Standard text analysis method and device, computer equipment, medium and program product

Abstract

The invention relates to a standard text analysis method and device, computer equipment, a medium and a program product. The normative text analysis method comprises the steps of obtaining text information of a normative text in response to a detected input operation of the normative text; converting the text information into to-be-screened information according to a preset conversion mode; displaying first prompt information, wherein the first prompt information is used for displaying at least one piece of first-level information to a user; in response to the detected input operation of the first prompt information, second prompt information is displayed, and the second prompt information is used for displaying at least one piece of second-level information to the user; and in response to the detected input operation of the second prompt information, outputting an analysis result according to the degree of correlation between the input first prompt information and the to-be-screened information as well as between the input second prompt information and the to-be-screened information. By displaying the first prompt information and the second prompt information in sequence, the analysis requirement of the user can be determined hierarchically, and the analysis requirement is matched with the to-be-screened information to determine the analysis result of the power grid specification text, so that the reliability of the analysis result is improved.

Description

Technical Field

[0001] This disclosure relates to the field of power grid technology, and in particular to a method, apparatus, computer equipment, medium, and program product for parsing standard text. Background Technology

[0002] Currently, power grid specifications stipulate the mandatory standards that grid-connected equipment must meet to ensure the stability and security of the power grid. However, due to differences in the specifications of power grids in different countries, users may encounter inaccurate results when interpreting these specifications. Summary of the Invention

[0003] To overcome the problems existing in related technologies, this disclosure provides a standard text parsing method, apparatus, computer equipment, medium, and program product.

[0004] According to a first aspect of the present disclosure, a canonical text parsing method is provided, the canonical text parsing method comprising:

[0005] In response to the detection of input of canonical text, the text information of the canonical text is obtained;

[0006] According to a preset conversion method, the text information is converted into information to be filtered;

[0007] Display a first prompt message, the first prompt message being used to display at least one first-level information to the user;

[0008] In response to the detection of the input operation of the first prompt information, a second prompt information is displayed, the second prompt information being used to display at least one second-level information to the user;

[0009] In response to the input operation that detects the second prompt information, the parsing result is output based on the relevance of the input first prompt information and the input second prompt information to the information to be filtered;

[0010] Each of the first-level information corresponds to at least one second-level information.

[0011] In some embodiments of this disclosure, the preset conversion method includes a vector conversion method and / or a keyword determination method; the information to be filtered includes first vector information and / or first keyword information.

[0012] The step of converting the text information into information to be filtered according to a preset conversion method includes:

[0013] According to the vector conversion method, the text information is converted into at least one of the first vector information;

[0014] Based on the keyword determination method, at least one of the first keyword information is determined from the text information;

[0015] In response to the input operation that detects the second prompt information, the parsing result is output based on the relevance of the input first prompt information, the input second prompt information, and the information to be filtered, including:

[0016] Based on the degree of relevance between the input first prompt information and the input second prompt information and each of the first vector information and / or each of the first keyword information, the parsing result is output.

[0017] In some embodiments of this disclosure, the step of responding to the input operation of detecting the second prompt information and outputting a parsing result based on the relevance of the input first prompt information and the input second prompt information to the information to be filtered includes:

[0018] Based on the input first prompt information, the input second prompt information, and each of the first vector information, determine the first matching degree of each of the first vector information;

[0019] Based on the input first prompt information, the input second prompt information, and each of the first keyword information, determine the second matching degree of each of the first keyword information;

[0020] The target text information of the standardized text is determined based on the text information corresponding to each of the first matching degrees and the text information corresponding to each of the second matching degrees;

[0021] Based on the target text information, output the parsing result.

[0022] In some embodiments of this disclosure, converting the text information into at least one of the first vector information according to the vector conversion method includes:

[0023] The text information is input into a vectorization model, the vectorization model is used to analyze the text information, and at least one first vector information corresponding to the text information is output.

[0024] In some embodiments of this disclosure, before converting the text information into information to be filtered according to a preset conversion method, the method further includes:

[0025] According to a preset segmentation method, each piece of text information in the text information is divided into at least one segment information;

[0026] The step of converting the text information into information to be filtered according to a preset conversion method includes:

[0027] According to the vector conversion method, each of the fragment information is converted into at least one of the first vector information;

[0028] Based on the keyword determination method, at least one of the first keyword information is determined from each of the segment information.

[0029] In some embodiments of this disclosure, before obtaining the text information of the standardized text, the method further includes:

[0030] Identify the language text to be converted from the specified text;

[0031] Convert the text to be converted into the target language text;

[0032] The step of obtaining text information of the standardized text in response to the detection of input of standardized text includes:

[0033] Obtain the text information of the target language text.

[0034] In some embodiments of this disclosure, the step of responding to the input operation of detecting the second prompt information and outputting a parsing result based on the relevance of the input first prompt information and the input second prompt information to the information to be filtered includes:

[0035] According to the vector conversion method, the input first prompt information and the input second prompt information are converted into second vector information;

[0036] Based on the keyword determination method, the second keyword information is determined from the first input prompt information and the second input prompt information;

[0037] The parsing result is output based on the degree of relevance between the second vector information and / or the second keyword information and the information to be filtered.

[0038] In some embodiments of this disclosure, the step of responding to the input operation of detecting the second prompt information and outputting a parsing result based on the relevance of the input first prompt information and the input second prompt information to the information to be filtered includes:

[0039] The first input prompt, the second input prompt, and the information to be filtered are input into a large language model. The large language model analyzes the first input prompt, the second input prompt, and the information to be filtered, and outputs the parsing result.

[0040] In some embodiments of this disclosure, the step of responding to the input operation of detecting the second prompt information and outputting the parsing result based on the relevance of the input first prompt information and the input second prompt information to the information to be filtered includes:

[0041] Output the target text information of the specified text;

[0042] The target text information is output in a preset language.

[0043] Output the confidence level of the target text information.

[0044] According to a second aspect of the present disclosure, a canonical text parsing apparatus is provided, the canonical text parsing apparatus comprising:

[0045] An acquisition module is configured to acquire text information of the standard text in response to an input operation that detects the standard text.

[0046] A conversion module is configured to convert the text information into information to be filtered according to a preset conversion method.

[0047] A first display module is configured to display first prompt information, the first prompt information being used to display at least one first-level information to the user;

[0048] A second display module is configured to display a second prompt message in response to an input operation that detects the first prompt message. The second prompt message is used to display at least one second-level information to the user.

[0049] An output module is configured to respond to an input operation that detects the second prompt information, and output a parsing result based on the relevance of the input first prompt information and the input second prompt information to the information to be filtered;

[0050] Each of the first-level information corresponds to at least one second-level information.

[0051] According to a third aspect of the present disclosure, a computer device is provided, the computer device comprising:

[0052] processor;

[0053] Memory used to store the processor's executable instructions;

[0054] The processor is configured to execute the canonical text parsing method described above.

[0055] According to a fourth aspect of the present disclosure, a non-transitory computer-readable storage medium is provided, which, when instructions in the storage medium are executed by a processor of a terminal, enables the terminal to perform the canonical text parsing method as described above.

[0056] According to a fifth aspect of the present disclosure, a computer program product is provided, including a computer program or instructions, which, when executed by a processor, implement the canonical text parsing method as described above.

[0057] The technical solutions provided by the embodiments of this disclosure may include the following beneficial effects:

[0058] In response to the detection of input of standard text, the system acquires the text information of the standard text to obtain the power grid standard that the user needs to parse. According to a preset conversion method, the text information is converted into at least one piece of information to be filtered, so as to match its relevance with the information the user needs to parse. A first prompt is displayed to indicate the category of information that can be parsed. In response to the detection of input of the first prompt, a second prompt is displayed to indicate the specific information that can be parsed under that information category. In response to the detection of input of the second prompt, the system outputs the parsing result based on the relevance of the input first and second prompts to the information to be filtered, thus determining and outputting the parsing result based on the relevance of the specific information under the information category to the information to be filtered. By sequentially displaying the first and second prompts and outputting the parsing result based on the relevance of the input first and second prompts to the information to be filtered, the system can determine the user's parsing needs hierarchically and match them with the information to be filtered to determine the parsing result of the power grid standard text, thereby improving the reliability of the parsing result.

[0059] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0060] The accompanying drawings, which are incorporated in and form a part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure.

[0061] Figure 1 This is a flowchart illustrating a canonical text parsing method according to an exemplary embodiment;

[0062] Figure 2 This is a flowchart illustrating a canonical text parsing method according to another exemplary embodiment;

[0063] Figure 3 This is a flowchart illustrating a canonical text parsing method according to another exemplary embodiment;

[0064] Figure 4 This is a flowchart illustrating a canonical text parsing method according to another exemplary embodiment;

[0065] Figure 5 This is a flowchart illustrating a canonical text parsing method according to another exemplary embodiment;

[0066] Figure 6 This is a flowchart illustrating a canonical text parsing method according to another exemplary embodiment;

[0067] Figure 7 This is a flowchart illustrating a canonical text parsing method according to another exemplary embodiment;

[0068] Figure 8 This is a block diagram illustrating a canonical text parsing apparatus according to an exemplary embodiment.

[0069] In the picture:

[0070] 100 - Acquisition module; 200 - Conversion module; 300 - First display module; 400 - Second display module; 500 - Output module. Detailed Implementation

[0071] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this disclosure. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this disclosure as detailed in the appended claims. It should also be understood that the term “and / or” as used in this disclosure refers to any or all possible combinations including one or more of the associated listed items.

[0072] Currently, power grid specifications stipulate mandatory standards that grid-connected equipment (such as power generation equipment, loads, and energy storage systems) must meet when connecting to the grid. These standards include connection technology requirements, operation and management requirements, protection system requirements, and testing requirements, to ensure the stability, security, and quality of transmitted power. Connection technology requirements include the voltage / frequency tolerance range, active / reactive power regulation capabilities, and fault ride-through capabilities of the grid-connected equipment. Operation and management requirements include dispatch command response speed and reserve capacity range. Protection system requirements include relay protection operating time and grounding methods. Testing requirements include simulation verification and low-voltage ride-through testing.

[0073] One embodiment of this disclosure provides a method for parsing standard text. After obtaining the standard text of a power grid, the method acquires the parsing requirements input by the user. Based on the user's parsing requirements, the method determines the paragraph information in the standard text and outputs this paragraph information as the parsing result. However, the user's input question may be redundant, such as "extract the frequency range of the power grid, only the frequency range of wind turbine equipment types." The determined relevant paragraphs may include: paragraph information related to "frequency," paragraph information related to "range," paragraph information related to "power grid," paragraph information related to "wind turbine," paragraph information related to "equipment," and paragraph information related to "type," etc. The output paragraph information is quite disorganized, and there are paragraphs that are essentially irrelevant to the user's question as parsing results, leading to inaccuracies in the parsing results. In such cases, users usually need to perform secondary checks, reducing the parsing efficiency of the standard text.

[0074] This disclosure provides another method for parsing standard text, such as... Figure 1 As shown, the method includes:

[0075] S100: In response to the detection of input operation of standard text, obtain the text information of the standard text.

[0076] S200. Convert the text information into information to be filtered according to the preset conversion method.

[0077] S300, Display first prompt information, the first prompt information is used to display at least one first-level information to the user.

[0078] S400, in response to the detection of the input operation of the first prompt information, a second prompt information is displayed, the second prompt information being used to display at least one second-level information to the user.

[0079] S500: In response to the input operation that detects the second prompt information, output the parsing result based on the relevance of the input first prompt information and the input second prompt information to the information to be filtered.

[0080] Each first-level information corresponds to at least one second-level information.

[0081] In this embodiment, in response to the detection of a standard text input operation, the text information of the standard text is obtained to acquire the power grid standard that the user needs to parse. According to a preset conversion method, the text information is converted into at least one piece of information to be filtered, so as to match its relevance with the information the user needs to parse. A first prompt is displayed to indicate the category of information that can be parsed. In response to the detection of the first prompt input operation, a second prompt is displayed to indicate the specific information that can be parsed under that information category. In response to the detection of the second prompt input operation, based on the relevance of the input first and second prompts to the information to be filtered, a parsing result is output, thereby determining and outputting the parsing result based on the relevance of the specific information under the information category to the information to be filtered. By sequentially displaying the first and second prompts and outputting the parsing result based on the relevance of the input first and second prompts to the information to be filtered, the user's parsing needs can be determined hierarchically, and matched with the information to be filtered to determine the parsing result of the power grid standard text, thereby improving the reliability of the parsing result.

[0082] For example, users can enter the specification text by uploading a document, uploading an image, or entering a link.

[0083] For example, both the first and second prompts are preset information, which may include connection technology requirements, operation and management requirements, protection system requirements, and testing requirements. For instance, the first prompt may include "extract frequency range" and "extract grounding method," while the second prompt may include "power generation equipment" and "wind turbine equipment." It is understood that the first prompt may also include "wind turbine equipment" and "power generation equipment," and the second prompt may also include "extract frequency range" and "extract grounding method." As long as the first and second prompts form a hierarchical relationship, it is acceptable; further details are omitted here.

[0084] For example, when the first displayed prompt includes "Extract Frequency Range" and "Extract Grounding Method," and the user enters "Extract Frequency Range," the second displayed prompt includes "Power Generation Equipment" and "Wind Turbine Equipment." When the user enters "Wind Turbine Equipment," the system outputs analysis results based on the relevance of "Extract Frequency Range" and "Wind Turbine Equipment" to the information to be filtered. It is understood that each output analysis result includes both "Frequency Range" and "Wind Turbine Equipment," improving the reliability of the analysis results, eliminating the need for secondary checks by the user, and increasing the efficiency of the analysis process.

[0085] For example, in response to an input operation that detects a first prompt, first text information and second display information in the standard text related to the first prompt can be displayed. In response to an input operation that detects a second prompt, the displayed first text information is filtered to determine and display second text information that is related to both the input first and second prompts. Then, based on the degree of relevance between the second text information and the filtered information, the parsing result is output.

[0086] For example, in response to an input operation that detects a first prompt, the first text information related to the first prompt may be temporarily withheld, and a second prompt may be displayed directly. Upon detecting an input operation that detects a second prompt, the second text information related to both the first and second prompts is determined and displayed. Then, based on the relevance of the second text information to the filtered information, the parsing result is output.

[0087] It is understood that in step S500, in response to the input operation that detects the second prompt information, a third prompt information may also be displayed. The third prompt information is used to display at least one third-level information to the user. In response to the input operation that detects the third prompt information, the parsing result is output based on the relevance of the input first prompt information, the input second prompt information, and the input third prompt information to the information to be filtered. Each second-level information corresponds to at least one first- or third-level information. Alternatively, some second-level information may correspond to at least one third-level information, while other second-level information may not have corresponding third-level information. It is understood that fourth and fifth prompt information may also be included, etc., which will not be elaborated here.

[0088] For example, an index box can be displayed when or before the first prompt message is displayed. When the user clicks on the first and second prompt messages, the entered first and second prompt messages will be displayed in the index box. The user's click on the first and second prompt messages can be detected.

[0089] In one embodiment, such as Figure 2 As shown, the preset conversion methods include vector conversion and / or keyword determination. The information to be filtered includes first vector information and / or first keyword information. In step S200, converting text information into information to be filtered according to the preset conversion methods can be determined in the following way:

[0090] S210. According to the vector conversion method, convert the text information into at least one first vector information.

[0091] S220. Based on the keyword determination method, determine at least one first keyword from the text information.

[0092] In step S500, in response to the input operation that detects the second prompt information, the output parsing result can be determined as follows, based on the relevance of the input first prompt information, the input second prompt information, and the information to be filtered:

[0093] Based on the correlation between the first and second input prompts and each first vector and / or each first keyword, the parsing result is output.

[0094] In this embodiment, text information is converted into at least one first vector information according to a vector transformation method, thereby transforming the text information at the vector level. At least one first keyword information is determined from the text information according to a keyword determination method, thereby determining the keywords in the text information at the keyword level. By outputting the parsing result based on the relevance of the input first prompt information, the input second prompt information, and each first vector information and / or each first keyword information, the relevance between the user's parsing requirements and each first vector information and / or each first keyword information can be determined from different levels, thus identifying the portion of text information that meets the parsing requirements and improving the reliability of the parsing result.

[0095] For example, keyword identification methods are used to identify technical terms in text information. Vector conversion methods can convert sentences in text information into vectors, facilitating the matching of relevance to the user's parsing requirements.

[0096] In one embodiment, such as Figure 3 As shown, in step S500, in response to the input operation that detects the second prompt information, the output parsing result can also be determined in the following way based on the relevance of the input first prompt information, the input second prompt information, and the information to be filtered:

[0097] S510. Based on the first input prompt information, the second input prompt information, and each first vector information, determine the first matching degree of each first vector information.

[0098] S520. Based on the first prompt information, the second prompt information, and each first keyword information, determine the second matching degree of each first keyword information.

[0099] S530. Determine the target text information of the standardized text based on the text information corresponding to each first matching degree and each second matching degree.

[0100] S540. Output the parsing results based on the target text information.

[0101] In this embodiment, based on the input first prompt information, the input second prompt information, and each first vector information, a first matching degree of each first vector information is determined to determine the degree of matching between the user's parsing requirements and each first vector information. Based on the input first prompt information, the input second prompt information, and each first keyword information, a second matching degree of each first keyword information is determined to determine the degree of matching between the user's parsing requirements and each first keyword information. Based on the text information corresponding to each first matching degree and each second matching degree, the target text information of the standardized text is determined. Then, based on the partial text information corresponding to both the first keyword information and the first vector information with high matching degrees, the target text information in the standardized text with a high similarity to the parsing requirements is determined. Based on the target text information, the parsing result is output to display the parsing result to the user. By determining the target text information based on the matching degree between the user's parsing requirements and each first keyword and each first vector information, the target text information with a high matching degree in the standardized text can be determined from two levels, thereby improving the reliability of the parsing result.

[0102] For example, when each first vector information includes first sub-vector information, second sub-vector information, and third sub-vector information, and each first keyword information includes first sub-keyword information, second sub-keyword information, and third sub-keyword information, the matching degree of the first sub-vector information is determined to be 10%, the matching degree of the second sub-vector information is 20%, and the matching degree of the third sub-vector information is 90%. The matching degree of the first sub-keyword information is determined to be 2%, the matching degree of the second sub-keyword information is 10%, and the matching degree of the third sub-vector information is 90%. When the first sub-vector information and the second sub-keyword information correspond to the same text statement (hereinafter referred to as the first text statement), and when the third sub-vector information and the third sub-keyword information correspond to the same text statement (hereinafter referred to as the second text statement), the second sub-vector information corresponds to the third text statement, and the first keyword information corresponds to the fourth text statement. Then, the matching degree of the above text statements is weighted according to the weights of the first vector information and the first keyword information (taking 50% each as an example). Therefore, the matching degree of the first text statement is 10%, the matching degree of the second text statement is 90%, the matching degree of the third text statement is 10%, and the matching degree of the fourth text statement is 1%. The matching scores are sorted, and the second text statement with the highest matching score is output.

[0103] For example, the above text statements can also be displayed in a list format according to their matching degree for user reference. The weighting ratios described above are merely examples, and different weighting ratios can be assigned based on the accuracy of the vector conversion method and the keyword determination method.

[0104] In one embodiment, the conversion of text information into at least one first vector information according to the vector conversion method in step S210 is determined in the following way:

[0105] The text information is input into a vectorization model, which analyzes the text information and outputs at least one first vector information corresponding to the text information.

[0106] In this embodiment, since the vectorization model can convert the input text information into vector information with high efficiency and accuracy, the reliability of the vector conversion method is improved by converting it into the first vector information through the vectorization model.

[0107] In one embodiment, such as Figure 4 As shown, before converting the text information into the information to be filtered according to the preset conversion method in step S200, the method further includes:

[0108] According to the preset segmentation method, each piece of text information is divided into at least one segment.

[0109] The step S200, which converts text information into information to be filtered according to a preset conversion method, can also be determined in the following way:

[0110] S230. According to the vector conversion method, convert each segment of information into at least one first vector information.

[0111] S240. Based on the keyword determination method, determine at least one primary keyword from each segment of information.

[0112] In this embodiment, since the text information contains multiple clauses, each clause in the text information is divided into at least one segment. The vector conversion method and the keyword determination method can process each segment, thereby improving the efficiency and accuracy of the vector conversion method and the keyword determination method in processing the text information.

[0113] For example, one or more sentences in each article can be divided into a segment. Alternatively, each paragraph in the text can be divided into at least one segment, and one or more sentences in each paragraph can be divided into a segment.

[0114] In one embodiment, such as Figure 5 As shown, in step S500, in response to the input operation that detects the second prompt information, the output parsing result can also be determined in the following way based on the relevance of the input first prompt information, the input second prompt information, and the information to be filtered:

[0115] S550. According to the vector conversion method, convert the input first prompt information and the input second prompt information into second vector information.

[0116] S560. Determine the second keyword information from the first input prompt information and the second input prompt information according to the keyword determination method.

[0117] S570. Output the parsing results based on the relevance of each second vector information and / or each second keyword information to the information to be filtered.

[0118] In this embodiment, based on a vector transformation method, the input first and second prompts are converted into multiple second vector information to transform the parsing requirements at the vector level. Based on a keyword determination method, multiple second keyword information is determined from the input first and second prompts to determine the keywords in the parsing requirements at the keyword level. By outputting the parsing results based on the relevance of each second vector information and / or each second keyword information to the information to be filtered, the relevance between the user's parsing requirements and the text information can be determined from different levels, thus identifying the portion of text information that meets the parsing requirements and improving the reliability of the parsing results.

[0119] For example, since vector transformation can convert text information, first prompt information, and second prompt information into vector information, semantically similar statements are close to each other in the vector space, while semantically dissimilar statements are far apart. By converting text information into first vector information and first and second prompt information into second vector information, the part of text information most relevant to the parsing requirements can be determined based on the proximity of the first and second vector information (generally represented by cosine). Similarly, since keyword determination can identify technical terms in both text information and parsing requirements, by determining the first keyword information from the text information and the second keyword information from the first and second prompt information, the part of text information most relevant to the parsing requirements can be determined based on the similarity between the first and second keywords.

[0120] For example, statement A in the specification text is converted into vector a through vectorization, statement B is converted into vector b through vectorization, and the first and second prompts are converted into vector c through vectorization. By comparing the cosine values ​​of vector a and vector c, as well as the cosine values ​​of vector b and vector c, it can be determined whether statement A or statement B is more relevant to the parsing requirements.

[0121] For example, statement A in the standard text is identified using the keyword identification method, which determines keyword m. Statement B in the standard text is identified using the same method, which determines keyword n. The first and second prompts are identified using the same method, which determines keyword k. By comparing the relevance of keywords m and k, and the relevance of keywords n and k, it can be determined whether statement A or statement B is more relevant to the parsing requirements.

[0122] For example, the second vector information may contain one or more vector information, and the second keyword information may contain one or more keyword information.

[0123] In one embodiment, such as Figure 6 As shown, before obtaining the text information of the standard text in step S100, the method further includes:

[0124] S600: Identify the language text to be converted from the standard text.

[0125] S700: Convert the text to be converted into the target language text.

[0126] In step S100, in response to the detection of canonical text input, the text information of the canonical text is determined as follows:

[0127] Obtain text information in the target language.

[0128] In this embodiment, since different countries use different languages, the language text to be converted in the standard text is identified to determine the language type of the standard text that the user needs to parse. The language text to be converted is then converted into the target language text, so that standard texts in different languages ​​are all converted into standard texts in the same language. This reduces the difficulty of converting text information into information to be filtered and matching the degree of relevance between the information to be filtered and the first and second prompts, thereby improving the efficiency and reliability of the standard text parsing method.

[0129] For example, the target language text can be either English or Chinese. The first and second prompt messages can be in the same language as the target language text.

[0130] In one embodiment, in step S500, in response to the input operation that detects the second prompt information, the output parsing result can be determined in the following way based on the relevance of the input first prompt information and the input second prompt information to the information to be filtered:

[0131] Input the first prompt, the second prompt, and the information to be filtered into the large language model. The large language model analyzes the first prompt, the second prompt, and the information to be filtered, and outputs the parsing results.

[0132] In this embodiment, since the large language model can deeply understand language text, it can understand, analyze and compare the first prompt information, the second prompt information and the information to be filtered by the large language model, and determine the information to be filtered with the highest degree of relevance to the first prompt information and the second prompt information as the parsing result and output it, thereby improving the reliability of the output result.

[0133] For example, in step S500, in response to the input operation that detects the second prompt information, the output parsing result can also be determined in the following way based on the relevance of the input first prompt information and the input second prompt information to the information to be filtered:

[0134] Output the target text information of the standard text.

[0135] Output the target text information in a preset language.

[0136] Output the confidence level of the target text information.

[0137] In this embodiment, the target text information of the standard text is output to display the original content of the standard text corresponding to the parsing requirements to the user. The target text information is output in a preset language to display the target text information to the user in the preset language. The confidence level of the target text information is output to display the reliability of the target text information corresponding to the parsing requirements to the user. Since the user's commonly used language may differ from the language of the standard text, outputting the target text information in a preset language allows the meaning of the target text information to be displayed to the user in their commonly used language. Furthermore, by outputting the confidence level, the reliability of the output target text information can be displayed to the user. If the confidence level is low, the user can re-verify whether the target text information corresponds to the parsing requirements; if the confidence level is high, the user can directly extract relevant information without verification, thereby improving the reliability of the standard text parsing.

[0138] For example, users can set the language of the preset text according to their needs to meet the usage requirements of users in different countries. The language of the first and second prompt messages can be the same as the language of the preset text.

[0139] This disclosure provides an exemplary embodiment of a canonical text parsing method, such as... Figure 7 As shown, canonical text parsing methods include:

[0140] S10. In response to the detection of input operation of canonical text, identify the language text to be converted from the canonical text.

[0141] S11. Convert the text to be converted into the target language text.

[0142] S12. According to the preset segmentation method, divide each piece of information in the target language text into at least one segment.

[0143] S13. Input each segment information into the vectorization model, analyze each segment information using the vectorization model, and output at least one first vector information corresponding to each segment information.

[0144] S14. Based on the keyword determination method, determine at least one primary keyword from each segment of information.

[0145] S15. Display first prompt information, which is used to display at least one first-level information to the user.

[0146] S16. In response to the detection of the input operation of the first prompt information, a second prompt information is displayed, the second prompt information being used to display at least one second-level information to the user.

[0147] S17. In response to the input operation of detecting the second prompt information, input the first prompt information and the second prompt information into the vectorization model, analyze the first prompt information and the second prompt information using the vectorization model, and output the second vector information.

[0148] S18. Based on the keyword determination method, determine the second keyword information from the first input prompt information and the second input prompt information.

[0149] S19. Input each first vector information, each first keyword information, the second vector information, and the second keyword information into the large language model, and use the large language model to analyze the first vector information, each first keyword information, the second vector information, and the second keyword information.

[0150] S20, the target text information of the large language model output standard text, the target text information output in preset language text, and the confidence level of the output target text information.

[0151] In this embodiment, by sequentially displaying the first and second prompts and outputting the parsing results based on the relevance of the input first and second prompts to the information to be filtered, the user's parsing needs can be determined hierarchically, and matched with the information to be filtered to determine the parsing results of the power grid specification text, thereby improving the reliability of the parsing results.

[0152] In one exemplary embodiment, a canonical text parsing apparatus is provided for implementing the method described above. (Reference) Figure 8 As shown, the specification text parsing device may include an acquisition module 100, a conversion module 200, a first display module 300, a second display module 400, and an output module 500. During the implementation of the above method,

[0153] The acquisition module 100 is configured to acquire the text information of the canonical text in response to the detection of input of canonical text.

[0154] The conversion module 200 is configured to convert text information into information to be filtered according to a preset conversion method.

[0155] The first display module 300 is configured to display first prompt information, which is used to display at least one first-level information to the user.

[0156] The second display module 400 is configured to display a second prompt message in response to an input operation that detects the first prompt message. The second prompt message is used to display at least one second-level information to the user.

[0157] The output module 500 is configured to respond to an input operation that detects a second prompt message, and output the parsing result based on the relevance of the input first prompt message and the input second prompt message to the information to be filtered.

[0158] Each first-level information corresponds to at least one second-level information.

[0159] In one exemplary embodiment, a canonical text parsing apparatus is provided, wherein a conversion module 200 is configured to:

[0160] According to the vector transformation method, the text information is converted into at least one first vector information.

[0161] Based on the keyword determination method, at least one primary keyword is determined from the text information.

[0162] In one exemplary embodiment, a canonical text parsing apparatus is provided, wherein an output module 500 is configured to:

[0163] Based on the correlation between the first and second input prompts and each first vector and / or each first keyword, the parsing result is output.

[0164] In one exemplary embodiment, a canonical text parsing apparatus is provided, wherein an output module 500 is configured to:

[0165] Based on the first input prompt information, the second input prompt information, and each first vector information, determine the first matching degree of each first vector information.

[0166] Based on the first prompt information, the second prompt information, and each first keyword information, determine the second matching degree of each first keyword information.

[0167] Based on the text information corresponding to each first matching degree and each second matching degree, the target text information of the standardized text is determined.

[0168] Output the parsing results based on the target text information.

[0169] In one exemplary embodiment, a canonical text parsing apparatus is provided, wherein a conversion module 200 is configured to:

[0170] The text information is input into a vectorization model, which analyzes the text information and outputs at least one first vector information corresponding to the text information.

[0171] In one exemplary embodiment, a canonical text parsing apparatus is provided, wherein an acquisition module 100 is configured to:

[0172] According to the preset segmentation method, each piece of text information is divided into at least one segment.

[0173] In one exemplary embodiment, a canonical text parsing apparatus is provided, wherein an acquisition module 100 is configured to:

[0174] Based on the vector transformation method, each segment of information is converted into at least one first vector information.

[0175] Based on the keyword determination method, at least one primary keyword is determined from each segment of information.

[0176] In one exemplary embodiment, a canonical text parsing apparatus is provided, wherein an acquisition module 100 is configured to:

[0177] Identify the language text to be converted from the standard text.

[0178] Convert the text in the language to be converted into the target language text.

[0179] In one exemplary embodiment, a canonical text parsing apparatus is provided, wherein an acquisition module 100 is configured to:

[0180] Obtain text information in the target language.

[0181] In one exemplary embodiment, a canonical text parsing apparatus is provided, wherein an output module 500 is configured to:

[0182] Based on the vector transformation method, the first input prompt information and the second input prompt information are converted into second vector information.

[0183] Based on the keyword determination method, the second keyword information is determined from the first input prompt information and the second input prompt information.

[0184] Based on the relevance of the second vector information and / or the second keyword information to the information to be filtered, the analysis results are output.

[0185] In one exemplary embodiment, a canonical text parsing apparatus is provided, wherein an output module 500 is configured to:

[0186] Input the first prompt, the second prompt, and the information to be filtered into the large language model. The large language model analyzes the first prompt, the second prompt, and the information to be filtered, and outputs the parsing results.

[0187] In one exemplary embodiment, a canonical text parsing apparatus is provided, wherein an output module 500 is configured to:

[0188] Output the target text information of the standard text.

[0189] Output the target text information in a preset language.

[0190] Output the confidence level of the target text information.

[0191] In one exemplary embodiment, a computer device is provided, including a processor and a memory for storing processor-executable instructions. The processor is configured to perform the above-described canonical text parsing method.

[0192] In one exemplary embodiment, a non-transitory computer-readable storage medium including instructions is also provided, which, when executed by a terminal's processor, enable the terminal to perform the canonical text parsing method shown in the above embodiments or combinations thereof.

[0193] In one exemplary embodiment, a computer program product is provided, including a computer program or instructions that, when executed by a processor, implement the above-described canonical text parsing method.

[0194] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This disclosure is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this disclosure are indicated by the claims.

[0195] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this disclosure. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0196] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this disclosure, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0197] It should be understood that the various forms of processes shown above can be used to rearrange, add, or delete steps. For example, the steps described in this disclosure can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this disclosure can be achieved, and this is not limited herein.

[0198] The specific embodiments described above do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure should be included within the scope of protection of this disclosure.

Claims

1. A method for parsing standard text, characterized in that, The canonical text parsing method includes: In response to the detection of input of canonical text, the text information of the canonical text is obtained; According to a preset conversion method, the text information is converted into information to be filtered; Display a first prompt message, the first prompt message being used to display at least one first-level information to the user; In response to the detection of the input operation of the first prompt information, a second prompt information is displayed, the second prompt information being used to display at least one second-level information to the user; In response to the input operation that detects the second prompt information, the parsing result is output based on the relevance of the input first prompt information and the input second prompt information to the information to be filtered; Each of the first-level information corresponds to at least one second-level information.

2. The method for parsing standard text according to claim 1, characterized in that, The preset conversion method includes vector conversion method and / or keyword determination method; the information to be filtered includes first vector information and / or first keyword information; The step of converting the text information into information to be filtered according to a preset conversion method includes: According to the vector conversion method, the text information is converted into at least one of the first vector information; Based on the keyword determination method, at least one of the first keyword information is determined from the text information; In response to the input operation that detects the second prompt information, the parsing result is output based on the relevance of the input first prompt information, the input second prompt information, and the information to be filtered, including: Based on the degree of relevance between the input first prompt information and the input second prompt information and each of the first vector information and / or each of the first keyword information, the parsing result is output.

3. The standard text parsing method according to claim 2, characterized in that, In response to the input operation that detects the second prompt information, the parsing result is output based on the relevance of the input first prompt information and the input second prompt information to the information to be filtered, including: Based on the input first prompt information, the input second prompt information, and each of the first vector information, determine the first matching degree of each of the first vector information; Based on the input first prompt information, the input second prompt information, and each of the first keyword information, determine the second matching degree of each of the first keyword information; The target text information of the standardized text is determined based on the text information corresponding to each of the first matching degrees and the text information corresponding to each of the second matching degrees; Based on the target text information, output the parsing result.

4. The standard text parsing method according to claim 2, characterized in that, The step of converting the text information into at least one of the first vector information according to the vector conversion method includes: The text information is input into a vectorization model, the vectorization model is used to analyze the text information, and at least one first vector information corresponding to the text information is output.

5. The standard text parsing method according to claim 2, characterized in that, Before converting the text information into filterable information according to a preset conversion method, the method further includes: According to a preset segmentation method, each piece of text information in the text information is divided into at least one segment information; The step of converting the text information into information to be filtered according to a preset conversion method includes: According to the vector conversion method, each of the fragment information is converted into at least one of the first vector information; Based on the keyword determination method, at least one of the first keyword information is determined from each of the segment information.

6. The method for parsing standard text according to claim 1, characterized in that, Before obtaining the text information of the standardized text, the method further includes: Identify the language text to be converted from the specified text; Convert the text to be converted into the target language text; The step of obtaining text information of the standardized text in response to the detection of input of standardized text includes: Obtain the text information of the target language text.

7. The method for parsing standard text according to claim 1, characterized in that, In response to the input operation that detects the second prompt information, the parsing result is output based on the relevance of the input first prompt information and the input second prompt information to the information to be filtered, including: According to the vector conversion method, the input first prompt information and the input second prompt information are converted into second vector information; Based on the keyword determination method, the second keyword information is determined from the first input prompt information and the second input prompt information; The parsing result is output based on the degree of relevance between the second vector information and / or the second keyword information and the information to be filtered.

8. The method for parsing standard text according to claim 1, characterized in that, In response to the input operation that detects the second prompt information, the parsing result is output based on the relevance of the input first prompt information and the input second prompt information to the information to be filtered, including: The first input prompt, the second input prompt, and the information to be filtered are input into a large language model. The large language model analyzes the first input prompt, the second input prompt, and the information to be filtered, and outputs the parsing result.

9. The canonical text parsing method according to any one of claims 1 to 8, characterized in that, In response to the input operation that detects the second prompt information, the parsing result is output based on the relevance of the input first prompt information, the input second prompt information, and the information to be filtered, including: Output the target text information of the specified text; The target text information is output in a preset language. Output the confidence level of the target text information.

10. A standard text parsing device, characterized in that, The canonical text parsing device includes: An acquisition module is configured to acquire text information of the standard text in response to an input operation that detects the standard text. A conversion module is configured to convert the text information into information to be filtered according to a preset conversion method. A first display module is configured to display first prompt information, the first prompt information being used to display at least one first-level information to the user; A second display module is configured to display a second prompt message in response to an input operation that detects the first prompt message. The second prompt message is used to display at least one second-level information to the user. An output module is configured to respond to an input operation that detects the second prompt information, and output a parsing result based on the relevance of the input first prompt information and the input second prompt information to the information to be filtered; Each of the first-level information corresponds to at least one second-level information.

11. A computer device, characterized in that, The computer device includes: processor; Memory used to store the processor's executable instructions; The processor is configured to perform the canonical text parsing method as described in any one of claims 1 to 9.

12. A non-transitory computer-readable storage medium, characterized in that, When the instructions in the storage medium are executed by the processor of the terminal, the terminal is able to perform the canonical text parsing method as described in any one of claims 1 to 9.

13. A computer program product, comprising a computer program or instructions, characterized in that, When the computer program or instructions are executed by the processor, they implement the canonical text parsing method as described in any one of claims 1 to 9.

Images