Low-code template identification method and system for chemical experiment design

Abstract

The invention provides a low-code template identification method and system for chemical experiment design, and the method comprises the steps: binding a sub-form template to a detection project document, and designing the sub-form template on a low-code platform; based on a template recognition engine, analyzing information in the sub-form template according to the input business document; performing data identification on the analyzed sub-form template; and carrying out mark confirmation on the identification data result, and outputting the identification data result carrying a mark as an identifiable document. By introducing a low-code platform and a template recognition technology, the template can be designed and automatically recognized, the threshold of experimental design is lowered, and the accuracy and flexibility of design are improved.

Description

Technical Field

[0001] The present invention relates to the field of code development, and more specifically, to a low-code template recognition method and system for chemical experiment design. Background Art

[0002] In the chemical industry, experimental design is an important process for developing new products and optimizing existing products. Traditionally, experimental design relies on professional engineers to manually write codes or configure complex software systems, which is not only time-consuming and costly, but also prone to errors. As the chemical industry's requirements for experimental design efficiency and accuracy are increasing, traditional methods can no longer meet actual needs.

[0003] In the prior art, chemical experiment design is usually completed through professional software, which provides templates and tools for experimental design, but users still need to have certain programming knowledge or software operation experience. In addition, the prior art lacks effective recognition and error checking mechanisms for experimental design templates, and cannot automatically discover potential problems in templates, which may lead to data recording errors or experimental failures during the experiment. Summary of the invention

[0004] In view of the technical problems existing in the prior art, the present invention provides a low-code template recognition method and system for chemical experiment design.

[0005] According to a first aspect of the present invention, a low-code template recognition method for chemical experiment design is provided, comprising: Bind a sub-form template to the inspection project document, and design the sub-form template on a low-code platform; Based on the template recognition engine, the information in the subform template is parsed according to the input business document; Performing data recognition on the parsed sub-form template; The recognition data result is marked and confirmed, and the recognition data result carrying the mark is output as a recognizable document.

[0006] Based on the above technical solution, the present invention can also make the following improvements.

[0007] Optionally, binding a sub-form template to the detection project document and designing the sub-form template on a low-code platform include: Enter the design page of the subform template on the low-code platform, select the style component and the formula component by dragging and dropping, and design the subform template based on the style component and the formula component; The sub-form template designed for the detection project document is bound, and the calculation formula and information recording method of each detection sub-item are configured in the sub-form template.

[0008] Optionally, the template recognition engine is based on parsing the information in the subform template according to the input business document, including: According to the input business document, the designed subform template is parsed to parse out the information in the subform template, wherein the information includes proper nouns, detection indicators, formulas, tables, pictures, placeholders, fields and formats; The information extracted from the subform template is normalized.

[0009] Optionally, standardizing the information extracted from the subform template includes: Remove duplicate information; If missing information is identified, the information is automatically or manually corrected; Unify the format and unit of unit or date information; Using a scale scaling algorithm, feature information at different scales is scaled to the same range, wherein the feature information includes detection indicators and images; Use clustering algorithms to group information under different conditions and types.

[0010] Optionally, the performing data identification on the parsed sub-form template includes: Identifying specific entities in the parsed subform template, wherein the specific entities include detection indicators, test conditions, and test results, wherein the detection indicators include elongation and aging rate, the test conditions include temperature and pressure, and the test results include density and oxygen index; Performing syntactic analysis and semantic understanding on the subform template to extract values ​​of various fields; The calculation formula in the subform template is parsed using a symbolic calculation library parsing formula to generate an abstract syntax tree thereof.

[0011] Optionally, marking and confirming the recognition data result includes: Comparing each field value extracted from the subform template with each reference field value in the business document to determine whether each field value extracted from the subform template is correct; If there is no field value corresponding to the field extracted from the subform template in the business document, a reference field value is determined according to the empirical value of the field or the context of the business document, and each field value extracted from the subform template is compared with the reference field value to determine whether each field value extracted from the subform template is correct; If the field value extracted from the subform template is incorrect, marking the field value as an error; For the parsed calculation formula, based on the generated abstract syntax tree, determine whether the grammatical structure of the calculation formula is correct, and based on the rule engine, verify the logic of the calculation formula and mark the erroneous calculation formula as an error.

[0012] Optionally, while outputting the recognition data result carrying the mark as a recognizable document, the method further includes: The erroneous or improper recognition data results identified from the sub-form template are displayed in an intuitive manner, wherein the displayed content includes the location of the misplaced recognition data result, the error type and the impact caused by the error.

[0013] Optionally, the step of outputting the recognition data result carrying the mark as a recognizable document further includes: Based on the identified erroneous or improper identification data results, modification suggestions and adjustment guidance are given, and the modification suggestions and adjustment guidance include suggestions for modifying calculation formulas, adjustment methods for data consistency, and optimization suggestions for the subform template.

[0014] According to a second aspect of the present invention, there is provided a low-code template recognition system for chemical experiment design, comprising a low-code design platform and a template recognition engine; The low-code design platform is used for users to design sub-form templates thereon and bind the sub-form templates to the inspection project documents; The template recognition engine is used to parse the information in the sub-form template according to the input business document, perform data recognition on the parsed sub-form template; mark and confirm the recognition data result, and output the recognized data result carrying the mark as a recognizable document.

[0015] Optionally, the system further includes an error display module and an adjustment guidance module; The error display module is used to display the erroneous or improper recognition data results identified from the sub-form template in an intuitive manner, wherein the displayed content includes the location of the misplaced recognition data result, the error type and the impact caused by the error; The adjustment guidance module is used to provide modification suggestions and adjustment guidance based on the identified erroneous or improper identification data results, and the modification suggestions and adjustment guidance include suggestions for modifying calculation formulas, methods for adjusting data consistency, and optimization suggestions for the subform template.

[0016] The present invention provides a low-code template recognition method and system for chemical experiment design, which introduces a low-code platform and template recognition technology, designs a sub-form template on the low-code platform, binds the sub-form template to the test project document; based on the template recognition engine, parses the information in the sub-form template according to the input business document; performs data recognition on the parsed sub-form template; marks and confirms the recognition data result, and outputs the recognition data result with the mark as a recognizable document. The present invention can realize the design and automatic recognition of templates, lower the threshold of experimental design, and improve the accuracy and flexibility of design. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 A flow chart of a low-code template recognition method for chemical experiment design provided by the present invention; Figure 2 The overall flow chart of the low-code template recognition method designed for chemical experiment; Figure 3 A structural schematic diagram of a low-code template recognition system for chemical experiment design provided by the present invention. DETAILED DESCRIPTION

[0018] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention. In addition, the technical features in the various embodiments or single embodiments provided by the present invention can be arbitrarily combined with each other to form a feasible technical solution. This combination is not subject to the constraints of the sequence of steps and / or the structural composition mode, but must be based on the ability of ordinary technicians in this field to achieve. When the combination of technical solutions is contradictory or cannot be achieved, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

[0019] Figure 1 A flow chart of a low-code template recognition method for chemical experiment design provided by the present invention, such as Figure 1 and Figure 2 As shown, the method includes: Step 1: bind the sub-form template to the inspection project document and design the sub-form template on the low-code platform.

[0020] It is understandable that users can bind sub-form templates to inspection project documents and design sub-form templates on the low-code platform. Specifically, designing sub-form templates on the low-code platform includes: Enter the design page of the subform template on the low-code platform, and select the style component by dragging and dropping to design the style of the subform template according to the style components and formula components provided by the low-code platform, and select the formula component by dragging and dropping to design the formula in the subform template. Bind the designed subform template to the detection project document, and configure the calculation formula and information recording method of each detection sub-item in the subform template.

[0021] Among them, users can design subform templates on the low-code platform by dragging and dropping, without having to write complex codes. The low-code platform provides a variety of styles and formula components, which users can select and configure according to experimental needs.

[0022] Step 2: Based on the template recognition engine, the information in the subform template is parsed according to the input business document.

[0023] It is understandable that for the designed sub-form template, the template recognition engine is used to parse the information in the sub-form template. The parsing process includes: parsing the designed sub-form template according to the input business document, parsing the information in the sub-form template, the information including proper nouns, detection indicators, formulas, tables, pictures, placeholders, fields and formats; and standardizing the information extracted from the sub-form template.

[0024] The template recognition engine is pre-trained by collecting and processing a large amount of business documents and related professional information, proper nouns, etc., to provide sufficient sample data for machine learning and optimization of template recognition models. The template recognition engine is trained based on the sample data.

[0025] The trained template recognition engine parses the information in the designed subform template, extracts the proper nouns, indicators, formulas, tables, pictures, placeholders, fields, formats and other information in the subform template, and standardizes the information. Standardization includes: removing duplicate information; if missing information is identified, automatically or manually correcting the information; unifying the format and unit of unit or date information; using a scaling algorithm to scale feature information at different scales to the same range, the feature information includes detection indicators and pictures; using a clustering algorithm to group information under different conditions and types to facilitate the discovery and merging of similar records, etc. Use the above standardization method to prepare for the subsequent template recognition.

[0026] Step 3: performing data recognition on the parsed sub-form template.

[0027] The step of performing data identification on the parsed sub-form template includes: The specific entity in the parsed sub-form template is identified, the specific entity including detection indicators, test conditions and test results, the detection indicators including elongation and aging rate, the test conditions including temperature and pressure, and the test results including density and oxygen index.

[0028] Performing syntactic analysis and semantic understanding on the subform template to extract values ​​of various fields; The calculation formula in the subform template is parsed using a symbolic calculation library parsing formula to generate an abstract syntax tree thereof.

[0029] It is understandable that the recognition of subform templates combines machine learning and NLP technology, focusing on improving adaptability to complex documents and formats. It mainly includes the following recognitions: (1) Training a specific named entity recognition (NER) to identify specific entities in the subform template, such as test indicators (elongation, aging rate, etc.), test conditions (temperature, pressure, etc.), and test results (density, oxygen index, etc.). (2) Perform syntactic analysis and semantic understanding to identify the structure of the subform template as accurately as possible and extract the values ​​of each field. (3) Use the symbolic computing library (SymPy) to parse the calculation formula of each test item and generate its abstract syntax tree (AST).

[0030] Step 4: Mark and confirm the recognition data result, and output the recognition data result carrying the mark as a recognizable document.

[0031] It is understandable that after the template recognition engine recognizes the data in the subform template, it is necessary to determine whether the recognition result is correct. The recognition data result is marked and confirmed, including: The various field values ​​extracted from the subform template are compared with the various reference field values ​​in the business document to determine whether the various field values ​​extracted from the subform template are correct. If there is no field value corresponding to the field extracted from the subform template in the business document, the reference field value is determined according to the empirical value of the field or the context of the business document, and the various field values ​​extracted from the subform template are compared with the reference field value to determine whether the various field values ​​extracted from the subform template are correct. If the field value extracted from the subform template is incorrect, the field value is marked as an error. For the parsed calculation formula, based on the generated abstract syntax tree, it is determined whether the grammatical structure of the calculation formula is correct, and based on the rule engine, the logic of the calculation formula is verified, and the erroneous calculation formula is marked as an error.

[0032] After recognizing the data in the subform template, if the recognition is inaccurate, feedback can be provided, such as marking the wrong field through the interface and giving the correct value. These feedback data are automatically collected and marked as "wrong" or "corrected" samples for subsequent template recognition engine training. Automated feedback: The template recognition engine can also collect feedback through automated methods. For example, when an uncertain field or a low-confidence field is encountered during the recognition process, confirmation can be automatically requested from the user or administrator. These confirmed data can also be used to optimize the template recognition engine.

[0033] The template recognition engine can continuously update the recognition engine model through online learning or incremental learning when new data and feedback arrive, thereby optimizing the accuracy of field recognition. For example, new sample data can be used to train the incremental recognition engine model without retraining the entire recognition engine model.

[0034] The present invention also displays the erroneous or improper identification data results identified from the subform template in an intuitive manner, wherein the displayed content includes the location of the misplaced identification data results, the error type, and the impact of the error. In addition, based on the identified errors or improprieties, the adjustment guidance module provides modification suggestions and adjustment guidance, and the guidance content includes suggestions for modifying formulas, methods for adjusting data consistency, and optimization suggestions for template logic.

[0035] Adjust the design of the subform template according to the guidance and modification suggestions, and then resubmit it to the template recognition engine for recognition until the data recognized from the subform template is accurate.

[0036] See also Figure 3 , a low-code template recognition system for chemical experiment design provided by the present invention, the system includes a low-code design platform 301, a template recognition engine 302, an error display module 303 and an adjustment guidance module 304; The low-code design platform 301 is used for users to design sub-form templates thereon and bind the sub-form templates to the inspection project documents; The template recognition engine 302 is used to parse the information in the sub-form template according to the input business document, perform data recognition on the parsed sub-form template; mark and confirm the recognition data result, and output the recognition data result with the mark as a recognizable document; The error display module 303 is used to display the erroneous or improper recognition data results identified from the sub-form template in an intuitive manner, wherein the displayed content includes the location of the misplaced recognition data result, the error type and the impact caused by the error; The adjustment guidance module 304 is used to provide modification suggestions and adjustment guidance according to the identified erroneous or improper identification data results. The modification suggestions and adjustment guidance include suggestions for modifying calculation formulas, methods for adjusting data consistency, and optimization suggestions for the subform template.

[0037] It can be understood that the low-code template recognition system for chemical experiment design provided by the present invention corresponds to the low-code template recognition method for chemical experiment design provided by the aforementioned embodiments. The relevant technical features of the low-code template recognition system for chemical experiment design can refer to the relevant technical features of the low-code template recognition method for chemical experiment design, which will not be repeated here.

[0038] The low-code template recognition method and system for chemical experiment design provided by the embodiment of the present invention have the following beneficial effects: (1) Lowering the threshold: Through the low-code design platform, users can complete experimental design without writing complex codes, which lowers the threshold for experimental design.

[0039] (2) Improved accuracy: The template recognition engine and error display module can automatically identify errors in the template, thereby improving the accuracy of the experimental design.

[0040] (3) Enhanced flexibility: Users can freely design templates according to experimental requirements, which enhances the flexibility of experimental design.

[0041] (4) Improved efficiency: Automated error checking and adjustment guidance mechanisms reduce the time for manual inspection and improve the efficiency of experimental design.

[0042] It should be noted that in the above embodiments, the description of each embodiment has its own emphasis, and for parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0043] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0044] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0045] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0046] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0047] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0048] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A low-code template recognition method for chemical experiment design, characterized in that: include: Bind a sub-form template to the inspection project document, and design the sub-form template on a low-code platform; Based on the template recognition engine, the information in the subform template is parsed according to the input business document; Performing data recognition on the parsed sub-form template; The recognition data result is marked and confirmed, and the recognition data result carrying the mark is output as a recognizable document.

2. The low-code template recognition method according to claim 1 is characterized in that: Binding the sub-form template to the detection project document and designing the sub-form template on the low-code platform include: Enter the design page of the subform template on the low-code platform, select the style component and the formula component by dragging and dropping, and design the subform template based on the style component and the formula component; The sub-form template designed for the detection project document is bound, and the calculation formula and information recording method of each detection sub-item are configured in the sub-form template.

3. The low-code template recognition method according to claim 1 is characterized in that: The template recognition engine parses the information in the subform template according to the input business document, including: According to the input business document, the designed subform template is parsed to parse out the information in the subform template, wherein the information includes proper nouns, detection indicators, formulas, tables, pictures, placeholders, fields and formats; The information extracted from the subform template is normalized.

4. The low-code template recognition method according to claim 3 is characterized in that: Standardize the information extracted from the subform template, including: Remove duplicate information; If missing information is identified, the information is automatically or manually corrected; Unify the format and unit of unit or date information; Using a scale scaling algorithm, feature information at different scales is scaled to the same range, wherein the feature information includes detection indicators and images; Use clustering algorithms to group information under different conditions and types.

5. The low-code template recognition method according to claim 1 is characterized in that: The step of performing data identification on the parsed sub-form template includes: Identifying specific entities in the parsed subform template, wherein the specific entities include detection indicators, test conditions, and test results, wherein the detection indicators include elongation and aging rate, the test conditions include temperature and pressure, and the test results include density and oxygen index; Performing syntactic analysis and semantic understanding on the subform template to extract values ​​of various fields; The calculation formula in the subform template is parsed using a symbolic calculation library to generate a corresponding abstract syntax tree.

6. The low-code template recognition method according to claim 5 is characterized in that: The marking and confirming of the recognition data result includes: Comparing each field value extracted from the subform template with each reference field value in the business document to determine whether each field value extracted from the subform template is correct; If there is no reference field value corresponding to the field extracted from the subform template in the business document, the reference field value is determined according to the experience value of the field or the context of the business document, and each field value extracted from the subform template is compared with the reference field value to determine whether each field value extracted from the subform template is correct; If the field value extracted from the subform template is incorrect, marking the field value as an error; For the parsed calculation formula, whether the grammatical structure of the calculation formula is correct is determined based on the generated abstract syntax tree, and the logic of the calculation formula is verified based on the rule engine, and the erroneous calculation formula is marked as an error.

7. The low-code template recognition method according to claim 1, characterized in that: The step of outputting the recognition data result carrying the mark as a recognizable document also includes: The erroneous or improper recognition data results identified from the sub-form template are displayed in an intuitive manner, wherein the displayed content includes the location of the misplaced recognition data result, the error type and the impact caused by the error.

8. The low-code template recognition method according to claim 7, characterized in that: The step of outputting the recognition data result carrying the mark as a recognizable document further includes: Based on the identified erroneous or improper identification data results, modification suggestions and adjustment guidance are given, and the modification suggestions and adjustment guidance include suggestions for modifying calculation formulas, adjustment methods for data consistency, and optimization suggestions for the subform template.

9. A low-code template recognition system for chemical experiment design, characterized in that: Includes a low-code design platform and template recognition engine; The low-code design platform is used for users to design sub-form templates thereon and bind the sub-form templates to the inspection project documents; The template recognition engine is used to parse the information in the sub-form template according to the input business document, perform data recognition on the parsed sub-form template; mark and confirm the recognition data result, and output the recognized data result carrying the mark as a recognizable document.

10. The low-code template recognition system according to claim 9, characterized in that: The system also includes an error display module and an adjustment guidance module; The error display module is used to display the erroneous or improper recognition data results identified from the sub-form template in an intuitive manner, wherein the displayed content includes the location of the misplaced recognition data result, the error type and the impact caused by the error; The adjustment guidance module is used to provide modification suggestions and adjustment guidance based on the identified erroneous or improper identification data results, and the modification suggestions and adjustment guidance include suggestions for modifying calculation formulas, methods for adjusting data consistency, and optimization suggestions for the subform template.

Images