Regulation clause adaptive reference processing method and processing program

Abstract

The invention discloses a regulation and clause adaptive reference processing method and a processing program, the processing method comprises the following steps: S1, reading and identifying a regulation and regulation system, S2, analyzing the regulation and regulation system into tree structured clause data and storing the tree structured clause data, the structured clause data comprises a regulation code, a parent code, a regulation title and regulation content, s3, setting an examination data set, and constructing a mapping relationship between each examination task in the examination data set and the structured clause data, and S4, analyzing keywords in the examination work form, matching the examination data set, and associating clause data corresponding to the examination tasks in the examination data set with corresponding items in the examination work form. According to the method, various regulations are subjected to tree structured analysis to form corresponding terms, meanwhile, the content in the review worksheet is analyzed and decomposed into the review data set, and then each review task in the review data set is mapped with the regulation terms, so that the convenience of reference of related regulations and regulations is effectively improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of data processing, and in particular relates to a method and procedure for processing adaptive references of regulations and clauses. Background Art

[0002] At present, in the traditional event compliance evaluation method, the compliance evaluation of the event is often based on the company's internal system and past experience. If the event does not comply with the company's internal system, it is considered that there is a compliance risk in the event, and the compliance risk is graded based on the degree of impact of the historical compliance event.

[0003] However, the traditional event compliance evaluation method only considers the problem of mismatch between the company's internal system and the actual event. When facing the audit of industrial products, such as drone products, there are many types and complicated relevant rules and regulations. The efficiency of manual comparison is too low. Specifically, personnel need to manually retrieve various rules and regulations documents to search and match them one by one with the naked eye, which greatly increases the time and labor costs. Moreover, there is no set of standards to measure the degree of matching of manual comparison. It is affected by subjective human factors, resulting in unstable matching results and a lot of repetitive work. Summary of the invention

[0004] The purpose of the present invention is to overcome the deficiencies of the prior art and provide a regulations processing system that automatically analyzes regulations and effectively improves the marking speed and accuracy of the regulations of the review form by establishing a mapping relationship.

[0005] The present invention is achieved through the following technical solutions:

[0006] A method for processing adaptive references of regulatory clauses, comprising the following steps:

[0007] S1 reads identification rules and regulations,

[0008] S2 parses the rules and regulations into tree-structured clause data and stores them, wherein the structured clause data includes a rule code, a parent code, a rule title and a rule content.

[0009] S3 sets the review data set and constructs a mapping relationship between each review task in the review data set and the structured clause data.

[0010] S4 analyzes the keywords in the review work form and matches the review data set, and associates the clause data corresponding to the review task in the review data set with the corresponding items in the review work form.

[0011] As one of the preferred solutions, the reading and identifying of rules and regulations includes the steps of periodically obtaining updated rules and regulations from a rules and regulations library or a predetermined website, and specially marking the modified or added new clause data.

[0012] As one of the preferred solutions, the method also includes a step of performing OCR recognition on non-text files.

[0013] As one of the preferred solutions, the analysis of the rules and regulations in S2 is to identify and analyze the structure and hierarchical paragraphs of the rules and regulations based on the beginning number information of the paragraphs in the rules and regulations.

[0014] As one of the preferred solutions, it also includes a mapping relationship updating step.

[0015] Compare the referenced clause information in the review worksheet with the associated information in step S4,

[0016] Statistical analysis review task updates, additions, deletions or changes to corresponding mappings, including:

[0017] If the update rate of the approved reference terms and associated information of the review task exceeds the predetermined ratio, the review task is redefined according to the work item in the review work form corresponding to the review task and a new mapping relationship is established;

[0018] If the terms of the same review task are added, deleted or modified more than the predetermined number of times, the mapping relationship is updated.

[0019] As one of the preferred solutions, it also includes a step of matching the review form with the review task in the review data or determining a new review task based on the semantic recognition of AI.

[0020] As one of the preferred options, the regulations are regulations applicable to unmanned aerial vehicles.

[0021] A processing procedure for executing the regulatory clause adaptive reference processing method.

[0022] The advantages and beneficial effects of the present invention are:

[0023] A method for adaptively citing regulations and clauses of the present invention analyzes various regulations into corresponding clauses in a tree-like structure, and simultaneously analyzes and decomposes the content in the review work form into a review data set, and then maps each review task in the review data set with the regulations and clauses, thereby effectively improving the convenience of citing relevant regulations and systems, and establishing associations based on a database with high robustness and reliability. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 This is a screenshot of part of the regulatory document "Airworthiness Standards for Transport Category Airplanes [CCAR-25-R4]".

[0025] Figure 2 It is the parsed content obtained by processing the regulatory documents.

[0026] Figure 3 It is the formatted regulations content.

[0027] Figure 4 A schematic diagram of the mapping relationship.

[0028] Figure 5 Automatically match schematics to create the "Fuel System Check" review business set.

[0029] Figure 6 The following is a schematic diagram showing the structure of some relevant structured clauses.

[0030] Figure 7 This is a schematic diagram of the correspondence between the final review business set and the structured terms after matching.

[0031] For ordinary technicians in this field, other relevant drawings can be obtained based on the above drawings without any creative work. DETAILED DESCRIPTION

[0032] In order to enable those skilled in the art to better understand the technical solution of the present invention, the technical solution of the present invention is further described below in conjunction with the accompanying drawings and specific embodiments.

[0033] A method for processing adaptive references to regulations and clauses of the present invention is mainly applicable to occasions where a large number of regulations and systems are required during the review. Different review items will be customized in the review work form according to actual product requirements. Each review item has one or more corresponding regulations and clauses as review standards or system support. For example, the Civil Aviation Administration of China publishes regulations on the application of unmanned aerial vehicles, and it is necessary to review them one by one according to many contents in the review work form. However, there are many regulations and systems, and they are updated and modified from time to time, which brings many inconveniences to the review work. In order to improve the efficiency of regulation processing and identification, and thus ensure the rapid and accurate review, the method for processing adaptive references to regulations and clauses of the present invention includes the following steps:

[0034] S1 reads and identifies rules and regulations, wherein the reading includes manual update, or automatic update from the rules and regulations library, or linking directly to the corresponding website to automatically crawl and update the system, and also includes the step of OCR recognition of non-text files to ensure the timely update of various rules and regulations;

[0035] S2 parses the rules and regulations into tree-structured clause data and stores them. The structured clause data includes the rule code, parent code, rule title and rule content. The rules and regulations are parsed and processed into tree-structured structured data and stored in the database so that the system can reference the clause content of the specific rule file. Figure 1 As shown in the figure, taking the regulatory document "Airworthiness Standards for Transport Category Aircraft [CCAR-25-R4]" as an example, after importing and analyzing it, the regulatory structure is automatically identified according to the paragraph beginning number information. For example, Section A and Section B are the first level; Article XX.X is the second level; (a), (b)... are the third level; (1) is the fourth level; (i), (ii)... are the fifth level, and so on); the identified content is stored in the service memory, including four columns: regulation code, parent code, regulation title, and regulation content. For example, Figure 2 Then, according to the parsed content in the content, SQL statements for Insert into are automatically generated in batches and inserted into the database. After the insertion is completed, the formatted content of the regulations can be viewed in the system; Figure 3 shown.

[0036] S3 sets the review data set, and constructs a mapping relationship between the review tasks in the review data set and the structured clause data. First, based on the review work resume review data set, the review data set is mainly a collection of review types. Taking drone review as an example, the review data set includes "flight area restrictions", "altimeter correction procedures", "operation within visual range" and other review tasks that auditors need to pay attention to during actual reviews. After the review data set is constructed, a mapping relationship with the structured clause data is established based on the review tasks therein, wherein one review task can correspond to multiple clause data, or multiple clause data can correspond to multiple review tasks.

[0037] S4 analyzes the keywords in the review work form and matches the review data set, and associates the clause data corresponding to the review task in the review data set with the corresponding items in the review work form. Because there are many rules and regulations, proactive matching can effectively and quickly obtain information on rules and regulations and speed up the review. The associated clause data can limit the name of the rules and regulations, the specific clauses involved in the rules and regulations, and the specific clause content, etc. as needed.

[0038] Specifically, when the review team reviews the project, they will create action items for the inspection items, that is, create review data sets, such as fuel system inspections, engines, propellers, noise, etc. Figure 5-7 As shown in the figure, taking the fuel system inspection as an example, after the creation of the action item is completed, the system will automatically associate the corresponding structured clauses based on the action item, that is, the keywords of the review task in the review data set, using semantic recognition and the results of large model training. Due to the large amount of clauses, manual selection is inefficient and prone to errors. However, this system can quickly and accurately associate all clauses related to this content in multiple regulations, greatly improving efficiency and accuracy.

[0039] Furthermore, in order to adapt to the update of the system, the reading and identification of rules and regulations includes the steps of regularly obtaining updated rules and regulations from the rules and regulations library or the predetermined website, and specially marking the modified or added new clause data. The special mark can be in the form of bold, color or underline, etc. After a new update or a predetermined time, such as 3 months or 1 year later, the special mark will be automatically cancelled. The special reminder can effectively promote the implementation and understanding of the new system.

[0040] Because the content of each review work form may be different or have different enforcement strengths, the regulatory clause adaptive reference processing method of the present invention also includes a mapping relationship update step.

[0041] Compare the referenced clause information in the review worksheet with the associated information in step S4,

[0042] Statistical analysis review task updates, additions, deletions or changes to corresponding mappings, including:

[0043] If the update rate of the approved reference terms and associated information of the review task exceeds a predetermined ratio, such as 70%, the review task is redefined according to the work item in the review work form corresponding to the review task and a new mapping relationship is established;

[0044] If the terms of the same review task are added, deleted or modified more than the predetermined number of times, the mapping relationship is updated.

[0045] Among them, when doing the review, the system automatically brings out the formatted content of the regulations and clauses based on the previously set mapping relationship. At the same time, the reviewers can delete or re-formulate new cited regulations and clauses on this basis. The approved reference clauses in the review work form are the actual reference clauses after the review is completed. The data in the review work form after actual use are used to update and modify the regulations and systems recommended by the above method. Among them, if the change is too large, the review task is identified as the content of a new review data set, and a new mapping relationship is established, which realizes the supplementary iteration and improvement of the review data set. If the same review task is done several times, such as deleting the same or several related clause data more than three times, the mapping relationship of the review task is updated. The adjustment can be changed based on the actual review trend, etc. After statistical analysis, it is adjusted and iterated into the review data set, which effectively improves the timeliness of the processing method and ensures the best policy status.

[0046] For the above-mentioned update system, a neural network algorithm can also be used for updating. After completing the sample library training of the initial data and the construction of the neural network, the review task of verifying the referenced clause information and the related information in step S4 is updated, and the corresponding mapping is added, deleted or changed. The statistical analysis result is used as the correction value for sample expansion and then merged with part of the original sample content as the new sample data of the above-mentioned neural network to iterate the neural network model, which can effectively improve the accuracy of the above-mentioned mapping.

[0047] In order to improve the response speed, it also includes the steps of matching the review form with the review task in the review data or determining the new review task based on the semantic recognition of AI. Since the review form is for auditors and the text expression may be different, the paragraph text of the review work form can also be semantically understood and stemmed before mapping the regulations. The steps are as follows:

[0048] The natural language processing (NLP) artificial intelligence technology is combined to analyze the syntax of the paragraph, that is, the dependency relationship between words in the sentence is used to represent the syntactic structure information of the words and the tree structure is used to represent the structure of the whole sentence; the subject-predicate structure analysis and the method of removing modifiers are combined to extract the stem, and the syntactic structure information includes the structural relationship between subject-predicate, verb-object, attributive-predicate, and the structure of the whole sentence includes subject-predicate-object, attributive-adverbial-complement. The paragraph classification model learning and training is based on deep learning of big data (existing mature deep learning) to automatically learn deep semantic and word order features, and meet the review task classification to cover 95% of the review data set. Paragraph parsing: Paragraph parsing includes three functions: word segmentation, dependency syntax analysis, and named entity recognition. By identifying the basic vocabulary (word segmentation) in the text string, the vocabulary is reorganized and annotated to obtain the word part of speech. Dependency syntax is used to identify named entities in order to identify entities with specific meanings in the review work form. Key information extraction: Based on named entity recognition, the main part of the paragraph syntax is identified, and the dependency relationship of the sentence is used to identify the corresponding structural information, so as to extract information from different types of regulations and clauses. Through systematic analysis of a large number of regulations, the relationship between the review work form and the clauses is finally summarized and established.

[0049] To sum up, the processing method of the present invention enables the regulations formatting system to systematically batch convert the management procedures, regulations and other pdf files issued by relevant national departments, and then realizes automatic recognition of the regulation structure according to the numbering information in the regulation content, and parses the structured data into a tree structure and stores it in the database, which can be used for administrative approval and review later to quickly quote the terms and contents of specific regulations.

[0050] Especially in the field of drones, with the rise of the low-altitude economy, drones and aircraft of various specifications and models will appear, which requires a lot of flight airworthiness review work. By structuring the regulations and analyzing the labels, we can automatically match the appropriate regulations according to different models and different application scenarios, such as flight tests, aircraft power tests, aircraft safety tests and other application scenarios. Structured data can cross different regulations and directly generate a matching checklist, that is, a review work form. At the same time, combined with big data processing capabilities and AI analysis capabilities, the recommendation and feedback optimization recommendation logic and label marking, as well as the automatic interpretation of application scenarios, analyze the characteristics of different models, and provide data analysis capabilities on the demand side. The dual efforts of technology can effectively improve the overall use effect.

[0051] The present invention is described above by way of example. It should be noted that, without departing from the core of the present invention, any simple deformation, modification or other equivalent replacement that can be made by those skilled in the art without inventive effort falls within the protection scope of the present invention.

Claims

1. A method for processing adaptive references of regulations, characterized in that: The following steps are included: S1 reads identification rules and regulations, S2 parses the rules and regulations into tree-structured clause data and stores them, wherein the structured clause data includes a rule code, a parent code, a rule title and a rule content. S3 sets the review data set and constructs a mapping relationship between each review task in the review data set and the structured clause data. S4 analyzes the keywords in the review work form and matches the review data set, and associates the clause data corresponding to the review task in the review data set with the corresponding items in the review work form.

2. The method for processing adaptive reference of regulations and clauses according to claim 1, characterized in that: The reading and identifying of rules and regulations includes the steps of periodically obtaining updated rules and regulations from a rules and regulations library or a predetermined website, and specially marking the modified or added new clause data.

3. The method for processing adaptive reference of regulations and clauses according to claim 1, characterized in that: It also includes the step of performing OCR recognition on non-text files.

4. The method for processing adaptive reference of regulations and clauses according to claim 1, characterized in that: The analysis of the rules and regulations in S2 is to identify and analyze the structure and hierarchical paragraphs of the rules and regulations based on the beginning number information of the paragraphs in the rules and regulations.

5. The method for processing adaptive reference of regulations and clauses according to claim 1, characterized in that: It also includes a mapping relationship update step, Compare the referenced clause information in the review worksheet with the associated information in step S4, Statistical analysis review task updates, additions, deletions or changes to corresponding mappings, including: If the update rate of the approved reference terms and associated information of the review task exceeds the predetermined ratio, the review task is redefined according to the work item in the review work form corresponding to the review task and a new mapping relationship is established; If the terms of the same review task are added, deleted or modified more than the predetermined number of times, the mapping relationship is updated.

6. The method for processing adaptive reference of regulations and clauses according to claim 5, characterized in that: It also includes steps for matching review forms with review tasks in review data or determining new review tasks based on AI's semantic recognition.

7. The method for processing adaptive reference of regulations and clauses according to claim 5, characterized in that: The regulations mentioned are those applicable to unmanned aerial vehicles.

8. A processing program for executing the method for processing adaptive reference of regulatory clauses as described in any one of claims 1 to 7.

Images