Processing method and device of inspection standard text, electronic equipment and storage medium

Abstract

The invention provides an inspection standard text processing method and device, electronic equipment and a storage medium, and relates to the technical field of inspection standard processing. The method comprises the following steps: converting an inspection standard text into a natural text paragraph; splitting the natural text paragraph into a plurality of pieces of structured metadata according to text semantics, and dividing each piece of structured metadata in the plurality of pieces of structured metadata into inspection object type structured metadata, inspection item type structured metadata, inspection method type structured metadata and inspection index type structured metadata; based on the test object type structured metadata, the test item type structured metadata, the test method type structured metadata and the test index type structured metadata, establishing an association relationship corresponding to the test object type structured metadata; and establishing a tree structure corresponding to the test object class structured metadata based on the association relationship. The tree structure can be automatically established according to the inspection standard text, and efficiency and accuracy are high.

Description

Technical Field

[0001] Embodiments of the present invention relate to the technical field of inspection standard processing, and in particular to a method and device for processing inspection standard text, an electronic device, and a storage medium. Background Art

[0002] The management of inspection standards is a core component of the laboratory management system, which directly affects the laboratory's operational efficiency and data reliability. It can ensure that laboratory operations comply with established standards and specifications, and is an important guarantee for laboratory quality control and compliance management.

[0003] The core and difficulty of inspection standard management is to parse the unstructured inspection standard text into structured data. At present, this work is done manually. The specific process is roughly as follows: the standard administrator will split the inspection standard text into independent structured metadata such as inspection object, inspection item, inspection method, etc., and then associate these independent structured metadata through foreign keys.

[0004] However, manually completing the above contents may lead to problems such as human errors and subjective consistency deviations. In addition, manually completing the above contents may also lead to problems such as low work efficiency, data redundancy, difficulty in designing and maintaining association relationships, and changes requiring a lot of maintenance work. Summary of the invention

[0005] In view of the above problems, embodiments of the present invention are proposed to provide a method and device for processing inspection standard text, an electronic device and a storage medium that overcome the above problems or at least partially solve the above problems.

[0006] In a first aspect, an embodiment of the present application discloses a method for processing a verification standard text, comprising:

[0007] Convert the standard text into natural text paragraphs;

[0008] Splitting the natural text paragraph into a plurality of structured metadata according to text semantics, and dividing each of the plurality of structured metadata into inspection object structured metadata, inspection item structured metadata, inspection method structured metadata, and inspection indicator structured metadata;

[0009] Establishing the association relationship corresponding to the inspection object class structured metadata based on the inspection object class structured metadata, the inspection item class structured metadata, the inspection method class structured metadata and the inspection indicator class structured metadata;

[0010] A tree structure corresponding to the structured metadata of the inspection object class is established based on the association relationship.

[0011] Optionally, the method further comprises:

[0012] Performing similarity identification on each of the structured metadata and the existing structured metadata in the existing metadata table;

[0013] For one of the structured metadata, when the similarity between the structured metadata and the existing structured metadata in the existing metadata table is less than or equal to a preset similarity, the structured metadata is stored in the existing metadata table.

[0014] Optionally, the method further comprises:

[0015] Obtain the inspection object class structured metadata, inspection item class structured metadata, inspection method class structured metadata, and inspection indicator class structured metadata, and their corresponding numbers in the existing metadata table;

[0016] In the tree structure, the inspection object class structured metadata, the inspection item class structured metadata, the inspection method class structured metadata and the inspection indicator class structured metadata are respectively replaced with the corresponding numbers to form an inspection standard library relationship table, and the inspection standard library relationship table is stored in the inspection standard library.

[0017] Optionally, the step of splitting the natural text paragraph into a plurality of structured metadata according to text semantics, and dividing each of the plurality of structured metadata into inspection object structured metadata, inspection item structured metadata, inspection method structured metadata, and inspection indicator structured metadata includes:

[0018] A named entity recognition model is adopted to split the natural text paragraph into multiple structured metadata according to text semantics, and each structured metadata in the multiple structured metadata is divided into inspection object type structured metadata, inspection item type structured metadata, inspection method type structured metadata and inspection indicator type structured metadata.

[0019] Optionally, the establishing of the association relationship corresponding to the inspection object class structured metadata based on the inspection object class structured metadata, the inspection item class structured metadata, the inspection method class structured metadata and the inspection indicator class structured metadata includes:

[0020] The bidirectional encoding model of Transformer is adopted to establish the association relationship corresponding to the structured metadata of the inspection object class based on the structured metadata of the inspection object class, the structured metadata of the inspection item class, the structured metadata of the inspection method class and the structured metadata of the inspection indicator class.

[0021] Optionally, converting the inspection standard text into a natural text paragraph includes:

[0022] The inspection standard text is converted into natural text paragraphs through text table parsing tools.

[0023] Optionally, the establishing of a tree structure corresponding to the structured metadata of the inspection object class based on the association relationship includes:

[0024] An object nesting method is adopted to establish a tree structure corresponding to the structured metadata of the inspection object class based on the association relationship.

[0025] In a second aspect, a processing device for verifying a standard text is provided, comprising:

[0026] A conversion module, used to convert the inspection standard text into natural text paragraphs;

[0027] A splitting and classification module, used for splitting the natural text paragraph into a plurality of structured metadata according to text semantics, and dividing each of the plurality of structured metadata into inspection object type structured metadata, inspection item type structured metadata, inspection method type structured metadata and inspection indicator type structured metadata;

[0028] An association relationship establishment module, used to establish an association relationship corresponding to the inspection object class structured metadata based on the inspection object class structured metadata, the inspection item class structured metadata, the inspection method class structured metadata and the inspection indicator class structured metadata;

[0029] The tree structure establishment module is used to establish a tree structure corresponding to the structured metadata of the inspection object class based on the association relationship.

[0030] According to a third aspect, an electronic device is provided, including:

[0031] A processor, a memory, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, any one of the aforementioned methods for processing inspection standard texts is implemented.

[0032] In a fourth aspect, a readable storage medium is provided, which, when the instructions or transactions in the storage medium are executed by a processor of an electronic device, enables the electronic device to execute any of the aforementioned methods for processing inspection standard texts.

[0033] In a fifth aspect, a computer program product is provided, comprising instructions or transactions, which, when executed by a processor in an electronic device, enable the electronic device to execute any one of the aforementioned methods for processing inspection standard texts.

[0034] In an embodiment of the present application, a tree structure corresponding to the structured metadata of the inspection object class can be automatically established based on the inspection standard text, which reduces human errors and subjective biases, keeps the data concise, realizes automatic disassembly and relationship modeling, and improves the work efficiency and accuracy of inspection standard maintenance. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 It is a flowchart of a method for processing a verification standard text provided by an embodiment of the present invention;

[0036] Figure 2 is a schematic diagram of a test standard text provided by an embodiment of the present invention;

[0037] Figure 3 is a schematic diagram of another inspection standard text provided by an embodiment of the present invention;

[0038] Figure 4 is a flowchart of steps for training a NER model provided by an embodiment of the present invention;

[0039] Figure 5 is a flowchart of splitting and classification steps provided by an embodiment of the present invention;

[0040] Figure 6 is a flowchart of the steps of training a BERT model provided by an embodiment of the present invention;

[0041] Figure 7 is a schematic diagram of a tree structure provided by an embodiment of the present invention;

[0042] Figure 8 This is a flowchart of the steps of establishing an association relationship and generating a tree structure provided by an embodiment of the present invention;

[0043] Fig. 9 is a flowchart of another method for processing a verification standard text provided by an embodiment of the present invention;

[0044] Fig.10 is a flowchart of similarity matching steps provided by an embodiment of the present invention;

[0045] Fig.11 It is a partial schematic diagram of an existing metadata table provided by an embodiment of the present invention;

[0046] Fig.12 It is a schematic diagram of a test standard library relationship table provided by an embodiment of the present invention;

[0047] Fig.13 It is a structural schematic diagram of a processing device for checking standard text provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0048] The exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided in order to enable a more thorough understanding of the present invention and to enable the scope of the present invention to be fully communicated to those skilled in the art.

[0049] The present application provides a method for preparing a processing method for testing standard text, referring to Figure 1 , the method comprises the following steps.

[0050] Step A1, converting the inspection standard text into natural text paragraphs.

[0051] An inspection standard document is a document that specifies in detail how to inspect and test a product, service, or process. For example, the inspection standard document may be an inspection standard document applied in the medical field, etc. There is no limitation on the specific field of the inspection standard document. The inspection standard document usually includes inspection objects, inspection items, inspection methods, inspection indicators, etc. For example, Figure 2 and Figure 3 These are two examples of inspection standard texts. The inspection index here refers to the normal range of the inspection results of the corresponding inspection items.

[0052] The test standard text can be converted into natural text paragraphs by using natural language processing models, etc. Figure 2 and Figure 3 ,Usually, the inspection standard text contains both text and tables. Optionally, a text / table parsing tool (such as Python’s fdfplumber, AI’s Amazon Textract, etc.) can be used to parse the inspection standard document into ,natural text paragraphs conveniently and accurately.

[0053] Step A2, splitting the natural text paragraph into multiple structured metadata according to text semantics, and dividing each structured metadata in the multiple structured metadata into inspection object class structured metadata, inspection item class structured metadata, inspection method class structured metadata and inspection indicator class structured metadata.

[0054] Here, the natural text paragraphs obtained by converting the inspection standard text are first split according to semantics to obtain multiple structured metadata. Some natural speech models can be used to achieve automatic and accurate splitting. Then, each structured metadata in the multiple structured metadata is divided into inspection object structured metadata, inspection item structured metadata, inspection method structured metadata and inspection indicator structured metadata. This classification action can also be achieved by means of some natural speech models to achieve automatic and accurate division.

[0055] For example, for Figure 2 The structured metadata obtained by splitting the test standard text shown in the figure are: background count; WBC; RBC; Hb; PIt; (≦0.2×10 9 / L); (≦0.02×10 112 / L); (≦1g / L); (≦5×10 9 / L);

[0056] {Use diluent.......;}

[0057] For example, for Figure 3 The structured metadata obtained by splitting the inspection standard text shown is: carryover contamination; WBC; RBC; Hb; PIt; (≦1.0%); (≦1.0%); (≦1.0%); (≦1.0%);

[0058] {For different test items, select.....}

[0059] For example, for Figure 2 , Figure 3 The results of multiple structured metadata classifications of the inspection standard text shown are: Inspection object: background count; carryover contamination

[0060] Test items: WBC; RBC; Hb; PIt

[0061] Index: (~, 0.2×10 9 ]; (~, 0.02×10 12 ];(~,1];(~,5×10 9 ]; (~, 1.0%]; (~, 1.0%]; (~, 1.0%]; (~, 1.0%];

[0062] Test method: {Use diluent.......}

[0063] {For different test items, select.....}

[0064] Figure 2 and Figure 3 The inspection items of the inspection standard text shown correspond to the same.

[0065] Optionally, the aforementioned step A2 can be completed automatically, quickly and accurately using a named entity recognition (NER) model. The NER model is mainly used to identify and classify named entities from text. Named entities are usually specific objects. Through the trained NER model, objects with specific features can be well identified and classified according to their characteristic properties. It should be noted that before using the NER model, it is necessary to train the NER model first. The training of the NER model can refer to Figure 4 , specifically, including the following sub-steps 101 to 105. 101: You can prepare the inspection standard text for training first. 102: Use common text / table parsing tools (such as Python's fdfplumber, AI's Amazon Textract, etc.) to parse the inspection standard text into natural text paragraphs. 103: Manually prepare two types of labeled training data for the parsed natural text paragraphs, namely "labeled data for extracting key metadata" and "labeled data for information classification". The classification here is classified into structured metadata of inspection object class, structured metadata of inspection item class, structured metadata of inspection method class and structured metadata of inspection indicator class. 104: Install the SpaC library, obtain the NER component, load the text and labeled data for training, adjust the training parameters and carry out iterative training. 105: After the training is completed, save the model to get the pre-trained NER model.

[0066] The process of splitting and classifying structured metadata using the NER model can be as follows: Figure 5 As shown, it specifically includes sub-steps 303 to 304. 301: Load the inspection standard document; 302: Generate a natural text paragraph after processing by a text parsing tool, which corresponds to the aforementioned step A1. 303: Deploy the pre-trained NER model and load the natural text paragraph into the NER model. 304: The NER model extracts and classifies the key metadata or structured metadata of the document, and classifies it. It should be noted that the key metadata mentioned in the full text and the accompanying drawings are or structured metadata.

[0067] Step A3: establishing an association relationship corresponding to the inspection object class structured metadata based on the inspection object class structured metadata, the inspection item class structured metadata, the inspection method class structured metadata and the inspection indicator class structured metadata.

[0068] The association relationship here refers to establishing an association relationship between the inspection object type structured metadata, inspection item type structured metadata, inspection method type structured metadata and inspection indicator type structured metadata corresponding to the same inspection object. This step can also be achieved by means of some natural speech models, etc., to automatically and accurately establish the association relationship.

[0069] For example, for Figure 2 In the example shown, the association established in this step can be: subject: "background count"

[0070] Item: {WBC; (0.2×10 9 ]}

[0071] Item: {RBC; (~, 0.02×10 12 ]}

[0072] item:{Hb;(~,1]}

[0073] item: {PIt; (~, 5×10 9 ]}

[0074] Methods: {Use diluent....}

[0075] Optionally, step A3 can adopt the Transformer's bidirectional encoding (BERT) model to establish an association relationship based on the aforementioned test object class structured metadata, test item class structured metadata, test method class structured metadata, and test indicator class structured metadata. The Transformer's bidirectional encoding model understands the text through a bidirectional (i.e., considering both the left and right contexts) encoder, which can capture more contextual information than the traditional unidirectional language model, and the association relationship established thereby is more accurate.

[0076] It should be noted that the Transformer bidirectional encoding model needs to be trained before it can be used. Figure 6 Specifically, sub-steps 201 to 203 may be included. 201: Prepare a data set for training, including text paragraphs, key metadata, and metadata associations. 202: Install the Transformers library, load BERT and the training model and configuration, load the training data set, and fine-tune to the relationship extraction training task. 203: After the training is completed, save the model to obtain a pre-trained BERT model.

[0077] Step A4: establishing a tree structure corresponding to the structured metadata of the inspection object class based on the association relationship.

[0078] This step is mainly to convert the aforementioned association relationship into a tree structure that is easy to understand and apply. Figure 2 and Figure 3 For example, the tree structure established can refer to Figure 7 .

[0079] Optional, see Figure 8, an object nesting method can be used to establish a tree structure corresponding to the structured metadata of the inspection object class based on the association relationship. The object nesting method allows objects to be organized in a tree structure to form a part-whole relationship, thereby forming the tree structure efficiently and accurately.

[0080] To sum up, in the embodiments of the present application, a tree structure corresponding to the structured metadata of the inspection object class can be automatically established based on the inspection standard text, which reduces human errors and subjective biases, keeps the data concise, realizes automatic disassembly and relationship modeling, and improves the work efficiency and accuracy of inspection standard maintenance.

[0081] This application provides another method for preparing a processing method for testing standard text, referring to Fig. 9 , the method comprises the following steps.

[0082] Step B1, converting the inspection standard text into natural text paragraphs.

[0083] Step B2, splitting the natural text paragraph into multiple structured metadata according to text semantics, and dividing each structured metadata in the multiple structured metadata into inspection object type structured metadata, inspection item type structured metadata, inspection method type structured metadata and inspection indicator type structured metadata.

[0084] Step B3: establishing an association relationship corresponding to the inspection object class structured metadata based on the inspection object class structured metadata, inspection item class structured metadata, inspection method class structured metadata and inspection indicator class structured metadata.

[0085] Step B4: establishing a tree structure corresponding to the structured metadata of the inspection object class based on the association relationship.

[0086] Steps B1 to B4 herein may correspond to steps A1 to A4 described above, and no specific limitation is given to avoid repetition.

[0087] Step B5: performing similarity identification on each of the structured metadata and the existing structured metadata in the existing metadata table.

[0088] Here, the main purpose is to identify text similarity. Before step B5, the application has already established a metadata table for storing existing structured metadata. Fig.10 In step 403, a similarity function may be used to adjust a set threshold value, and similarity judgment may be performed between the classification and structured metadata obtained from the NER model and the existing structured metadata in the table corresponding to the classification of the existing metadata table.

[0089] Step B6: for one of the structured metadata, if the similarity between the structured metadata and the existing structured metadata in the existing metadata table is less than or equal to a preset similarity, the structured metadata is stored in the existing metadata table.

[0090] The preset similarity here can be set according to actual needs, and no specific limitation is made to this. Here, reference can be made to sub-steps 404 and 405 in 10, 404: dissimilarity indicates that the structured metadata does not exist in the existing metadata table, and the structured metadata is updated to the classification table corresponding to the existing metadata table. 405: similarity indicates that the structured metadata already exists in the existing metadata table, and there is no need to update the existing metadata table to avoid data redundancy.

[0091] Step B7, obtaining the inspection object type structured metadata, inspection item type structured metadata, inspection method type structured metadata and inspection indicator type structured metadata, and their corresponding numbers in the existing metadata table.

[0092] You can refer to here Fig.10 In sub-step 406, similarity can be used to match the split structured metadata, or the classified test object class structured metadata, test item class structured metadata, test method class structured metadata and test indicator class structured metadata with the existing structured metadata in the corresponding table of the existing metadata table. The database table field ID (number) with the highest similarity of the structured metadata is the number corresponding to the structured metadata in the aforementioned existing metadata table.

[0093] Step B8, replace the inspection object class structured metadata, inspection item class structured metadata, inspection method class structured metadata and inspection indicator class structured metadata in the tree structure with the corresponding numbers respectively to form an inspection standard library relationship table, and store the inspection standard library relationship table in the inspection standard library.

[0094] You can refer to here Figure 8 Sub-steps 505 and 506 in the above are as follows: 505: replace the structured metadata or the structured metadata of the inspection object class, the structured metadata of the inspection item class, the structured metadata of the inspection method class and the structured metadata of the inspection indicator class with the metadata table ID; 506: maintain the metadata table in the inspection standard library relation table. For more details, refer to Fig.11 , Fig.11 This is a partial diagram of the existing metadata table. Figure 7 The tree structure shown is transformed into Fig.12The inspection standard library relation table is then stored in the inspection standard library. By replacing the structured metadata or the inspection object class structured metadata, the inspection item class structured metadata, the inspection method class structured metadata, and the inspection indicator class structured metadata with the metadata table ID, the storage space can be reduced and the relation table looks more concise.

[0095] Reference Fig. 9 , the present application is further explained below in conjunction with specific embodiments.

[0096] The first step is to convert the inspection standard text into natural text paragraphs through text table parsing tools.

[0097] The second step is to pre-train the NER model. Figure 4 Sub-steps 101 to 105 in .

[0098] The third step is to pre-train the BERT model. For details, refer to the third step. Figure 6 Sub-steps 201 to 203 in .

[0099] In the fourth step, the NER model trained in the second step is used to split the aforementioned natural text paragraph into multiple structured metadata according to the text semantics, and each of the multiple structured metadata is divided into test object structured metadata, test item structured metadata, test method structured metadata and test indicator structured metadata. Figure 5 Sub-steps 301 to 304 in .

[0100] The fifth step is to identify the similarity between each of the aforementioned structured metadata and the existing structured metadata in the existing metadata table; for a structured metadata, if the similarity between the structured metadata and the existing structured metadata in the existing metadata table is less than or equal to the preset similarity, the structured metadata is stored in the existing metadata table, and the corresponding numbers of each structured metadata or the test object type structured metadata, the test item type structured metadata, the test method type structured metadata and the test indicator type structured metadata in the existing metadata table are obtained. The fifth step can be specifically referred to Fig.10 Sub-steps 401 to 406 in .

[0101] The sixth step is to use the BERT model trained in the third step to establish an association relationship based on the aforementioned test object class structured metadata, test item class structured metadata, test method class structured metadata, and test indicator class structured metadata, and then use the object nesting method to establish a tree structure corresponding to the test object class structured metadata based on the association relationship, and replace the test object class structured metadata, test item class structured metadata, test method class structured metadata, and test indicator class structured metadata in the tree structure with corresponding numbers, respectively, to form a test standard library relationship table, and store the test standard library relationship table in the test standard library. The sixth step can be specifically referred to Figure 8 Sub-steps 501 to 506 in .

[0102] It should be noted that Fig. 9 The meta-object table and database table in refer to the existing metadata table, and metadata is the abbreviation of structured metadata.

[0103] The present application also provides a processing device for checking standard text, referring to Fig.13 ,include:

[0104] A conversion module C1, for converting the inspection standard text into a natural text paragraph;

[0105] A splitting and classification module C2 is used to split the natural text paragraph into multiple structured metadata according to text semantics, and divide each of the multiple structured metadata into inspection object type structured metadata, inspection item type structured metadata, inspection method type structured metadata and inspection indicator type structured metadata;

[0106] An association relationship establishment module C3, used to establish an association relationship corresponding to the inspection object class structured metadata based on the inspection object class structured metadata, the inspection item class structured metadata, the inspection method class structured metadata and the inspection indicator class structured metadata;

[0107] The tree structure establishing module C4 is used to establish a tree structure corresponding to the structured metadata of the inspection object class based on the association relationship.

[0108] Optionally, the device further comprises:

[0109] A similarity identification module, used to identify similarities between each of the structured metadata and the existing structured metadata in the existing metadata table;

[0110] The storage module is used for storing, for one of the structured metadata, the structured metadata into the existing metadata table when the similarity between the structured metadata and the existing structured metadata in the existing metadata table is less than or equal to a preset similarity.

[0111] Optionally, the device further comprises:

[0112] A number acquisition module, used to acquire the corresponding numbers of the inspection object type structured metadata, the inspection item type structured metadata, the inspection method type structured metadata and the inspection indicator type structured metadata in the existing metadata table;

[0113] The inspection standard library relationship table forming module is used to replace the inspection object class structured metadata, inspection item class structured metadata, inspection method class structured metadata and inspection indicator class structured metadata in the tree structure with the corresponding numbers respectively to form an inspection standard library relationship table, and store the inspection standard library relationship table in the inspection standard library.

[0114] Optionally, the splitting and classification module C2 includes:

[0115] The splitting and classification submodule is used to adopt a named entity recognition model to split the natural text paragraph into multiple structured metadata according to text semantics, and divide each structured metadata in the multiple structured metadata into inspection object class structured metadata, inspection item class structured metadata, inspection method class structured metadata and inspection indicator class structured metadata.

[0116] Optionally, the association relationship establishing module C3 includes:

[0117] The association relationship establishment submodule is used to adopt the bidirectional encoding model of Transformer to establish the association relationship corresponding to the inspection object class structured metadata based on the inspection object class structured metadata, inspection item class structured metadata, inspection method class structured metadata and inspection indicator class structured metadata.

[0118] Optionally, the conversion module C1 includes:

[0119] The conversion submodule is used to convert the inspection standard text into natural text paragraphs through a text table parsing tool.

[0120] Optionally, the tree structure establishing module C4 includes:

[0121] The tree structure establishment submodule is used to establish a tree structure corresponding to the structured metadata of the inspection object class based on the association relationship by adopting an object nesting method.

[0122] The processing device of the inspection standard text can refer to the relevant records of the aforementioned processing method of the inspection standard text, and can achieve the same or similar beneficial effects. In order to avoid repetition, it will not be described here.

[0123] The present application also provides an electronic device, comprising: a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, any one of the aforementioned methods for processing inspection standard texts is implemented.

[0124] The present application also provides a readable storage medium, when the instructions or transactions in the storage medium are executed by the processor of the electronic device, the electronic device can execute any of the aforementioned methods for processing inspection standard texts.

[0125] The present application also provides a computer program product, including instructions or transactions, which, when executed by a processor in an electronic device, enable the electronic device to execute any of the aforementioned methods for processing inspection standard texts.

[0126] It should be noted that the control system, electronic device, readable storage medium and computer program product can all refer to the description of the aforementioned control method and have the same or similar beneficial effects. In order to avoid repetition, they will not be described here.

[0127] It should be noted that, for the sake of simplicity, the method embodiments are described as a series of action combinations, but those skilled in the art should be aware that the embodiments of the present invention are not limited by the order of the actions described, because according to the embodiments of the present invention, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions involved are not necessarily required by the embodiments of the present invention.

[0128] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the application disclosed herein. The present application is intended to cover any modification, use or adaptation of the present application, which follows the general principles of the present application and includes common knowledge or customary techniques in the art that are not disclosed in the present disclosure. The specification and examples are intended to be exemplary only, and the true scope and spirit of the present application are indicated by the following claims.

[0129] It should be understood that the present application is not limited to the precise structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present application is limited only by the appended claims.

Claims

1. A method for processing a test standard text, characterized in that: include: Convert the standard text into natural text paragraphs; Splitting the natural text paragraph into a plurality of structured metadata according to text semantics, and dividing each of the plurality of structured metadata into inspection object structured metadata, inspection item structured metadata, inspection method structured metadata, and inspection indicator structured metadata; Establishing the association relationship corresponding to the inspection object class structured metadata based on the inspection object class structured metadata, the inspection item class structured metadata, the inspection method class structured metadata and the inspection indicator class structured metadata; A tree structure corresponding to the structured metadata of the inspection object class is established based on the association relationship.

2. The method for processing inspection standard text according to claim 1, characterized in that: Also includes: Performing similarity identification on each of the structured metadata and the existing structured metadata in the existing metadata table; For one of the structured metadata, when the similarity between the structured metadata and the existing structured metadata in the existing metadata table is less than or equal to a preset similarity, the structured metadata is stored in the existing metadata table.

3. The method for processing inspection standard text according to claim 2, characterized in that: Also includes: Obtain the inspection object class structured metadata, inspection item class structured metadata, inspection method class structured metadata, and inspection indicator class structured metadata, and their corresponding numbers in the existing metadata table; In the tree structure, the inspection object class structured metadata, the inspection item class structured metadata, the inspection method class structured metadata and the inspection indicator class structured metadata are respectively replaced with the corresponding numbers to form an inspection standard library relationship table, and the inspection standard library relationship table is stored in the inspection standard library.

4. The method for processing the inspection standard text according to any one of claims 1 to 3, characterized in that: The step of splitting the natural text paragraph into a plurality of structured metadata according to text semantics, and dividing each of the plurality of structured metadata into inspection object structured metadata, inspection item structured metadata, inspection method structured metadata, and inspection indicator structured metadata includes: A named entity recognition model is adopted to split the natural text paragraph into multiple structured metadata according to text semantics, and each structured metadata in the multiple structured metadata is divided into inspection object type structured metadata, inspection item type structured metadata, inspection method type structured metadata and inspection indicator type structured metadata.

5. The method for processing the inspection standard text according to any one of claims 1 to 3, characterized in that: The establishing of the association relationship corresponding to the inspection object class structured metadata based on the inspection object class structured metadata, the inspection item class structured metadata, the inspection method class structured metadata and the inspection indicator class structured metadata includes: The bidirectional encoding model of Transformer is adopted to establish the association relationship corresponding to the structured metadata of the inspection object class based on the structured metadata of the inspection object class, the structured metadata of the inspection item class, the structured metadata of the inspection method class and the structured metadata of the inspection indicator class.

6. The method for processing the inspection standard text according to any one of claims 1 to 3, characterized in that: The step of converting the inspection standard text into a natural text paragraph includes: The inspection standard text is converted into natural text paragraphs through text table parsing tools.

7. The method for processing the inspection standard text according to any one of claims 1 to 3, characterized in that: The step of establishing a tree structure corresponding to the structured metadata of the inspection object class based on the association relationship includes: An object nesting method is adopted to establish a tree structure corresponding to the structured metadata of the inspection object class based on the association relationship.

8. A processing device for checking standard text, characterized in that: include: A conversion module, used to convert the inspection standard text into natural text paragraphs; A splitting and classification module, used for splitting the natural text paragraph into a plurality of structured metadata according to text semantics, and dividing each of the plurality of structured metadata into inspection object type structured metadata, inspection item type structured metadata, inspection method type structured metadata and inspection indicator type structured metadata; An association relationship establishment module, used to establish an association relationship corresponding to the inspection object class structured metadata based on the inspection object class structured metadata, the inspection item class structured metadata, the inspection method class structured metadata and the inspection indicator class structured metadata; The tree structure establishment module is used to establish a tree structure corresponding to the structured metadata of the inspection object class based on the association relationship.

9. An electronic device, characterized in that: include: A processor, a memory, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the method for processing the inspection standard text as described in any one of claims 1 to 7 is implemented.

10. A readable storage medium, characterized in that: When the instructions or transactions in the storage medium are executed by a processor of an electronic device, the electronic device is enabled to execute the method for processing a test standard text according to any one of claims 1 to 7.

11. A computer program product, characterized in that It comprises instructions or transactions, which, when executed by a processor in an electronic device, enable the electronic device to execute the method for processing the inspection standard text according to any one of claims 1 to 7.

Images