Intelligent analysis and entry method and system for structured rule data, and medium

Abstract

The invention relates to an intelligent analysis and input method and system for structured rule data and a medium, and belongs to the technical field of structured data receiving. The method comprises the following steps: calling a pre-trained large model on the basis of to-be-maintained structured rule data selected and input by a user; reasoning the to-be-maintained structured rule data through the pre-trained large model to obtain the reasoned structured rule data; performing verification and filtering identification processing on the reasoned structured rule data; and packaging the processed data into structured rule data for page maintenance, and outputting the structured rule data. According to the method, through application of the pre-trained large model, the data processing speed is greatly improved, the pre-trained large model can quickly identify OriginalText text information, data classification and object analysis are automatically completed, and the processing time of a rule freight rate is greatly shortened to several minutes.

Description

Technical Field

[0001] This invention belongs to the field of structured data retrieval technology, and specifically relates to a method, system, and medium for intelligent parsing and inputting structured rule data. Background Technology

[0002] In the field of international freight rate publication, the DIT (International Freight Rate Input Front-End) system serves as the core operating platform, with its StructuredRule module undertaking the crucial function of maintaining structured rule data. This module needs to perform refined maintenance on each structured rule data under various scenarios (including Refund, Change, No Show, General, and OtherInfo for multilingual information management). These freight rate maintenance scenarios collectively constitute the foundational data system for the effectiveness and transmission of international freight rate rules.

[0003] However, the current maintenance of structured rule data mainly relies on manual maintenance, which suffers from inefficiency and quality risks. On the one hand, the manual maintenance process is cumbersome and time-consuming. Business users need to verify the correspondence between the OriginalText text and the freight rates under various conditions one by one, manually enter the rule parameters for Refund, Change, and other situations, and at the same time, manually translate and enter multilingual text for OtherInfo nodes. The complete maintenance of a single data entry often consumes a lot of manpower, making it difficult to adapt to the high-frequency update requirements of international freight rate business. On the other hand, manual operation is prone to data deviation. Since the OriginalText text often contains complex descriptions of freight rate conditions (such as change fees, validity period restrictions, etc.), and multilingual translations must strictly follow industry terminology standards, semantic misunderstandings, parameter errors, and translation inconsistencies are prone to occur during manual judgment and entry, thus affecting the accuracy and compliance of freight rate rules. One existing preprocessing method and system for rule-based fares, for example, processes physically dispersed but business-related rule-based fares data. This advances the large-scale querying, matching, and filtering work of the real-time processing stage to the preprocessing stage, thereby minimizing duplicate queries of rule-based fares without sacrificing computational accuracy, thus improving the system's real-time processing efficiency. However, this method still suffers from cumbersome and complex procedures, requiring significant time and effort from staff, and is prone to human error. Summary of the Invention

[0004] To address the above problems, this invention provides a method and apparatus for intelligent parsing and inputting structured rule data.

[0005] The first objective of this invention is to provide a method for intelligent parsing and inputting structured rule-based data, comprising: Based on the structured rule data to be maintained selected and entered by the user, the pre-trained large model is invoked; The pre-trained large model is used to infer the structured rule data to be maintained, resulting in inferred structured rule data. The structured rule data after reasoning is validated and filtered for identification. The processed data is encapsulated into structured rule data maintained by the page and then output.

[0006] In a specific embodiment of the present invention, the structured rule data to be maintained is text information in the structured rule data or multilingual maintenance node text in the structured rule data.

[0007] In a specific embodiment of the present invention, the step of reasoning through the pre-trained large model on the structured rule data to be maintained to obtain the reasoned structured rule data includes: The pre-trained large model is used to parse the structured rule data to be maintained, and to enhance the weighting of keywords; The large model performs retrieval and matching in the knowledge base based on the extracted keywords; The large model fills the matching results into the structured rule data to be maintained, resulting in the inferred structured rule data.

[0008] In a specific embodiment of the present invention, the knowledge base includes a rule-based translation dataset and a multilingual translation dataset.

[0009] In a specific embodiment of the present invention, the construction of the rule translation dataset includes: Extracting entity-relation-entity tripartite knowledge from historical data; The extracted ternary knowledge is stored in a structured manner to form a rule-based translation dataset.

[0010] The second objective of this invention is to provide a structured rule data intelligent parsing and input system, including an intelligent interaction module, a model reasoning module, a data parsing service module, and a data intelligent maintenance module; The intelligent interaction module is used to call the pre-trained large model based on the structured rule data to be maintained selected and entered by the user. The model inference module is used to infer the structured rule data to be maintained through the pre-trained large model, and obtain the inferred structured rule data. The data parsing service module is used to verify, filter, and identify the inferred structured rule data; and to encapsulate the processed data into structured rule data maintained on the page. The data intelligent maintenance module is used to output structured rule data encapsulated into page maintenance.

[0011] In a specific embodiment of the present invention, the structured rule data intelligent parsing and input system further includes an input front-end page; The data entry front-end page is used to accept user-selected data for confirmation or re-maintenance.

[0012] In a specific embodiment of the present invention, the data entry front-end page is also used to accept user-selected data confirmation or re-maintenance.

[0013] In a specific embodiment of the present invention, the intelligent interaction module is further used to send the structured rule data to be maintained to the model inference module; The model inference module is also used to receive the structured rule data being maintained, and to pass the inferred structured rule data to the data parsing service module. The data parsing service module is also used to receive the inferred structured rule data and send the encapsulated page maintenance structured rule data to the data intelligent maintenance module. The data intelligent maintenance module is also used to receive encapsulated structured rule data for page maintenance and output and display it on the input front-end page.

[0014] A third object of the present invention is to provide an electronic device comprising: a processor coupled to a memory; The memory is used to store computer programs; The processor is configured to execute the computer program stored in the memory, so that the electronic device performs the method.

[0015] A fourth objective of the present invention is to provide a computer-readable storage medium, characterized in that the computer-readable storage medium stores a program or instructions that, when the program or instructions are run on a computer, cause the computer to perform the method described thereon.

[0016] A fifth objective of this invention is to provide a computer program product comprising a computer program / instructions, characterized in that, when the computer program / instructions are executed by a processor, they implement the method described herein.

[0017] The beneficial effects of this invention are: The present invention provides a structured rule data intelligent parsing and input method, system, and medium. By applying a pre-trained large model, the data processing speed is greatly improved. The pre-trained large model can quickly identify OriginalText information and automatically complete data classification and object parsing, significantly reducing the processing time of a rule fare to a few minutes.

[0018] In terms of data quality, the pre-trained model is trained on a large amount of data and has accurate semantic analysis and pattern recognition capabilities. It can accurately identify the correspondence between OriginalText and rule categories in various situations, and between OtherInfo and multilingual translations.

[0019] Moreover, this invention not only achieves automatic maintenance of structured data through the data entry system, but also automatically populates and displays the maintained structured data. Staff only need to supervise and review, reducing labor costs, improving work stability and reliability, and minimizing the impact of human factors on the data. In addition, this solution optimizes the user experience and improves customer satisfaction.

[0020] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures pointed out in the description, claims, and drawings. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 A flowchart of a method for intelligent parsing and inputting structured rule data according to an embodiment of the present invention is shown; Figure 2 A framework diagram of a structured rule data intelligent parsing and input system according to an embodiment of the present invention is shown; Figure 3 A flowchart of a method for entering structured rules for freight rates according to an embodiment of the present invention is shown; Figure 4 A frame diagram of an electronic device according to an embodiment of the present invention is shown; In the diagram: 10, Intelligent Selection Interaction Module; 20, Model Inference Module; 30, Data Parsing Service Module; 40, Data Intelligent Maintenance Module; 300, Electronic Equipment; 301, Processor; 302, Memory. Detailed Implementation

[0023] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, 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, 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.

[0024] The technical terms and their explanations used in the following examples are detailed in Table 1.

[0025] Table 1

[0026] like Figure 1 Therefore, according to certain embodiments of the present invention, a method for intelligent parsing and inputting structured rule data includes: S1. Based on the structured rule data to be maintained selected and entered by the user, call the pre-trained large model; S2. The pre-trained large model is used to infer the structured rule data to be maintained, and the inferred structured rule data is obtained. S3. Perform verification, filtering, and identification processing on the inferred structured rule data; S4. Encapsulate the processed data into structured rule data maintained by the page and output it.

[0027] Analysis of existing maintenance data reveals a correlation between the OriginalText and fare maintenance under Refund, Change, NoShow, and General conditions. Similarly, the text entered by the same user when maintaining OtherInfo corresponds to the fare maintenance in the translated multilingual cases. Therefore, in some embodiments of this invention, the structured rule data to be maintained is either text information within the structured rule data (hereinafter referred to as OriginalText) or multilingual maintenance node text within the structured rule data (hereinafter referred to as OtherInfo).

[0028] In some embodiments of the present invention, step S2 includes: S2-1. The pre-trained large model is used to parse the structured rule data to be maintained, and the keywords are weighted. S2-2, The large model performs retrieval and matching in the knowledge base based on the extracted keywords; S2-3. The large model fills the matching results into the structured rule data to be maintained, resulting in the inferred structured rule data.

[0029] In some embodiments of the present invention, in step S2-2, the knowledge base includes a rule-based translation dataset and a multilingual translation dataset.

[0030] In some embodiments of the present invention, the construction of the rule translation dataset includes: Extracting entity-relation-entity tripartite knowledge from historical data; The extracted ternary knowledge is stored in a structured manner to form a rule-based translation dataset.

[0031] In some embodiments of the present invention, the historical data is historical freight rate maintenance data information stored in DIT; In this case, in the entity-relationship-entity ternary knowledge, the two types of entities are OriginalText and the processing rules for Refund, Change, No Show, and General (examples are shown in the table below). The relationship is the mapping relationship between keywords in OriginalText and Refund, Change, No Show, and General. For example, the mapping relationship is shown in Table 1.

[0032] Table 1

[0033] In some embodiments of the present invention, the multilingual translation dataset is constructed by translating the original OtherInfo text into Chinese translations, and using the Chinese translations as relations to build a structured knowledge dataset of OtherInfo text (entities) – relations – multilingual translations (entities), which is the multilingual translation dataset.

[0034] In some embodiments of the present invention, in step S2, the large model is typically Qwen2.5-32B, but it can also be a multi-voice large model with equivalent or higher performance.

[0035] In some embodiments of the present invention, step S3, the verification and filtering identification process, includes: Validate the data information content (data structure and the validity of field content); Filter and identify abnormal data items such as duplicates and incomplete information; like Figure 2 As shown, a structured rule data intelligent parsing and input system according to certain embodiments of the present invention includes an intelligent interaction module 10, a model reasoning module 20, a data parsing service module 30, and a data intelligent maintenance module 40. The intelligent interaction module 10 is used to call the pre-trained large model based on the structured rule data to be maintained selected and entered by the user. The model inference module 20 is used to infer the structured rule data to be maintained through the pre-trained large model, and obtain the inferred structured rule data. The data parsing service module 30 is used to verify, filter, and identify the inferred structured rule data; and to encapsulate the processed data into structured rule data maintained on the page. The data intelligent maintenance module 40 is used to output structured rule data encapsulated into page maintenance.

[0036] In some embodiments of the present invention, the structured rule data intelligent parsing and input system further includes an input front-end page; The data entry front-end page is used to accept user-selected data for confirmation or re-maintenance.

[0037] In some embodiments of the present invention, the intelligent interaction module 10 is further configured to send the structured rule data to be maintained to the model inference module 20; The model reasoning module 20 is also used to receive the structured rule data being maintained, and to pass the reasoned structured rule data to the data parsing service module 30. The data parsing service module 30 is also used to receive the inferred structured rule data and send the encapsulated page maintenance structured rule data to the data intelligent maintenance module 40. The data intelligent maintenance module 40 is also used to receive encapsulated structured rule data for page maintenance and output and display it on the input front-end page.

[0038] In some embodiments of the present invention, the input front-end page is also used to accept the results of the data confirmation step selected by the user.

[0039] In some embodiments of the present invention, the result of the data verification step is either yes or no.

[0040] In some embodiments of the present invention, the data entry system terminates the maintenance process based on the determined data confirmation step results.

[0041] In some embodiments of the present invention, the data entry system repeats the "call", "reasoning", "processing and encapsulation", "input and display" and "data confirmation" steps again based on the result of the data confirmation step, until the result of the data confirmation step is determined.

[0042] The following describes the application of the method described in the above embodiments to the intelligent data entry process of structured freight rate rules, namely the international freight rate entry front-end (hereinafter referred to as DIT). The specific process is as follows: Figure 3 As shown, it includes: Invocation: The front-end page accepts the structured rule data to be maintained selected by the user and invokes the pre-trained large model through the intelligent interaction module; the intelligent interaction module sends the structured rule data to be maintained to the model inference module; Inference: The model inference module receives the maintained structured rule data and infers from the structured rule data to be maintained using a pre-trained large model to obtain the inferred structured rule data; the model inference module then transmits the inferred structured rule data to the data parsing service module. Processing and Encapsulation: The data parsing service module receives the inferred structured rule data, performs verification and filtering on the inferred structured rule data, and encapsulates the processed data into page-maintained structured rule data; the data parsing service module then sends the encapsulated page-maintained structured rule data to the data intelligent maintenance module. Output and display: The intelligent data maintenance module receives the encapsulated structured rule data for page maintenance and outputs and displays it on the input front-end page; Data Confirmation: The data entry front-end page accepts the data confirmation step result selected by the user; the data entry system, based on the negative data confirmation step result, repeats the "call", "reasoning", "processing and encapsulation", "input and display" and "data confirmation" steps until a positive data confirmation step result is obtained; the data entry system ends the maintenance process based on the positive data confirmation step result.

[0043] In some embodiments of the present invention, the input front-end page accepts the structured rule data to be maintained selected by the user. When the user selects a structured rule data on the input front-end page and clicks to enter the maintenance interface, and clicks the selection button as needed, the input front-end page enters the loading waiting state (that is, enters the "call", "reasoning", "processing and encapsulation" process in the system background). The selection button mentioned above is, for example, a rule translation function button or a multilingual automatic translation function.

[0044] In some embodiments of the present invention, when the data intelligent maintenance module receives the encapsulated structured rule data for page maintenance, the loading waiting state of the input front-end page ends, and the input front-end page outputs and displays the encapsulated structured rule data for page maintenance (i.e., the structured data after reasoning and parsing).

[0045] The specific details of the DIT system data entry example mentioned above are as follows: Smart Selection Interaction Module 10: Retrieves the structured rule data to be maintained selected and entered by the user. The user selects the aforementioned data by logging into the DIT system, clicking the StructuredRule module, querying for a structured rule data with a status of ToDo, and clicking to claim it to enter the maintenance structured rule data details page. In one scenario (corresponding to "the structured rule data to be maintained is OriginalText"): An "Auto Analyze" button will appear on the structured rule data details page, allowing the user to complete their selection. Clicking the "Auto Analyze" button will initiate a loading process (i.e., entering the application entry system). The data subsequently transmitted to the pre-trained large model will be the OriginalText information of this structured rule data.

[0046] In another scenario (corresponding to the case where "the structured rule data to be maintained is OtherInfo text"): In the OtherInfo module of the structured rule data details page, for each OtherInfo text data entry, its OriginalText information is filled in. An Auto Translate button will appear below this text box, completing the user's selection. Clicking the Auto Translate button will initiate a loading process, and the data subsequently passed to the pre-trained large model will be OtherInfo text information.

[0047] Model Inference Module 20: One scenario: Automatic parsing: After receiving the OriginalText, the pre-trained large model, upon understanding the question, searches and matches it within its internal knowledge base (this knowledge base is composed of knowledge learned during the pre-training and fine-tuning phases. The model searches the knowledge base for relevant knowledge and information based on the key information of the question). After finding the relevant knowledge and information, the large model uses its generative capabilities to organize this information into corresponding rule-based translation results and returns them.

[0048] Another scenario: Automatic translation: After receiving the text data from OtherInfo, the pre-trained large model will search and match in its internal knowledge base after understanding the question. After finding relevant knowledge and information, the large model will use its generation capabilities to organize this information into corresponding multilingual translation results data, and then translate it from Simplified Chinese into Traditional Chinese, Korean and Japanese and return it together.

[0049] Data parsing service module 30: The model inference module 20 returns the data inferred from the pre-trained large model to the data parsing service module 30. The data parsing service module 30 translates or automatically translates the data according to the returned rules, verifies the data information content, filters and identifies abnormal data items such as duplicates, and encapsulates it into a structured rule-based data structure that can be automatically filled by the DIT system page. After the data parsing is completed, it is returned to the data intelligent maintenance module 40.

[0050] Data intelligent maintenance module 40: When the data intelligent maintenance module 40 receives the parsed data, the user's loading page ends and the maintained structured data is displayed.

[0051] Users can confirm and re-maintain the data based on the displayed structured data, thus completing the maintenance process for this structured rule data.

[0052] like Figure 4 As shown, in some embodiments of the present invention, an electronic device is provided, the electronic device 300 including: a processor 301 coupled to a memory 302; The memory 302 is used to store computer programs; The processor 301 is configured to execute the computer program stored in the memory 302, so that the electronic device performs the method described in the above embodiments.

[0053] In some embodiments of the present invention, a computer-readable storage medium is provided that stores a program or instructions that, when executed on a computer, cause the computer to perform the methods described in the above embodiments.

[0054] According to embodiments of the present invention, the computer-readable storage medium may be a non-volatile computer-readable storage medium, such as including but not limited to: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In the present invention, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, electronic device, or apparatus.

[0055] In some embodiments of the present invention, a computer program product is provided, including a computer program / instructions that, when executed by a processor, implement the methods described in the above embodiments.

[0056] Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for intelligent parsing and inputting structured rule-based data, characterized in that, include: Based on the structured rule data to be maintained selected and entered by the user, the pre-trained large model is invoked; The pre-trained large model is used to infer the structured rule data to be maintained, resulting in inferred structured rule data. The structured rule data after reasoning is validated and filtered for identification. The processed data is encapsulated into structured rule data maintained by the page and then output.

2. The method for intelligent parsing and inputting structured rule data according to claim 1, characterized in that, The structured rule data to be maintained is either text information in the structured rule data or multilingual maintenance node text in the structured rule data.

3. The method for intelligent parsing and inputting structured rule data according to claim 1 or 2, characterized in that, The process of reasoning through the pre-trained large model on the structured rule data to be maintained, resulting in inferred structured rule data, includes: The pre-trained large model is used to parse the structured rule data to be maintained, and to enhance the weighting of keywords; The large model performs retrieval and matching in the knowledge base based on the extracted keywords; The large model fills the matching results into the structured rule data to be maintained, resulting in the inferred structured rule data.

4. The method for intelligent parsing and inputting structured rule data according to claim 3, characterized in that, The knowledge base includes rule-based translation datasets and multilingual translation datasets.

5. The method for intelligent parsing and inputting structured rule data according to claim 4, characterized in that, The construction of the rule-based translation dataset includes: Extracting entity-relation-entity tripartite knowledge from historical data; The extracted ternary knowledge is stored in a structured manner to form a rule-based translation dataset.

6. A structured rule-based intelligent data parsing and input system, characterized in that, It includes an intelligent interaction module, a model reasoning module, a data parsing service module, and a data intelligent maintenance module; The intelligent interaction module is used to call the pre-trained large model based on the structured rule data to be maintained selected and entered by the user. The model inference module is used to infer the structured rule data to be maintained through the pre-trained large model, and obtain the inferred structured rule data. The data parsing service module is used to verify, filter, and identify the structured rule data after inference. And encapsulate the processed data into structured rule data for page maintenance. The data intelligent maintenance module is used to output structured rule data encapsulated into page maintenance.

7. The intelligent parsing and input system for structured rule data according to claim 6, characterized in that, This also includes inputting the front-end page; The data entry front-end page is used to accept user-selected data for confirmation or re-maintenance.

8. The intelligent parsing and input system for structured rule data according to claim 7, characterized in that, The data entry front-end page is also used to accept user-selected data confirmation or re-maintenance.

9. A structured rule-based data intelligent parsing and input system according to any one of claims 6-8, characterized in that, The intelligent interaction module is also used to send the structured rule data to be maintained to the model inference module; The model inference module is also used to receive the structured rule data being maintained, and to pass the inferred structured rule data to the data parsing service module. The data parsing service module is also used to receive the inferred structured rule data and send the encapsulated page maintenance structured rule data to the data intelligent maintenance module. The data intelligent maintenance module is also used to receive encapsulated structured rule data for page maintenance and output and display it on the input front-end page.

10. An electronic device, characterized in that, include: Processor, the processor being coupled to memory; The memory is used to store computer programs; The processor is configured to execute the computer program stored in the memory to cause the electronic device to perform the method as described in any one of claims 1 to 5.

11. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a program or instructions that, when executed on a computer, cause the computer to perform the method as described in any one of claims 1 to 5.

12. A computer program product comprising a computer program / instructions, characterized in that, When the computer program / instructions are executed by the processor, they implement the method of any one of claims 1 to 5.

Images