different matching call times and the average value of the proportion of the number of technical standards that do not adopt the preset parsing mode in the associated technical standards of the associated technical requirements, and determine the preprocessing method of the technical standards based on the parsing influence factors.

[0026] Embodiment 1 As Figure 1 shown, the present application provides a recommendation method, which specifically includes: S1 Determine the technical requirements of different engineers based on the analysis results of the Q&A data of the large model. Determine the complex parsing technical requirements in the technical requirements according to the number of technical standards associated with different technical requirements and in combination with the change situation of the demand information items of the engineers in different technical standards. S2 Obtain the call parsing data of the technical standards in different complex parsing technical requirements. When it is determined that the association degree between the technical standards and the complex parsing technical requirements meets the requirements, proceed to the next step. S3 Determine the associated technical requirements of the technical standards according to the call parsing data of the technical standards in different technical requirements. S4 Determine the preprocessing method of the technical standards based on the deviation situation of the demand data of the information items of the technical standards in different associated technical requirements and the composition data of the technical standards of the preset parsing mode. When the technical standards are called and processed by the technical requirements of the large model, perform the recommendation process with the preprocessing method of the technical standards.

[0027] Further, the technical requirements are determined according to the matching results of the Q&A data of the user and the preset keywords of different preset technical requirements. Specifically, the preset technical requirement with the largest number of matches is used as the technical requirement.

[0028] Table 1 shows the corresponding relationship between technical requirements and technical standards.

[0029]

[0030] Further, the technical standards associated with the technical requirements are determined according to the preset corresponding relationship between the technical requirements and the technical standards.

[0031] It can be understood that the change situation of the demand information items of the engineers is determined according to the deviation situation of the demand information items of different engineers under the technical requirements.

[0032] It should be noted that as Figure 2 shown, the method for determining the complex parsing technical requirements in the technical requirements is: Determine the number of associated technical standards for the technical requirement based on the technical standards associated with the technical requirement, and use it as the number of associated standards; Determine the identification of the technical standards for which the requirement information items of different engineering personnel are all inconsistent according to the change situation of the requirement information items of different associated technical standards for different engineering personnel, and use it as the technical standards for requirement changes; Determine whether the technical requirement is a complex parsing technical requirement based on the number of associated standards and the number of technical standards for requirement changes.

[0033] Furthermore, when the number of associated standards for the technical requirement is greater than the preset associated standard number threshold or the number of technical standards for requirement changes is greater than the change standard number threshold, it is determined that the technical requirement is a complex parsing technical requirement.

[0034] Optionally, the method for determining the complex parsing technical requirement in the technical requirement is as follows: Determine the number of associated technical standards for the technical requirement based on the technical standards associated with the technical requirement, and use it as the number of associated standards; Determine the identification of the technical standards for which the requirement information items of different engineering personnel are all inconsistent according to the change situation of the requirement information items of different associated technical standards for different engineering personnel, and use it as the technical standards for requirement changes; Determine the parsing complexity value of the technical requirement based on the average value of the number of associated standards and the number of technical standards for requirement changes, and determine whether the technical requirement is a complex parsing technical requirement based on the parsing complexity value.

[0035] Furthermore, when the parsing complexity value of the technical requirement is greater than the preset complexity value threshold, it is determined that the technical requirement is a complex parsing technical requirement.

[0036] Furthermore, the called parsing data includes the requirement information items of engineering personnel under different technical requirements.

[0037] Specifically, as Figure 3 shown, determining that the degree of association between the technical standard and the complex parsing technical requirement meets the requirements specifically includes: Determine the matching call times of the technical standard under different complex parsing technical requirements based on the called parsing data; Determine the associated complex requirements in the complex parsing technical requirement according to the matching call times; Determine whether the degree of association between the technical standard and the complex parsing technical requirement meets the requirements based on the number of associated complex requirements.

[0038] Further, the number of matching calls is the number of matching calls that match the requirements of engineering personnel under the complex parsing technology requirements.

[0039] It should be noted that the associated complex requirements are complex parsing technology requirements with the number of matching calls greater than the preset matching call number threshold.

[0040] Further, when the number of the associated complex requirements is greater than the preset associated complex requirement number threshold, it is determined that the degree of association between the technical standard and the complex parsing technology requirements does not meet the requirements.

[0041] It can be understood that when the degree of association between the technical standard and the complex parsing technology requirements does not meet the requirements, the preset parsing mode is used for preprocessing the technical standard.

[0042] Optionally, determining that the degree of association between the technical standard and the complex parsing technology requirements meets the requirements specifically includes: Using the call parsing data to determine the number of matching calls of the technical standard under different complex parsing technology requirements; Determining the total number of matching calls according to the sum of the number of matching calls under different complex parsing technology requirements; Determining whether the degree of association between the technical standard and the complex parsing technology requirements meets the requirements through the total number of matching calls.

[0043] Further, when the total number of matching calls is greater than the preset total number of matching calls threshold, it is determined that the degree of association between the technical standard and the complex parsing technology requirements does not meet the requirements.

[0044] Optionally, determining that the degree of association between the technical standard and the complex parsing technology requirements meets the requirements specifically includes: S21 Using the call parsing data to determine the number of matching calls of the technical standard under different complex parsing technology requirements, and determining the total number of matching calls according to the sum of the number of matching calls under different complex parsing technology requirements; Optionally, in one of the embodiments, the above step S21 further includes the following three situations: Situation 1: When the total number of matching calls does not meet the requirements, that is, is greater than the threshold, it is determined that the degree of association between the technical standard and the complex parsing technology requirements does not meet the requirements; Case 2: When the total number of matching calls meets the requirements, determine the complex parsing technology requirements with matching call data according to the number of matching calls under different complex parsing technology requirements. When the complex parsing technology requirements with matching call data are less than the preset requirement quantity threshold, if the number of matching calls under different complex parsing technology requirements is less than the preset matching call number threshold, it is determined that the degree of association between the technical standard and the complex parsing technology requirements meets the requirements; Case 3: When the complex parsing technology requirements with matching call data are not less than the preset requirement quantity threshold or the number of matching calls under different complex parsing technology requirements is not all less than the preset matching call number threshold: If the number of complex parsing technology requirements with matching call number not less than the preset matching call number threshold is large, that is, greater than the threshold, it is determined that the degree of association between the technical standard and the complex parsing technology requirements does not meet the requirements.

[0045] S22 Determine the processing time effect value under different complex parsing technology requirements according to the number of matching calls under different complex parsing technology requirements and in combination with the number of technical standards associated with different complex parsing technology requirements; Exemplarily, in one of the embodiments, the processing time effect value under the complex parsing technology requirements is determined according to the product of the number of matching calls and the number of associated technical standards.

[0046] In a possible embodiment, when there are complex parsing technology requirements where the processing time effect value does not meet the requirements, that is, is greater than the threshold, it is determined that the degree of association between the technical standard and the complex parsing technology requirements does not meet the requirements; In addition, if there are no complex parsing technology requirements where the processing time effect value does not meet the requirements, when the processing time effect values under different complex parsing technology requirements are all less than the preset processing time effect value threshold, if the complex parsing technology requirements with matching call data are less than the preset requirement quantity threshold, it is determined that the degree of association between the technical standard and the complex parsing technology requirements meets the requirements; If the complex parsing technology requirements with matching call data are not less than the preset requirement quantity threshold, when the sum of the processing time effect values under different complex parsing technology requirements does not meet the requirements, that is, is greater than the threshold, it is determined that the degree of association between the technical standard and the complex parsing technology requirements does not meet the requirements, and only when the sum of the processing time effect values under different complex parsing technology requirements meets the requirements, the time effect value is determined.

[0047] S23 determines the timeliness impact value of the technical standard by means of the total number of matching calls and the processing timeliness impact values under different complex parsing technology requirements, and determines whether the degree of association between the technical standard and the complex parsing technology requirements meets the requirements based on the timeliness impact value.

[0048] In a possible embodiment, the impact weight values of different complex parsing technology requirements are determined according to the product of different total numbers of matching calls and a preset proportionality factor, and the timeliness impact value of the technical standard is determined based on the sum of the products of the impact weight values and the processing timeliness impact values under different complex parsing technology requirements.

[0049] Optionally, the value range of the timeliness impact value is between 0 and 1. When the timeliness impact value is greater than the preset timeliness impact threshold, in a possible embodiment, when it is greater than 0.6, it is determined that the degree of association between the technical standard and the complex parsing technology requirements does not meet the requirements.

[0050] Furthermore, the associated technical requirements of the technical standard are the technical requirements with the number of matching calls greater than the preset value of the number of matching calls.

[0051] Specifically, the technical standard of the preset parsing mode is to perform pre-parsing processing and storage on all information items in the technical standard.

[0052] It should be noted that the deviation situation of the required data of the information items is determined according to the number of deviations of the required information items between different numbers of matching calls under the associated technical requirements.

[0053] It can be understood that as Figure 4 shown, the method for determining the preprocessing method of the technical standard is as follows: Based on the deviation situation of the required data of the information items of the technical standard in the associated technical requirements, determine the number of deviations of the required information items between different numbers of matching calls in the associated technical requirements, and determine the deviation call times according to the number of deviations of the required information items from other numbers of matching calls; Determine the proportion of the number of technical standards that do not adopt the preset parsing mode in the technical standards associated with the associated technical requirements, and use it as the proportion of the screening type quantity; Based on the average value of the proportion of the deviation call times under different associated technical requirements in the number of matching calls of the associated technical requirements and the proportion of the screening type quantity, determine the parsing impact factor under different associated technical requirements, and determine the preprocessing method of the technical standard based on the parsing impact factor.

[0054] Furthermore, determining the preprocessing method of the technical standard based on the parsing impact factor specifically includes: When there is an associated technical requirement whose parsing impact factor is greater than the preset impact factor threshold, the preprocessing of the technical standard is performed according to the preset parsing mode; When there is no associated technical requirement whose parsing impact factor is greater than the preset impact factor threshold, it is determined whether the sum of the parsing impact factors of different associated technical requirements is greater than the preset impact factor threshold. If so, the preprocessing of the technical standard is performed using the second preset parsing mode. If not, no preprocessing is required.

[0055] It can be understood that the requirement information item is an information item that matches the requirements of the engineering personnel under the associated call requirement.

[0056] Furthermore, the deviation call times are the call times whose sum of the deviation quantities of the requirement information items from other matching call times is greater than the preset requirement deviation quantity threshold.

[0057] Specifically, performing the preprocessing of the technical standard using the second preset parsing mode specifically includes: Based on the preset value of the impact factor, determine the associated technical requirements whose parsing impact factor is greater than the preset value of the impact factor, and use them as the impact technical requirements; Based on the sum of the matching call times of the information items of the technical standard under different impact technical requirements being greater than the preset times threshold, determine the target information item, and perform pre-parse processing and storage on the target information item of the technical standard.

[0058] Optionally, the method for determining the preprocessing method of the technical standard is: S41 Determine the number of technical standards that have not adopted the preset parsing mode among the associated technical standards of the associated technical requirements, and combine the number of technical standards that have adopted the preset parsing mode under the associated technical requirements to determine the call delay impact value of the associated technical requirements; In a possible embodiment, the call delay impact value of the associated technical requirements is determined according to the preset weight value of the technical standards that have not adopted the preset parsing mode, the number of preset weight values of the technical standards that have adopted the preset parsing mode under the associated technical requirements, and.

[0059] Optionally, when the number of associated technical requirements of the technical standard is large, that is, greater than a certain quantity threshold, the preprocessing of the technical standard is performed according to the preset parsing mode. When the number of associated technical requirements of the technical standard meets the requirements, based on the matching call times of the technical standard under different associated technical requirements, when the total number of matching calls under different associated technical requirements is large, that is, greater than a certain number of calls, the preprocessing of the technical standard is performed according to the preset parsing mode; Further, if the total number of matching calls under different associated technical requirements meets the requirements, when there is an associated technical requirement with a call delay impact value that does not meet the requirements, i.e., is greater than the preset threshold, then the preprocessing of the technical standard is performed according to the preset parsing mode. When there is no associated technical requirement with a call delay impact value that does not meet the requirements, the determination of the information item requirement deviation value is then carried out.

[0060] S42 determines the deviation quantity of the required information items between different matching call times in the associated technical requirement based on the deviation situation of the required data of the information items of the technical standard in the associated technical requirement. According to the deviation quantity of the required information items between different matching call times and in combination with the matching call times of the technical standard in the associated technical requirement, the information item requirement deviation value of the associated technical requirement is determined; In the above steps, in a possible embodiment, when it is determined that there is no deviation call time according to the deviation quantity of the required information items of other matching call times, that is, there is no call time with a deviation quantity greater than the preset threshold, and the sum of the call delay impact values of different associated technical requirements is within the preset delay impact value range, then the preprocessing of the technical standard is performed according to the second preset parsing mode; In addition, if the sum of the call delay impact values of different associated technical requirements is not within the preset delay impact value range, when the sum of the deviation call times of different associated technical requirements is relatively large, that is, greater than the threshold, then the preprocessing of the technical standard is performed according to the preset parsing mode. When the sum of the deviation call times of different associated technical requirements is not large, when there is an associated technical requirement with an information item requirement deviation value that does not meet the requirements, that is, is greater than the preset deviation threshold, then the preprocessing of the technical standard is performed according to the preset parsing mode.

[0061] Among them, the information item requirement deviation value of the associated technical requirement in the above steps is determined according to the sum of the deviation weight coefficients corresponding to the deviation quantity of the required information items between different matching call times, where the deviation weight coefficient is determined according to the product of the deviation quantity and the preset proportional factor.

[0062] S43 determines the parsing impact factor under different associated technical requirements based on the call delay impact value and the information item requirement deviation value under different associated technical requirements, and determines the preprocessing method of the technical standard based on the parsing impact factor.

[0063] Further, recommendation processing is performed using the preprocessing method of the technical standard, specifically including: Determine the matching required information items of the engineer based on the technical requirements matched by the engineer's search keywords; Obtain the matching required information items of the technical standard according to the preprocessing method of the technical standard; Based on the matching requirement information items of the technical standard, perform recommendation processing.

[0064] Table 2 is the technical standard for reference in the common design process of range hoods

[0065] Specifically, when the preset processing method of the technical standard is that no preprocessing recognition is required, when there are matching requirement information items, the parsing processing of the technical standard is performed in real time to obtain the matching requirement information items of the technical standard.

[0066] It should be noted that the matching requirement information item is an information item whose matching call times of the information item under the technical requirement are greater than the preset matching call times threshold.

[0067] Embodiment 2 In a second aspect, as Figure 5 shown, the present invention provides a technical standard intelligent assistant based on a large model, specifically including: A parsing method determination module, a parsing processing module, and a recommendation processing module; Among them, the parsing method determination module is responsible for determining the preprocessing methods of different technical standards; The parsing processing module is responsible for processing the information items of different technical standards based on the preprocessing method; The recommendation processing module is responsible for performing recommendation processing based on the information items of the technical standard.

[0068] Each embodiment in this specification is described in a progressive manner. For the same or similar parts among the embodiments, reference can be made to each other. The key point of each embodiment is to illustrate the differences from other embodiments. In particular, for the embodiments of devices, equipment, and non-volatile computer storage media, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can refer to the partial description of the method embodiments.

[0069] The above describes specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims can be performed in a different order than in the embodiments and still achieve the desired result. Additionally, the processes depicted in the figures do not necessarily require the specific order or continuous order shown to achieve the desired result. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0070] The above description is only for one or more embodiments of this specification and is not intended to limit this specification. For those skilled in the art, various modifications and variations can be made to one or more embodiments of this specification. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of one or more embodiments of this specification shall be included within the scope of the claims of this specification.

Claims

1. A recommendation method, characterized in that Specifically include: Based on the analysis results of the Q&A data of the large model, determine the technical requirements of different engineering personnel. According to the number of technical standards associated with different technical requirements, and combined with the change situation of the requirement information items of engineering personnel in different technical standards, determine the complex parsing technical requirements in the technical requirements; Obtain the call parsing data of the technical standards in different complex parsing technical requirements. When it is determined that the degree of association between the technical standards and the complex parsing technical requirements meets the requirements, proceed to the next step; Based on the call parsing data of the technical standards in different technical requirements, determine the associated technical requirements of the technical standards; Based on the deviation situation of the requirement data of the information items of the technical standards in different associated technical requirements and the composition data of the technical standards in the preset parsing mode, determine the preprocessing method of the technical standards. When the technical standards are called and processed by the technical requirements of the large model, perform recommendation processing using the preprocessing method of the technical standards.

2. The recommendation method according to claim 1, wherein The technical requirements are determined according to the matching results between the Q&A data of the user and the preset keywords of different preset technical requirements. Specifically, the preset technical requirement with the largest number of matches is used as the technical requirement.

3. The recommendation method according to claim 1, characterized in that, The change situation of the requirement information items of the engineering personnel is determined according to the deviation situation of the requirement information items of different engineering personnel under the technical requirements.

4. The recommendation method according to claim 1, characterized in that, The technical standards associated with the technical requirements are determined according to the preset corresponding relationship between the technical requirements and the technical standards.

5. The recommendation method according to claim 1, wherein The method for determining the complex parsing technical requirements in the technical requirements is: Based on the technical standards associated with the technical requirements, determine the number of associated technical standards of the technical requirements and use it as the associated standard number; According to the change situation of the requirement information items of different associated technical standards for different engineering personnel, determine the technical standards where the requirement information items of different engineering personnel are all inconsistent, and use it as the requirement change technical standard; Based on the associated standard number and the number of requirement change technical standards, determine whether the technical requirements are complex parsing technical requirements.

6. The recommendation method according to claim 1, characterized in that, The technical standards in the preset parsing mode are to perform pre-parsing processing and storage on all information items in the technical standards.

7. The recommendation method according to claim 1, wherein The deviation situation of the requirement data of the information items is determined according to the deviation quantity of the requirement information items between different matching call times under the associated technical requirements.

8. The recommendation method according to claim 1, wherein Determining that the degree of association between the technical standards and the complex parsing technical requirements meets the requirements specifically includes: Based on the call parsing data, determine the matching call times of the technical standards under different complex parsing technical requirements; Based on the matching call times, determine the associated complex requirements in the complex parsing technical requirements; Based on the number of the associated complex requirements, determine whether the degree of association between the technical standards and the complex parsing technical requirements meets the requirements.

9. The recommendation method according to claim 1, wherein The method for determining the preprocessing method of the technical standards is: Based on the deviation situation of the demand data of the information items of the technical standards in the associated technical requirements, determine the deviation quantity of the demand information items between different matching call times in the said associated technical requirements, and determine the deviation call times according to the deviation quantity of the demand information items of other matching call times; Determine the proportion of the number of technical standards that do not adopt the preset parsing mode in the associated technical standards of the associated technical requirements, and use it as the proportion of the screening type quantity; Based on the mean value of the proportion of the deviation call times under different associated technical requirements in the matching call times of the said associated technical requirements and the proportion of the screening type quantity, determine the parsing influence factor under different associated technical requirements, and determine the preprocessing method of the said technical standards based on the said parsing influence factor.

10. A technical standard intelligent assistant based on a large model, adopting a recommendation method according to any one of claims 1-9, specifically including: A parsing method determination module, a parsing processing module, and a recommendation processing module; Wherein the said parsing method determination module is responsible for determining the preprocessing methods of different technical standards; The said parsing processing module is responsible for processing the information items of different technical standards based on the said preprocessing method; The said recommendation processing module is responsible for performing recommendation processing based on the information items of the technical standards.

Citation Information

Patent Citations

  • Flood control scheduling rule question and answer recommendation method and system based on large language model

    CN118069797A

  • Typical skill recommendation method and device based on large model, storage medium and electronic equipment

    CN118194880A

  • Question and answer processing method and system based on large model

    CN118364091A

  • Marketing recommendation system and method based on big data analysis

    CN118428992A

  • Vector search and large language model-based talent recommendation method and device and medium

    CN119417430A