A compliance guide authoring system and method based on tagged regulations

Abstract

This invention relates to the technical field of structured data processing, and discloses a compliance guidance compilation system and method based on tagged regulations. The system autonomously and accurately performs denoising preprocessing of legal text data based on legal provision text information and legal provision text denoising algorithm matching information, achieving intelligent acquisition of denoised legal provision text data. It accurately identifies legal provision retrieval scenarios based on the text information characteristic of legal provision retrieval scenarios, combined with data analysis and scientifically stored standard requirement characteristic text information of legal provision retrieval scenarios. Based on the legal provision retrieval scenario identification information, combined with data analysis and the standard legal provision structured data compilation scheme information of legal provision retrieval scenarios established based on big data, it scientifically plans the legal provision structured data compilation scheme, realizing the scientific formulation of a regulatory tagged data structured data compilation strategy based on the requirements of regulatory retrieval scenarios.

Description

Technical Field

[0001] This invention relates to the technical field of structured data processing, specifically to a system and method for compiling compliance guidelines based on tagged regulations. Background Technology

[0002] The compilation of regulatory compliance guidelines involves transforming unstructured legal provisions into structured data that is understandable and computable by machines. The core technological approach is to construct a legal tagging system. Through in-depth analysis of the provisions, standardized and semantic tags are assigned to their chapters, clauses, and even specific legal elements. This process typically combines natural language processing with legal expert knowledge to transform text into structured data objects containing tags, attributes, and relationships. The resulting legal knowledge graph supports concept-based intelligent retrieval, provision association deduction, dynamic tracking of validity status, and automated compliance review, ultimately providing a high-quality underlying data foundation for the intelligent application of legal technology. Existing regulatory compliance guideline compilation processes cannot intelligently plan the compilation scheme of structured legal provisions based on the needs of legal provision retrieval scenarios, nor can they intelligently execute the compilation of legal provisions, thus reducing the intelligence and scientific nature of regulatory compliance guideline compilation.

[0003] Chinese invention patent CN101833545B, published on September 9, 2015, discloses a data indexing method in the digital resource processing process. This method involves scanning paper documents or opening existing electronic document images as a data package, performing necessary image processing, layout analysis, and adding indexing attributes to the layout frames. Through OCR recognition, the method automatically completes data indexing by utilizing the correspondence between the recognized text and the layout frames, outputting the index to which the text belongs and its position on the image. Based on this information, the image information can be correctly stored in the database, creating relevant retrieval information sources. In this invention, when processing document data, the user adds indexes simultaneously with the layout analysis process. The added indexes can be customized by the user, allowing for intuitive checking and modification of the text's index while verifying the recognition results. However, the above technical solutions cannot intelligently formulate data processing schemes and intelligently execute data processing tasks according to the needs of structured data processing. Summary of the Invention

[0004] (a) Technical problems to be solved To address the shortcomings of existing regulatory compliance guidance development processes, such as the inability to intelligently plan structured data compilation schemes for legal provisions based on legal provision retrieval scenarios and the lack of intelligent execution of legal provision compilation tasks, thus reducing the intelligence and scientific rigor of regulatory compliance guidance development, this new approach aims to achieve the following: intelligent preprocessing of legal provision data, accurate mining of legal provision retrieval scenario types, intelligent planning of structured data compilation schemes for legal provisions, efficient collection of structured data compilation tools for legal provisions, and autonomous and efficient execution of legal provision compilation tasks.

[0005] (II) Technical Solution This invention is achieved through the following technical solution: a method for compiling compliance guidelines based on labeling regulations, the method comprising the following steps: The process involves: collecting legal text information and text information describing the requirements of legal text retrieval scenarios; identifying and processing textual noise in the legal text data to obtain legal text noise identification information; matching legal text denoising algorithms to obtain legal text denoising algorithm matching information; performing textual data denoising preprocessing on the legal text to obtain legal text denoising data; identifying legal text retrieval scenarios based on the requirements of legal text retrieval scenarios to obtain legal text retrieval scenario identification information; planning a structured data compilation scheme for legal texts to obtain target legal text structured data compilation scheme planning information; collecting and processing structured data compilation tools for legal texts to obtain legal text structured data compilation tool information and executing the legal text structured data compilation operation.

[0006] Preferably, the following steps are taken: First, legal text information and text information representing the characteristics required for legal text retrieval scenarios are collected. Second, noise in the legal text data is identified and processed to obtain legal text noise identification information. Third, a matching process is performed on the legal text data denoising algorithm to obtain legal text denoising algorithm matching information. Fourth, the legal text data undergoes preprocessing for denoising to obtain denoised legal text data, including the following steps: Users can input the text format information of the target legal provision online through the data input box, and generate the legal provision text information. The text format includes any one or more of doc, pdf, docx and odt. Users can also input the text information describing their needs for the data retrieval scenario for the target legal provision online through the data input box, and generate the text information describing the needs for the legal provision retrieval scenario. Based on the legal text information, text data noise recognition processing of the legal text is performed to obtain a set of legal text noise recognition information; based on the set of legal text noise recognition information, text data denoising algorithm matching processing of the legal text is performed to obtain a set of legal text denoising algorithm matching information. Based on the legal text information and the matching information set of the legal text denoising algorithm, the legal text data is preprocessed for denoising to obtain denoised legal text data.

[0007] Preferably, the process of performing text data noise recognition processing on the legal provisions based on the legal provisions text information to obtain a legal provisions text noise recognition information set; and performing text data denoising algorithm matching processing on the legal provisions based on the legal provisions text noise recognition information set to obtain a legal provisions text denoising algorithm matching information set includes the following steps: The large language model is used to search and identify text editing noise and text layout noise in the legal provisions, and a set of legal provisions text noise identification information is generated. ,in Indicates the first The legal text noise identification information includes illegal characters, blanks, page numbers, watermarks, inconsistent full-width and half-width characters, mixed use of simplified and traditional characters, and typos; the large language model includes any one of Hunyuan, Xunfei Xinghuo, and Doubao. Based on a large language model and the aforementioned legal text noise recognition information set Noise recognition information for all legal provisions mentioned above. The program searches and processes text editing and layout noise information from legal provisions to generate a set of matching information for the legal provisions text noise reduction algorithm. ,in This indicates the noise identification information in the legal text. The corresponding legal text denoising algorithm matching information represents the text denoising algorithm running program data required for text noise processing of the target legal text. The legal text denoising algorithm matching information includes regular expression text denoising algorithm running program, string operation library running program, rule conversion text denoising algorithm running program, custom mapping table text denoising algorithm running program, and pre-trained language text denoising algorithm running program.

[0008] Preferably, the preprocessing of legal provision text data for denoising based on the legal provision text information and the matching information set of the legal provision text denoising algorithm to obtain denoised legal provision text data includes the following steps: Obtain the legal text information and the matching information set of the legal text denoising algorithm. ; The text data of the legal provisions is matched with the information set using the legal provisions text denoising algorithm. Matching information of all legal text text noise reduction algorithms mentioned above The corresponding text denoising algorithm runs to preprocess the text data for denoising and generates denoised legal text data.

[0009] Preferably, the legal provision retrieval scenario is identified and processed based on the legal provision retrieval scenario requirements information to obtain legal provision retrieval scenario identification information; the structured data compilation scheme of the legal provisions is planned and processed to obtain the target legal provision structured data compilation scheme planning information, including the following steps: Based on the legal provision retrieval scenario requirement feature text information and the legal provision retrieval scenario standard requirement feature text information matrix, legal provision retrieval scenario identification processing is performed to obtain legal provision retrieval scenario identification information; Based on the legal provision retrieval scenario identification information and the legal provision retrieval scenario standard legal provision structured data compilation scheme information matrix, the structured data compilation scheme planning process of the legal provisions is performed to obtain the target legal provision structured data compilation scheme planning information.

[0010] Preferably, the legal provision retrieval scenario identification process, based on the legal provision retrieval scenario requirement feature text information and the legal provision retrieval scenario standard requirement feature text information matrix, includes the following steps: Establish a text information matrix of standard requirements and features for legal provision retrieval scenarios. ,in Indicates the first The text information describes the standard requirements and features of legal provisions retrieval scenarios for different legal provision retrieval scenario types. These scenarios include conditional filtering, full-text keyword search, semantic intelligent retrieval, association graph retrieval, and clause-level precise positioning retrieval. The text information describes the standard requirements and features of legal provisions retrieval scenarios for different legal provision retrieval scenario types. The text information describing the requirements of the legal provision retrieval scenario is compared with the text information matrix describing the standard requirements of the legal provision retrieval scenario. The legal provisions retrieval scenario standard requirements feature text information described in the article Perform text information matching to search for text information that matches the standard requirement features of the legal provision retrieval scenario. The corresponding legal provision retrieval scenario type information is generated through data identification to form legal provision retrieval scenario identification information. The legal provision retrieval scenario identification information includes any one of the following: conditional filtering, full-text keyword search, semantic intelligent retrieval, association graph retrieval, and clause-level precise positioning retrieval.

[0011] Preferably, the process of planning and processing the structured data compilation scheme for legal provisions based on the legal provision retrieval scenario identification information and the standard legal provision structured data compilation scheme information matrix for the legal provision retrieval scenario includes the following steps: Establish a legal provision retrieval scenario standard, a structured data compilation scheme for legal provisions, and an information matrix. ,in Indicates the first This document provides information on standard legal provision structured data compilation schemes for different legal provision retrieval scenario types. The standard legal provision structured data compilation schemes represent the optimal schemes for compiling structured legal provisions for various legal provision retrieval scenario types. These schemes include strongly structured metadata + classification tags, full-text indexing + basic metadata, deep semantic vectors + knowledge enhancement, knowledge graph relationship networks, and clause-level metadata + tags. The document also includes information on data compilation steps and tools required for compiling structured legal provision data. Data compilation steps include one or more of the following: professional word segmentation, lexical processing, inverted index construction, vector storage and index configuration sorting, and graph data modeling and storage. Data compilation tools include one or more of Python, PDF parsing libraries, LabelStudio, Elasticsearch, Sentence-BERT, Milvus, Pinecone, DeepKE, and SPO triple extraction models. The legal provision retrieval scenario identification information is combined with the information matrix of the standard legal provision structure data compilation scheme for the legal provision retrieval scenario. Information on the legal provision retrieval scenario, standard legal provision structured data compilation scheme. Perform text matching for legal provision retrieval scenarios to search for the standard legal provision structure data compilation scheme information corresponding to the legal provision retrieval scenario identification information. The system generates a target legal provision structure data compilation scheme planning information, which represents the data compilation steps and data compilation tools required for compiling the optimal legal provision structure data for the target legal provision.

[0012] Preferably, the process of collecting and processing structured data compilation tools for legal provisions, obtaining information on these tools, and executing the structured data compilation task includes the following steps: Based on the data compilation tool information in the target legal provision structured data compilation scheme planning information, the software management platform downloads, collects, and processes the running program data of the legal provision structured data compilation tool, and generates a legal provision structured data compilation tool information matrix. ,in Indicates the collected number Information on tools for compiling structural data of legal provisions; the software management platform includes any one of Tencent Software Center, Huajun Software Park, and Pacific Download Center; The denoised legal text data is processed using a large language model, based on the data compilation steps and tool information in the target legal text structured data compilation scheme planning information. This controls the legal text structured data compilation tool information matrix. Information on the structured data compilation tools for legal provisions described in the document The corresponding data compilation tool runs in an orderly manner according to the data compilation steps, performing the text data structuring compilation of the target legal provisions.

[0013] A compliance guidance compilation system based on tagged regulations is provided to implement the aforementioned compliance guidance compilation method based on tagged regulations. The system includes a regulation data compilation preprocessing module, a regulation data compilation analysis module, and a regulation data compilation management module. The regulatory data compilation and preprocessing module includes a legal text acquisition unit, a legal text retrieval scenario requirement acquisition unit, a legal text noise identification unit, a legal text denoising algorithm matching unit, and a legal text denoising processing unit. The legal provision text acquisition unit acquires legal provision text information through a data input box; the legal provision retrieval scenario requirement acquisition unit acquires legal provision retrieval scenario requirement feature text information through a data input box; the legal provision text noise recognition unit performs text data noise recognition processing on the legal provision text based on the legal provision text information and in conjunction with a large language model to obtain legal provision text noise recognition information; the legal provision text denoising algorithm matching unit performs text data denoising algorithm matching processing on the legal provision text based on the legal provision text noise recognition information and in conjunction with a large language model to obtain legal provision text denoising algorithm matching information; the legal provision text denoising processing unit performs legal provision text data denoising preprocessing based on the legal provision text information and the legal provision text denoising algorithm matching information to obtain legal provision text denoising data; The regulatory data compilation and analysis module includes a storage unit for standard requirements and characteristics of legal text retrieval scenarios, a legal text retrieval scenario identification unit, a storage unit for a compilation scheme of standard legal text retrieval scenario structured data, and a compilation scheme formulation unit for legal text structured data. The legal provision retrieval scenario standard requirement feature information storage unit is used to store legal provision retrieval scenario standard requirement feature text information; the legal provision retrieval scenario identification unit performs legal provision retrieval scenario identification processing based on the legal provision retrieval scenario requirement feature text information and the legal provision retrieval scenario standard requirement feature text information to obtain legal provision retrieval scenario identification information; the legal provision retrieval scenario standard legal provision structured data compilation scheme storage unit is used to store legal provision retrieval scenario standard legal provision structured data compilation scheme information; the legal provision structured data compilation scheme formulation unit performs legal provision structured data compilation scheme planning processing based on the legal provision retrieval scenario identification information and the legal provision retrieval scenario standard legal provision structured data compilation scheme information to obtain target legal provision structured data compilation scheme planning information; The legal data compilation and management module includes a legal provision structure data compilation tool collection unit and a legal provision structure data compilation task execution unit; The legal provision structured data compilation tool collection unit, based on the target legal provision structured data compilation scheme planning information and in cooperation with the software management platform, collects and processes the legal provision structured data compilation tool information; the legal provision structured data compilation operation execution unit, based on the legal provision text noise reduction data, the target legal provision structured data compilation scheme planning information, and the legal provision structured data compilation tool information, and in cooperation with the large language model, executes the legal provision structured data compilation operation.

[0014] (III) Beneficial Effects This invention provides a system and method for developing compliance guidelines based on tagged regulations. It offers the following advantages: I. Accurately acquire legal text information and legal text retrieval scenario requirements through data input boxes, providing reliable data support for precise identification of legal text retrieval scenarios; intelligently identify textual noise in legal texts based on legal text information and in conjunction with a large language model; simultaneously, refine the matching of textual noise reduction algorithms for legal texts based on the identified noise information and in conjunction with the large language model, achieving adaptive intelligent selection of data noise reduction processing algorithms based on legal text noise, thus improving the quality of legal text information acquisition; autonomously and accurately perform legal text data noise reduction preprocessing based on legal text information and legal text noise reduction algorithm matching information, achieving efficient and intelligent acquisition of denoised legal text data, and improving the quality of structured data compilation based on tagged legal data.

[0015] Second, based on the textual information describing the characteristics of legal provision retrieval scenarios, combined with data analysis and scientific storage, the standard requirements for legal provision retrieval scenarios are accurately identified, enabling precise mining of legal provision retrieval scenario types based on big data. Furthermore, based on the identified legal provision retrieval scenario information, combined with data analysis and the standard legal provision structure data compilation scheme information established based on big data, the legal provision structure data compilation scheme is scientifically planned, enabling customized legal provision structure data compilation schemes based on legal provision retrieval scenarios. This also enables the scientific formulation of regulatory tagging data structure data compilation strategies based on the needs of regulatory retrieval scenarios, improving the intelligence and quality of digital application of regulations.

[0016] Third, based on the planning information of the target legal provision structure data compilation scheme and in conjunction with the software management platform, the structure data compilation tools for legal provisions are comprehensively and accurately collected. Based on the denoising data of legal provision texts, the planning information of the target legal provision structure data compilation scheme, and the information of the legal provision structure data compilation tools, and in conjunction with the large language model, the legal provision structure data compilation work is intelligently executed, realizing the intelligent and efficient compilation of legal provision structure data, realizing the scientific and rational establishment of digital legal data, and improving the efficiency and reliability of digital legal compilation based on tags. Attached Figure Description

[0017] Figure 1 A schematic diagram of a compliance guidance compilation system based on labeling regulations provided by the present invention; Figure 2 The flowchart illustrates a method for developing compliance guidelines based on labeling regulations, as provided by this invention. Detailed Implementation

[0018] 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 some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] An example of a compliance guidance compilation system and method based on labeling regulations is as follows: Example 1: Please see Figures 1-2 A method for developing compliance guidelines based on labeling regulations, comprising the following steps: The process involves: collecting legal text information and text information describing the requirements of legal text retrieval scenarios; identifying and processing textual noise in the legal text data to obtain legal text noise identification information; matching legal text denoising algorithms to obtain legal text denoising algorithm matching information; performing textual data denoising preprocessing on the legal text to obtain legal text denoising data; identifying legal text retrieval scenarios based on the requirements of legal text retrieval scenarios to obtain legal text retrieval scenario identification information; planning a structured data compilation scheme for legal texts to obtain target legal text structured data compilation scheme planning information; collecting and processing structured data compilation tools for legal texts to obtain legal text structured data compilation tool information and executing the legal text structured data compilation operation.

[0020] For further details, please refer to Figures 1-2 The process involves collecting legal text information and text information describing the characteristics required for legal text retrieval scenarios; identifying and processing textual noise in the legal text data to obtain legal text noise identification information; matching the textual data denoising algorithm of the legal text to obtain legal text denoising algorithm matching information; and performing textual data denoising preprocessing on the legal text to obtain legal text denoising data, including the following steps: Step 11: The user enters the text format information of the target legal provision online through the data input box, and generates the legal provision text information. The text format includes any one or more of doc, pdf, docx and odt. The user also enters the text information describing the user's needs for the data retrieval scenario for the target legal provision online through the data input box, and generates the text information describing the needs for the legal provision retrieval scenario. Step 12: Based on the textual information of the legal provisions, perform textual data noise recognition processing on the legal provisions to obtain a set of textual noise recognition information for the legal provisions; based on the set of textual noise recognition information for the legal provisions, perform textual data noise reduction algorithm matching processing on the legal provisions to obtain a set of textual noise reduction algorithm matching information for the legal provisions. Step 13: Based on the legal text information and the matching information set of the legal text denoising algorithm, perform denoising preprocessing on the legal text data to obtain denoised legal text data.

[0021] The process of identifying textual noise in legal provisions based on their textual information to obtain a set of textual noise identification information for legal provisions; and then performing textual noise reduction algorithm matching processing on the set of textual noise identification information to obtain a set of textual noise reduction algorithm matching information for legal provisions, includes the following steps: Step 121: Use a large language model to search and identify text editing noise and text layout noise in legal text information, and generate a set of legal text noise identification information. ,in Indicates the first The legal text noise identification information includes illegal characters, blanks, page numbers, watermarks, inconsistent full-width and half-width characters, mixed use of simplified and traditional characters, and typos; the large language model includes any one of Hunyuan, Xunfei Xinghuo, and Doubao. Step 122: Based on a large language model and combined with the noise recognition information set of legal text, identify the relevant information. Noise recognition information for all internal legal texts The program searches and processes text editing and layout noise information from legal provisions to generate a set of matching information for the legal provisions text noise reduction algorithm. ,in Represents noise identification information in legal texts. The corresponding legal text denoising algorithm matching information represents the text denoising algorithm running program data required for text noise processing of the target legal text. The legal text denoising algorithm matching information includes regular expression text denoising algorithm running program, string operation library running program, rule conversion text denoising algorithm running program, custom mapping table text denoising algorithm running program, and pre-trained language text denoising algorithm running program.

[0022] Based on the legal provision text information and the matching information set of the legal provision text denoising algorithm, the legal provision text data denoising preprocessing is performed to obtain the legal provision text denoising data, including the following steps: Step 131: Obtain the legal text information and the matching information set of the legal text denoising algorithm. ; Step 132: Match the text data in the legal provisions with the information set using a legal provisions text denoising algorithm. All legal text text noise reduction algorithm matching information The corresponding text denoising algorithm runs to preprocess the text data for denoising and generates denoised legal text data.

[0023] By accurately acquiring legal text information and legal text retrieval scenario requirements through data input boxes, reliable data support is provided for accurately identifying legal text retrieval scenarios. Based on legal text information and in conjunction with a large language model, intelligent noise recognition of legal text data is performed. Simultaneously, based on the legal text noise recognition information and in conjunction with the large language model, refined matching of legal text data denoising algorithms is performed, achieving adaptive intelligent selection of data denoising processing algorithms based on legal text noise, thus improving the quality of legal text information acquisition. Based on legal text information and legal text denoising algorithm matching information, autonomous and accurate preprocessing of legal text data denoising is performed, achieving efficient and intelligent acquisition of denoised legal text data, and improving the quality of structured data compilation based on tagged legal data.

[0024] For further details, please refer to Figures 1-2 Based on the requirements of legal provision retrieval scenarios, the legal provision retrieval scenarios are identified and processed to obtain legal provision retrieval scenario identification information; the structured data compilation scheme of legal provisions is planned and processed to obtain target legal provision structured data compilation scheme planning information, including the following steps: Step 21: Based on the text information of legal provision retrieval scenario requirements and the matrix of text information of legal provision retrieval scenario standard requirements, perform legal provision retrieval scenario identification processing to obtain legal provision retrieval scenario identification information; Step 22: Based on the legal provision retrieval scenario identification information and the legal provision retrieval scenario standard legal provision structure data compilation scheme information matrix, perform the legal provision structure data compilation scheme planning process to obtain the target legal provision structure data compilation scheme planning information.

[0025] Based on the textual information of legal provision retrieval scenario requirements and the matrix of standard legal provision retrieval scenario requirements, legal provision retrieval scenario identification processing is performed to obtain legal provision retrieval scenario identification information, including the following steps: Step 211: Establish a text information matrix of standard requirements and features for legal provision retrieval scenarios. ,in Indicates the first This document contains standard requirement feature text information for various legal provision retrieval scenarios, including: conditional filtering, full-text keyword search, semantic intelligent retrieval, association graph retrieval, and clause-level precise positioning retrieval. The standard requirement feature text information describes the standard legal provision data retrieval scenario requirements for different legal provision retrieval scenario types. Step 212: Combine the text information of legal provision retrieval scenario requirements with the matrix of standard legal provision retrieval scenario requirements. Chinese legal provisions retrieval scenario standard requirements characteristic text information Perform text information matching to search for text information that matches the standard requirements of legal provision retrieval scenarios. The corresponding legal provision retrieval scenario type information is generated through data identification to produce legal provision retrieval scenario identification information. The legal provision retrieval scenario identification information includes any one of the following: conditional filtering, full-text keyword search, semantic intelligent retrieval, association graph retrieval, and clause-level precise positioning retrieval.

[0026] Based on the legal provision retrieval scenario identification information and the information matrix of the standard legal provision structured data compilation scheme for the legal provision retrieval scenario, the structural data compilation scheme planning process for the legal provisions is carried out to obtain the target legal provision structured data compilation scheme planning information, including the following steps: Step 221: Establish a standard legal provision retrieval scenario and a structured data compilation scheme for legal provisions information matrix. ,in Indicates the first This document provides information on standard legal provision structured data compilation schemes for various legal provision retrieval scenarios. It also includes optimal schemes for different legal provision retrieval scenario types. These schemes utilize various methods, such as strongly structured metadata with classification tags, full-text indexing with basic metadata, deep semantic vectors with knowledge enhancement, knowledge graph relationship networks, and clause-level metadata with tags. The document further includes information on the data compilation steps and tools required for compiling the legal provision structured data. The data compilation steps include one or more of the following: professional word segmentation, lexical processing, inverted index construction, vector storage and index configuration sorting, and graph data modeling and storage. The data compilation tools include one or more of the following: Python, PDF parsing libraries, LabelStudio, Elasticsearch, Sentence-BERT, Milvus, Pinecone, DeepKE, and SPO triple extraction models. Step 222: Combine the legal provision retrieval scenario identification information with the information matrix of the standard legal provision structured data compilation scheme for the legal provision retrieval scenario. Information on standard legal provisions retrieval scenarios, structured data compilation schemes for legal provisions Perform text matching for legal provision retrieval scenarios to search for legal provision retrieval scenario identification information, and retrieve the standard legal provision structure data compilation scheme information corresponding to the legal provision retrieval scenario. It generates a target legal provision structure data compilation scheme planning information, which represents the data compilation steps and data compilation tools required for compiling the optimal legal provision structure data for the target legal provision.

[0027] Based on the textual information describing the characteristics of legal provision retrieval scenarios, combined with data analysis and scientific storage, this method accurately identifies legal provision retrieval scenarios, enabling precise mining of legal provision retrieval scenario types based on big data. Furthermore, based on the identified legal provision retrieval scenario information, combined with data analysis and the standard legal provision structure data compilation scheme information established based on big data, this method scientifically plans the legal provision structure data compilation scheme, enabling customized legal provision structure data compilation schemes based on legal provision retrieval scenarios. It also enables the scientific formulation of regulatory tagging data structure data compilation strategies based on the needs of regulatory retrieval scenarios, improving the intelligence and quality of digital application of regulations.

[0028] For further details, please refer to Figures 1-2 The process of collecting and processing structured data compilation tools for legal provisions, obtaining information about these tools, and executing structured data compilation tasks includes the following steps: Step 31: Based on the data compilation tool information in the target legal provision structured data compilation scheme planning information, download, collect, and process the running program data of the legal provision structured data compilation tool on the software management platform, and generate a legal provision structured data compilation tool information matrix. ,in Indicates the collected number Information on tools for compiling structural data of legal provisions; software management platforms include any one of Tencent Software Center, Huajun Software Park, and Pacific Download Center; Step 32: Using a large language model, the denoised legal text data is processed based on the data compilation steps and tools information in the target legal text structured data compilation scheme planning information, and the legal text structured data compilation tool information matrix is ​​controlled. Information on structural data compilation tools for Chinese legal provisions The corresponding data compilation tool runs in an orderly manner according to the data compilation steps, performing the text data structuring compilation of the target legal provisions.

[0029] Based on the planning information of the target legal provision structure data compilation scheme and in conjunction with the software management platform, the tool for compiling structure data of legal provisions is comprehensively and accurately collected. Based on the denoising data of legal provision text, the planning information of the target legal provision structure data compilation scheme, the information of the legal provision structure data compilation tool, and in conjunction with the large language model, the legal provision structure data compilation operation is intelligently executed, realizing the intelligent and efficient compilation of legal provision structure data, realizing the scientific and rational establishment of digital legal data, and improving the efficiency and reliability of digital legal compilation based on tags.

[0030] Example 2: Please see Figures 1-2 A compliance guidance compilation system based on tagged regulations is used to implement a method for compiling compliance guidelines based on tagged regulations. The system includes a regulatory data compilation preprocessing module, a regulatory data compilation analysis module, and a regulatory data compilation management module. The regulatory data compilation and preprocessing module includes a legal text collection unit, a legal text retrieval scenario requirement collection unit, a legal text noise identification unit, a legal text denoising algorithm matching unit, and a legal text denoising processing unit; The system comprises the following components: a legal provision text acquisition unit, which collects legal provision text information through a data input box; a legal provision retrieval scenario requirement acquisition unit, which collects text information with legal provision retrieval scenario requirement characteristics through a data input box; a legal provision text noise recognition unit, which performs text data noise recognition processing on the legal provisions based on the legal provision text information and in conjunction with a large language model, to obtain legal provision text noise recognition information; a legal provision text denoising algorithm matching unit, which performs text data denoising algorithm matching processing on the legal provisions based on the legal provision text noise recognition information and in conjunction with a large language model, to obtain legal provision text denoising algorithm matching information; and a legal provision text denoising processing unit, which performs legal provision text data denoising preprocessing based on the legal provision text information and legal provision text denoising algorithm matching information, to obtain legal provision text denoising data. The legal data compilation and analysis module includes a storage unit for standard requirements and characteristics of legal text retrieval scenarios, a legal text retrieval scenario identification unit, a storage unit for the compilation scheme of standard legal text retrieval scenario data, and a unit for formulating the compilation scheme of structured legal text retrieval data. The system includes: a legal provision retrieval scenario standard requirement feature information storage unit for storing text information on the standard requirement features of legal provision retrieval scenarios; a legal provision retrieval scenario identification unit for performing legal provision retrieval scenario identification processing based on the text information on the legal provision retrieval scenario requirement features and the text information on the standard requirement features of legal provision retrieval scenarios to obtain legal provision retrieval scenario identification information; a legal provision retrieval scenario standard legal provision structured data compilation scheme storage unit for storing information on the legal provision retrieval scenario standard legal provision structured data compilation scheme; and a legal provision structured data compilation scheme formulation unit for performing structured data compilation scheme planning processing on legal provisions based on the legal provision retrieval scenario identification information and the information on the standard legal provision retrieval scenario structured data compilation scheme to obtain target legal provision structured data compilation scheme planning information. The legal data compilation and management module includes a legal provision structure data compilation tool collection unit and a legal provision structure data compilation task execution unit; The legal provision structured data compilation tool collection unit, based on the target legal provision structured data compilation scheme planning information and in conjunction with the software management platform, collects and processes the legal provision structured data compilation tool information; the legal provision structured data compilation task execution unit, based on the legal provision text noise reduction data, the target legal provision structured data compilation scheme planning information, the legal provision structured data compilation tool information, and in conjunction with the large language model, executes the legal provision structured data compilation task.

[0031] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for developing compliance guidelines based on labeling regulations, characterized in that, The method includes the following steps: Legal text information and legal text retrieval scenario requirement text information are collected separately; noise in the legal text data is identified and processed to obtain legal text noise identification information; The text data denoising algorithm of the legal provisions is matched to obtain the matching information of the legal provisions text denoising algorithm; the text data denoising preprocessing of the legal provisions is performed to obtain the legal provisions text denoising data; Based on the requirements of legal provision retrieval scenarios, the legal provision retrieval scenarios are identified and processed to obtain legal provision retrieval scenario identification information; The structured data compilation scheme of legal provisions is planned and processed to obtain the planning information of the structured data compilation scheme of the target legal provisions; The structured data compilation tool for legal provisions is collected and processed to obtain information on the structured data compilation tool for legal provisions and to execute the structured data compilation task for legal provisions.

2. The method for compiling compliance guidelines based on labeling regulations according to claim 1, characterized in that: Legal text information and legal text retrieval scenario requirement text information are collected separately; noise in the legal text data is identified and processed to obtain legal text noise identification information; The text data denoising algorithm of the legal provisions is matched to obtain the matching information of the legal provisions text denoising algorithm; the text data denoising preprocessing of the legal provisions to obtain the legal provisions text denoising data includes the following steps: Users can input the text format information of the target legal provision online through the data input box, and the legal provision text information will be generated; users can also input the text information describing their needs for the data retrieval scenario for the target legal provision online through the data input box, and the legal provision retrieval scenario needs feature text information will be generated. Based on the legal text information, text data noise recognition processing of the legal text is performed to obtain a set of legal text noise recognition information; based on the set of legal text noise recognition information, text data denoising algorithm matching processing of the legal text is performed to obtain a set of legal text denoising algorithm matching information. Based on the legal text information and the matching information set of the legal text denoising algorithm, the legal text data is preprocessed for denoising to obtain denoised legal text data.

3. The method for compiling compliance guidelines based on labeling regulations according to claim 2, characterized in that: The process of performing textual noise recognition processing on the legal provisions based on the legal provisions textual information to obtain a set of legal provisions textual noise recognition information, and then performing textual denoising algorithm matching processing on the legal provisions textual noise recognition information set to obtain a set of legal provisions textual denoising algorithm matching information, includes the following steps: The large language model is used to search and identify text editing noise and text layout noise in the legal provisions, and a set of legal provisions text noise identification information is generated. The include ;in Indicates the first Information on noise recognition in legal texts; Based on the large language model and the above All internal descriptions The program searches and processes text editing and layout noise information from legal provisions to generate a set of matching information for the legal provisions text noise reduction algorithm. The include ;in Indicates the The corresponding legal text noise reduction algorithm matches the information.

4. The method for compiling compliance guidelines based on labeling regulations according to claim 3, characterized in that: Based on the legal provision text information and the matching information set of the legal provision text denoising algorithm, the legal provision text data denoising preprocessing is performed to obtain legal provision text denoising data, including the following steps: Obtain the legal text information and the ; The text data of the legal provisions is processed through the... All of the above The corresponding text denoising algorithm runs to preprocess the text data for denoising and generates denoised legal text data.

5. The method for compiling compliance guidelines based on labeling regulations according to claim 4, characterized in that: Based on the requirements of legal provision retrieval scenarios, the legal provision retrieval scenarios are identified and processed to obtain legal provision retrieval scenario identification information; the structured data compilation scheme of legal provisions is planned and processed to obtain target legal provision structured data compilation scheme planning information, including the following steps: Based on the legal provision retrieval scenario requirement feature text information and the legal provision retrieval scenario standard requirement feature text information matrix, legal provision retrieval scenario identification processing is performed to obtain legal provision retrieval scenario identification information; Based on the legal provision retrieval scenario identification information and the legal provision retrieval scenario standard legal provision structured data compilation scheme information matrix, the structured data compilation scheme planning process of the legal provisions is performed to obtain the target legal provision structured data compilation scheme planning information.

6. The method for compiling compliance guidelines based on labeling regulations according to claim 5, characterized in that: Based on the legal provision retrieval scenario requirement feature text information and the legal provision retrieval scenario standard requirement feature text information matrix, legal provision retrieval scenario identification processing is performed to obtain legal provision retrieval scenario identification information, including the following steps: Establish a text information matrix of standard requirements and features for legal provision retrieval scenarios. The include ;in Indicates the first The standard requirements and textual information of legal provisions retrieval scenarios corresponding to various legal provisions retrieval scenario types; The legal provisions retrieval scenario requirement feature text information and the... The above Perform text information matching to search for text information that matches the characteristics of the legal provision retrieval scenario. The corresponding legal provision retrieval scenario type information is used to generate legal provision retrieval scenario identification information through data identification.

7. The method for compiling compliance guidelines based on labeling regulations according to claim 6, characterized in that: Based on the legal provision retrieval scenario identification information and the legal provision retrieval scenario standard legal provision structured data compilation scheme information matrix, the legal provision structured data compilation scheme planning process is performed to obtain the target legal provision structured data compilation scheme planning information, including the following steps: Establish a legal provision retrieval scenario standard, a structured data compilation scheme for legal provisions, and an information matrix. The include ;in Indicates the first Information on the compilation scheme of standard legal provisions structured data for various legal provision retrieval scenarios; The legal provision retrieval scenario identification information and the... The above Perform text matching for legal provision retrieval scenarios to search for the legal provision retrieval scenario identification information corresponding to the provided text. It also generates structural data for the target legal provisions, and plans the compilation of such data.

8. The method for compiling compliance guidelines based on labeling regulations according to claim 7, characterized in that: The process of collecting and processing structured data compilation tools for legal provisions, obtaining information about these tools, and executing structured data compilation tasks includes the following steps: Based on the data compilation tool information in the target legal provision structured data compilation scheme planning information, the software management platform downloads, collects, and processes the running program data of the legal provision structured data compilation tool, and generates a legal provision structured data compilation tool information matrix. The include ;in Indicates the collected number Information on tools for compiling structural data of individual legal provisions; The denoised legal text data is processed using a large language model, based on the data compilation steps and tools information from the target legal text structure data compilation scheme planning information. This process controls the... The above The corresponding data compilation tool runs in an orderly manner according to the data compilation steps, performing the text data structuring compilation of the target legal provisions.

9. A compliance guidance compilation system based on label-based regulations, used to implement the compliance guidance compilation method based on label-based regulations as described in any one of claims 1-8, characterized in that: The system includes a regulatory data compilation preprocessing module, a regulatory data compilation analysis module, and a regulatory data compilation management module. The aforementioned regulatory data compilation and preprocessing module is used to collect legal text information and legal text retrieval scenario requirement feature text information; and to identify and process the text data noise of legal texts to obtain legal text noise identification information; The text data denoising algorithm of the legal provisions is matched to obtain the matching information of the legal provisions text denoising algorithm; the text data denoising preprocessing of the legal provisions is performed to obtain the legal provisions text denoising data; The legal data compilation and analysis module identifies and processes legal text retrieval scenarios based on the requirements of legal text retrieval scenarios, thereby obtaining legal text retrieval scenario identification information; it also plans and processes the structured data compilation scheme of legal texts, thereby obtaining target legal text structured data compilation scheme planning information. The legal data compilation and management module is used to collect and process the structured data compilation tools for legal provisions, obtain information on the structured data compilation tools for legal provisions, and execute the structured data compilation work for legal provisions.

Images