A multi-dimensional intelligent agent review method and system for flight review license reports

Abstract

The application discloses a kind of multi-dimension intelligent agent review methods and systems for flight evaluation license report, it is related to artificial intelligence technical field, the method includes: in response to report review request, obtain to be reviewed flight evaluation license report, text structure is analyzed to the to-be-reviewed flight evaluation license report by data extraction intelligent agent, to generate structured review dataset;By auditing intelligent agent, the structured review dataset is audited based on pre-constructed flight evaluation license report auditing standardization knowledge base, to obtain auditing result;By report generation intelligent agent, standardization auditing document is generated based on the auditing result, help to solve the problem that prior art cannot realize intelligent review flight evaluation license report.

Description

Technical Field

[0001] This invention relates to the field of artificial intelligence technology, and in particular to a multi-dimensional intelligent agent review method and system for air traffic assessment and permit reports. Background Technology

[0002] While my country's transportation administration system has achieved electronic and information-based processes, specialized approval areas such as navigation impact assessment permits still rely heavily on manual review, lacking sufficient intelligent capabilities. Navigation impact assessment reports are mostly unstructured data such as PDF text and engineering drawings. Existing systems can only store and browse these documents, unable to automatically parse and extract key parameters. Review relies on manual page-by-page verification and comparison with standards, resulting in low efficiency and a high risk of errors. Furthermore, each approval system is independently built, with inconsistent data standards and difficulties in cross-system integration. Application forms lack intelligent validation, leading to frequent revisions and high communication costs. Review standards vary significantly due to differences in personnel experience and expertise, resulting in inconsistent judgments for the same report, impacting the consistency and credibility of approvals.

[0003] Furthermore, data is not integrated between the business review and expert evaluation stages, opinions are not retained in a structured manner, and follow-up on rectification is difficult, making it hard to form a closed-loop management system. The traditional model cannot meet the needs of efficient, standardized, and unified approval. Therefore, it is urgent to introduce multi-agent collaborative technology to achieve structured parsing of reports, standardized review, and automated document generation, thereby improving the level of intelligence in the entire air traffic assessment review process. Summary of the Invention

[0004] In view of this, the present invention proposes a multi-dimensional intelligent agent review method and system for air assessment and permit reports, which can achieve data security isolation and privacy protection, and ensure the security and compliant use of data throughout the entire process.

[0005] To achieve the above objectives, the present invention provides the following technical solution: A multi-dimensional agent review method for air assessment and permit reports includes: In response to a report review request, the flight assessment permit report to be reviewed is obtained, and the data extraction agent performs text structured parsing on the flight assessment permit report to be reviewed to generate a structured review dataset; The auditing agent performs a preliminary audit of the structured review dataset based on a pre-built standardized knowledge base for reviewing flight assessment and permit reports, in order to obtain the audit results; The report-generating agent generates standardized audit documents based on the audit results.

[0006] Based on the above technical solution, the present invention can be further improved as follows: Optionally, the step of performing text-structured parsing of the airworthiness assessment permit report to be reviewed by a data extraction agent to generate a structured review dataset includes: The data extraction intelligent agent identifies the text content in the navigation assessment permit report to be reviewed, extracts the content of the text content to obtain core information, including hydrological parameters, quantitative data, waterway facility parameters, engineering design indicators and ecological impact information; The core information is categorized and labeled to obtain a structured review dataset.

[0007] Optionally, the auditing agent consists of a parameter calculation model, a logical reasoning model, an auditing rule model, and a formatted reward model; The quantitative data is verified for compliance through a parameter calculation model. The logic of the flight assessment and permit report to be reviewed is intelligently analyzed using a logical reasoning model. The completeness of the argumentation in the air assessment permit report to be reviewed is intelligently assessed using a logical reasoning model; The structured review dataset is verified for accuracy and completeness using an audit rule model. The preliminary review results are further optimized using the logical reasoning model and the formatted reward model to obtain the final review result.

[0008] Optionally, the compliance calculation and verification of the quantitative data through the parameter calculation model includes: The parameter difference is calculated using formula (1); Δ = PS Formula (1); In the formula, Δ is the parameter difference, P is the measured parameter value in the report, and S is the standard requirement value in the standard knowledge base; When Δ is greater than or equal to the preset value, it is considered compliant; when Δ is less than the preset value, it is considered non-compliant. The quantitative data includes navigable water level, pipeline burial depth, bridge elevation, and navigable clearance dimensions.

[0009] Optionally, the step of intelligently analyzing the logic of the flight assessment permit report to be reviewed through a logical reasoning model includes: The overall compliance score of the report is calculated using formula (2); L = (A / B)×W1+ (C / D)×W2-E×W3 Formula (2); In the formula, L is the overall compliance score of the report, A is the number of consistent data items in the report, B is the total number of data items to be verified, C is the number of items with complete argumentation chains, D is the total number of argumentation items required by the standard, E is the number of logical conflict points in the report, and W1, W2, and W3 are weighting coefficients, and W1+W2+W2=1. When L ≥ the preset threshold, the report logic is deemed compliant; when L < the preset threshold, the report logic is deemed defective.

[0010] Optionally, the step of intelligently assessing the completeness of the argumentation in the flight assessment permit report under review using a logical reasoning model includes: The completeness of the argument is determined by formula (3); Ψ= (U / V)-ε formula (3); In the formula, Ψ is the completeness judgment value of the argumentation, U is the number of technical points that have been demonstrated in the report, V is the total number of mandatory argumentation points required by the air assessment standard, and ε is the completeness correction coefficient. When Ψ≥ the preset judgment value, the judgment argument is complete; when Ψ< the preset judgment value, the judgment argument is missing.

[0011] Optionally, the multi-dimensional intelligent agent review method for air assessment and approval reports further includes: The standardized audit document, along with the associated historical comparison data, is sent to the business platform for push notification.

[0012] A multi-dimensional intelligent agent review system for air assessment and certification reports includes: A data extraction agent is used to respond to a report review request, obtain the flight assessment permit report to be reviewed, and perform text structured parsing on the flight assessment permit report to be reviewed to generate a structured review dataset; An intelligent review agent is used to conduct a preliminary review of the structured review dataset based on a pre-built standardized knowledge base for reviewing air assessment permit reports, in order to obtain the review results; A report-generating agent is used to generate standardized audit documents based on the audit results.

[0013] An electronic device includes a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor executes the computer program to implement the steps of the method described herein.

[0014] A non-transitory computer-readable storage medium having a computer program stored thereon, the computer program implementing the steps of the method when executed by a processor.

[0015] The present invention has the following advantages: The multi-dimensional intelligent agent review method for air assessment and permit reports in this invention completes the parsing, review and document generation of air assessment reports through multi-agent collaboration, realizing full-process automation of review and significantly improving review efficiency; relying on the standardized knowledge base for air assessment and permit report review to unify review standards and ensure consistent and accurate results; reducing reliance on manual labor and human error, reducing communication and time costs, and forming an efficient and standardized intelligent review closed loop. Attached Figure Description

[0016] For illustrative and not limiting purposes, the present invention will now be described in conjunction with embodiments and accompanying drawings, wherein: Figure 1 This is a flowchart illustrating the multi-dimensional intelligent agent review method for flight assessment and approval reports in an embodiment of the present invention. Figure 2 This is a schematic diagram of the main components of the multi-dimensional intelligent agent review system for air assessment and approval reports in an embodiment of the present invention; Figure 3 This is a schematic diagram of the physical structure of the electronic device provided by the present invention. Detailed Implementation

[0017] To enable those skilled in the art to better understand the present invention, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0018] It should be noted that the terms "first," "second," etc., in the specification and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be used interchangeably where appropriate for the embodiments of the invention described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0019] It should be noted that, where there is no conflict, the embodiments and features of the present invention can be combined with each other. The embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0020] Figure 1 This is a flowchart illustrating the multi-dimensional intelligent agent review method for flight assessment and approval reports according to an embodiment of the present invention. Figure 1 As shown, the multi-dimensional intelligent agent review method for air assessment and approval reports provided in this embodiment of the invention includes the following steps S101 to S103.

[0021] S101, in response to a report review request, obtains the flight assessment permit report to be reviewed, and performs text structured parsing of the flight assessment permit report to be reviewed through a data extraction agent to generate a structured review dataset.

[0022] The data extraction intelligent agent identifies the text content in the navigation assessment permit report to be reviewed, extracts the content of the text content to obtain core information, including hydrological parameters, quantitative data, waterway facility parameters, engineering design indicators and ecological impact information; The core information is categorized and labeled to obtain a structured review dataset.

[0023] S102, the auditing agent conducts a preliminary audit of the structured review dataset based on a pre-built standardized knowledge base for reviewing air assessment permit reports, in order to obtain the audit results.

[0024] The auditing agent consists of a parameter calculation model, a logical reasoning model, an auditing rule model, and a formatted reward model. The quantitative data is verified for compliance through a parameter calculation model. The logic of the flight assessment and permit report to be reviewed is intelligently analyzed using a logical reasoning model. The completeness of the argumentation in the air assessment permit report to be reviewed is intelligently assessed using a logical reasoning model; The structured review dataset is verified for accuracy and completeness using an audit rule model. The preliminary review results are further optimized using the logical reasoning model and the formatted reward model to obtain the final review result.

[0025] The compliance calculation and verification of the quantitative data through the parameter calculation model includes: The parameter difference is calculated using formula (1); Δ=PS Formula (1); In the formula, Δ is the parameter difference, P is the measured parameter value in the report, and S is the standard requirement value in the standard knowledge base; When Δ is greater than or equal to the preset value, it is considered compliant; when Δ is less than the preset value, it is considered non-compliant. The quantitative data includes navigable water level, pipeline burial depth, bridge elevation, and navigable clearance dimensions.

[0026] The required navigation clearance dimension for a certain bridge is S=7.0m, while the reported measured value is P=7.5m, and the preset threshold is 0. Δ=7.5-7.0=0.5m≥0, therefore it is deemed compliant.

[0027] The intelligent analysis of the logic of the flight assessment permit report under review through a logical reasoning model includes: The overall compliance score of the report is calculated using formula (2); L = (A / B)×W1+ (C / D)×W2-E×W3 Formula (2); In the formula, L is the overall compliance score of the report, A is the number of consistent data items in the report, B is the total number of data items to be verified, C is the number of items with complete argumentation chains, D is the total number of argumentation items required by the standard, E is the number of logical conflict points in the report, and W1, W2, and W3 are weighting coefficients, and W1+W1+W3=1. When L ≥ the preset threshold, the report logic is deemed compliant; when L < the preset threshold, the report logic is deemed defective.

[0028] The intelligent assessment of the completeness of the argumentation in the flight assessment permit report under review using a logical reasoning model includes: The completeness of the argument is determined by formula (3); Ψ=(U / V)-ε Formula (3); In the formula, Ψ is the completeness judgment value of the argumentation, U is the number of technical points that have been demonstrated in the report, V is the total number of mandatory argumentation points required by the air assessment standard, and ε is the completeness correction coefficient. When Ψ≥ the preset judgment value, the judgment argument is complete; when Ψ< the preset judgment value, the judgment argument is missing.

[0029] Let ε = 0.05, and the preset judgment value Ψ0 = 0.85.

[0030] Example: The proven key point U=22, the total number of key points that must be proven V=25Ψ=(22 / 25)-0.05=0.88-0.05=0.83<0.85 Judgment: The proof is incomplete.

[0031] The structured review dataset is verified for information accuracy using an audit rule model. R = (M / N) × 100%; In the formula: R is the information accuracy rate, M is the number of valid information items that have passed the verification in the structured review dataset, and N is the total number of information items to be verified in the structured review dataset.

[0032] When R ≥ the preset accuracy threshold, the information is considered accurate. When R < preset accuracy threshold, the judgment information is inaccurate / incorrect.

[0033] The preset accuracy threshold is 95%.

[0034] Example: Validated items M=38, total validated items N=40, R=(38 / 40)×100%=95%≥95% judgment: information is accurate.

[0035] The integrity of the structured review dataset is verified using an audit rule model. η=(K / T)×100% In the formula: η is the information completeness rate, K is the number of required fields provided in the structured review dataset, and T is the total number of required fields required by the flight assessment specifications.

[0036] Judgment rules: When η ≥ the preset integrity threshold, it is determined to be complete; When η < preset integrity threshold, it is judged as incomplete.

[0037] The preset integrity threshold is 98%.

[0038] Example: Required fields K=49 have been provided, total number of required fields T=50η=(49 / 50)×100%=98%≥98% Judgment: Information is complete.

[0039] The preliminary review results are further optimized using the logical reasoning model and the formatted reward model to obtain the final review result, including: First, the logical reasoning model performs consistency verification, conflict detection, and rationality deduction on the preliminary review results to eliminate parameter contradictions, logical loopholes, and argumentation gaps. Then, the formatted reward model, based on preset specifications, output formats, and interpretability requirements, structures and calibrates the review conclusions, problem lists, supporting clauses, and judgment reasons, eliminating vague expressions, strengthening supporting evidence, and unifying judgment criteria. After dual optimization, the final credible, traceable, and directly usable review results are obtained.

[0040] S103, the report generation agent generates standardized audit documents based on the audit results.

[0041] Leveraging the natural language understanding and generation capabilities of large-scale models, a human-machine collaborative intelligent review interface is constructed. Reviewers can quickly confirm field reports based on review results and supporting materials. Reviewers can ask professional questions to the intelligent agent via text input or voice interaction. The intelligent agent can access the standardized knowledge base and historical database for flight assessment and permit report review in real time, outputting accurate and standardized professional answers to assist human decision-making.

[0042] The multi-dimensional intelligent agent review method for air assessment and approval reports also includes: The standardized review documents, along with associated historical comparison data, are sent to the business platform for push notification. Reviewers can initiate real-time consultations with the intelligent agent regarding review questions through the feedback interface, ensuring efficient transmission of review results and timely response to questions, forming a seamless closed loop of "review-feedback-rectification". Warning information, tracking data, and handling records during the review process are uniformly archived and stored in the platform database for subsequent querying, statistical analysis, and model iteration.

[0043] Figure 2 This is a schematic diagram illustrating the main components of the multi-dimensional intelligent agent review system for flight assessment and approval reports according to an embodiment of the present invention. Figure 2 As shown, the multi-dimensional intelligent agent review system 1 for air assessment and approval reports provided in this embodiment of the invention includes a data extraction intelligent agent 10, an auditing intelligent agent 20, and a report generation intelligent agent 30.

[0044] Data extraction agent 10 is used to respond to a report review request, obtain the flight assessment permit report to be reviewed, and perform text structured parsing on the flight assessment permit report to be reviewed to generate a structured review dataset; The auditing agent 20 is used to conduct a preliminary audit of the structured review dataset based on a pre-built standardized knowledge base for flight assessment and permit report audits, in order to obtain the audit results; Report generating agent 30 is used to generate standardized audit documents based on the audit results.

[0045] Figure 3 This is a schematic diagram of the physical structure of an electronic device provided in an embodiment of the present invention, such as... Figure 3 As shown, the electronic device 40 includes: a processor 401, a memory 402, and a bus 403; The processor 401 and the memory 402 communicate with each other via the bus 403. The processor 401 is used to call program instructions in the memory 402 to execute the methods provided in the above-described method embodiments, and to execute the methods provided in the embodiments of the present invention.

[0046] This embodiment provides a non-transitory computer-readable storage medium that stores computer instructions, which cause a computer to execute the method provided in this embodiment of the invention.

[0047] Those skilled in the art will understand that all or part of the steps of the above method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it performs the steps of the above method embodiments. The aforementioned storage medium includes various storage media capable of storing program code, such as ROM, RAM, magnetic disk, or optical disk.

[0048] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can occur depending on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A multi-dimensional agent review method for a flight review license report, characterized in that, include: In response to a report review request, the flight assessment permit report to be reviewed is obtained, and the data extraction agent performs text structured parsing on the flight assessment permit report to be reviewed to generate a structured review dataset; The auditing agent performs a preliminary audit of the structured review dataset based on a pre-built standardized knowledge base for reviewing flight assessment and permit reports, in order to obtain the audit results; The report-generating agent generates standardized audit documents based on the audit results.

2. The multi-dimensional intelligent agent review method for air assessment and permit reports according to claim 1, characterized in that, The step of using a data extraction intelligent agent to perform text-structured parsing of the flight assessment permit report to be reviewed, in order to generate a structured review dataset, includes: The data extraction intelligent agent identifies the text content in the navigation assessment permit report to be reviewed, extracts the content of the text content to obtain core information, including hydrological parameters, quantitative data, waterway facility parameters, engineering design indicators and ecological impact information; The core information is categorized and labeled to obtain a structured review dataset.

3. The multi-dimensional intelligent agent review method for air assessment and permit reports according to claim 2, characterized in that, The auditing agent consists of a parameter calculation model, a logical reasoning model, an auditing rule model, and a formatted reward model. The quantitative data is verified for compliance through a parameter calculation model. The logic of the flight assessment and permit report to be reviewed is intelligently analyzed using a logical reasoning model. The completeness of the argumentation in the air assessment permit report to be reviewed is intelligently assessed using a logical reasoning model; The structured review dataset is verified for accuracy and completeness using an audit rule model. The preliminary review results are further optimized using the logical reasoning model and the formatted reward model to obtain the final review result.

4. The multi-dimensional intelligent agent review method for air assessment and permit reports according to claim 3, characterized in that, The compliance calculation and verification of the quantitative data through the parameter calculation model includes: The parameter difference is calculated using formula (1); Δ = PS Formula (1); In the formula, Δ is the parameter difference, P is the measured parameter value in the report, and S is the standard requirement value in the standard knowledge base; When Δ is greater than or equal to the preset value, it is considered compliant; when Δ is less than the preset value, it is considered non-compliant. The quantitative data includes navigable water level, pipeline burial depth, bridge elevation, and navigable clearance dimensions.

5. The multi-dimensional intelligent agent review method for air assessment and approval reports according to claim 3, characterized in that, The intelligent analysis of the logic of the flight assessment permit report under review through a logical reasoning model includes: The overall compliance score of the report is calculated using formula (2); L = (A / B)×W1+ (C / D)×W2-E×W3 Formula (2); In the formula, L is the overall compliance score of the report, A is the number of consistent data items in the report, B is the total number of data items to be verified, C is the number of items with complete argumentation chains, D is the total number of argumentation items required by the standard, E is the number of logical conflict points in the report, and W1, W2, and W3 are weighting coefficients, and W1+W2+W3=1. When L ≥ the preset threshold, the report logic is deemed compliant; when L < the preset threshold, the report logic is deemed defective.

6. The multi-dimensional intelligent agent review method for air assessment and permit reports according to claim 3, characterized in that, The intelligent assessment of the completeness of the argumentation in the flight assessment permit report under review using a logical reasoning model includes: The completeness of the argument is determined by formula (3); Ψ= (U / V)-ε formula (3); In the formula, Ψ is the completeness judgment value of the argumentation, U is the number of technical points that have been demonstrated in the report, V is the total number of mandatory argumentation points required by the air assessment standard, and ε is the completeness correction coefficient. When Ψ≥ the preset judgment value, the judgment argument is complete; when Ψ< the preset judgment value, the judgment argument is missing.

7. The multi-dimensional intelligent agent review method for air assessment and permit reports according to claim 3, characterized in that, The multi-dimensional intelligent agent review method for air assessment and approval reports also includes: The standardized audit document, along with the associated historical comparison data, is sent to the business platform for push notification.

8. A multi-dimensional intelligent agent review system for air traffic assessment and permit reports, characterized in that, include: A data extraction agent is used to respond to a report review request, obtain the flight assessment permit report to be reviewed, and perform text structured parsing on the flight assessment permit report to be reviewed to generate a structured review dataset; An intelligent review agent is used to conduct a preliminary review of the structured review dataset based on a pre-built standardized knowledge base for reviewing air assessment permit reports, in order to obtain the review results; A report-generating agent is used to generate standardized audit documents based on the audit results.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method as described in any one of claims 1 to 7.

10. A non-transitory computer-readable medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method as described in any one of claims 1 to 7.

Images