Methods, apparatus, media and electronic devices for generating text content tags

Abstract

The present application relates to the technical field of data processing, and specifically provides a text content label generation method, device, medium and electronic equipment, aiming to solve the problem of cold start of existing labels and label data. For this purpose, a text content label generation method of the present application, the method comprising: obtaining a target text to be labeled; generating a recommended label based on the target text through a preset labeling model and displaying the recommended label; determining whether the recommended label is modified; if the recommended label is modified, performing quality inspection on the modified label; if the quality inspection is passed, taking the modified label as the label of the target text, and storing the corresponding relationship between the modified label and the target text in the database. The corresponding relationship between the modified label and the target text can be stored in the database, facilitating labeling when similar target texts appear, and avoiding the problem of label data cold start caused by the lack of suitable labels.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and specifically provides a method, apparatus, medium, and electronic device for generating text content tags. Background Technology

[0002] In the internet age, users create content on platforms, which then personalize and distribute this content to other users based on their understanding of the content. Content tagging is a primary method of content understanding; its advantages include structure and interpretability, and it can typically build multi-level tags to cover a wide range of content. A content tagging system is a platform that tags content, thus demonstrating its significant value.

[0003] Traditional content tagging systems have a strict synchronous process: users produce content on the platform, and the platform tags the content based on its understanding. Depending on whether users are aware of the tagging results, they can be divided into backend tagging and frontend tagging. Backend tagging is completed entirely in the system's backend; users are unaware of the tagging process or the results. Frontend tagging, on the other hand, provides the tagging results after the user finishes editing the content and before publishing, displaying them on the content editing page, allowing the user to perceive the results. However, both backend and frontend tagging are performed independently by the tagging system, and both suffer from cold start issues with tags and tag data. This requires defining which tags exist and manually labeling samples to train the model. When new tags appear, if they are not detected promptly, much content will not be covered by existing tags. Even if new tags are discovered, sufficient samples still need to be manually collected to train the model.

[0004] Accordingly, there is a need in this field for a new technical solution to address the aforementioned problems. Summary of the Invention

[0005] The present invention aims to solve the above-mentioned technical problems and address the cold start issue of existing tags and tag data.

[0006] This invention provides a method for generating text content tags, the method comprising:

[0007] Obtain the target text to be labeled;

[0008] Based on the target text, recommended tags are generated and displayed using a preset tagging model;

[0009] Determine whether the recommended tags have been modified;

[0010] If the recommended tags are modified, the modified tags will be subject to quality inspection.

[0011] If the quality inspection passes, the modified tag will be used as the tag of the target text, and the correspondence between the modified tag and the target text will be stored in the database.

[0012] With the above technical solution, the tagging model can generate and display recommended tags. Users can select recommended tags as tags for target text, avoiding cold start issues with tag data. Furthermore, if a user modifies a recommended tag, the modified tag undergoes quality inspection. By using the modified tag as the target text's tag, the target text can be annotated with the new, quality-inspected tag. The correspondence between modified tags and target text is stored in the database, facilitating annotation of similar target text when they appear, while also avoiding the cold start problem caused by a lack of suitable tags.

[0013] In a specific implementation of the above-mentioned text content tag generation method, the step of "generating and displaying recommendation tags based on the target text using a preset tagging model" further includes:

[0014] Based on the target text, select one or more first candidate tags from a preset tag library;

[0015] One or more of the first candidate tags are displayed to the user as the recommended tags.

[0016] In a specific implementation of the above-mentioned text content tag generation method, the step of "if the recommended tag is modified, then perform quality inspection on the modified tag" further includes:

[0017] Determine whether the modified tag belongs to the tag library;

[0018] If the modified tag belongs to the tag library, the modified tag is inspected using a preset quality inspection model.

[0019] When using the above technical solution, a quality inspection model is used to inspect the modified labels to ensure that the modified labels correspond to the target text.

[0020] In a specific implementation of the above text content tag generation method, the step of "if the recommended tag is modified, then the modified tag is subjected to quality inspection" further includes: if the modified tag does not belong to the tag library, then based on the modified tag, a second alternative tag is recommended through a preset machine recommendation model;

[0021] Determine whether the second alternative label is adopted;

[0022] Based on the assessment results, a second quality inspection will be conducted.

[0023] When using the above technical solution, if the modified tag does not belong to the tag library, other tags can be recommended through the machine recommendation model, so that the tag can accurately represent the text content and conform to the tag specification.

[0024] In a specific implementation of the above-mentioned text content tag generation method, the step of "performing quality inspection again based on the judgment result" further includes:

[0025] If the second candidate label is adopted, the second candidate label is subjected to quality inspection through the quality inspection model;

[0026] If the second alternative label is not adopted, the modified label will be subject to manual quality inspection.

[0027] In a specific implementation of the above-mentioned text content tag generation method, the step of "performing quality inspection on the second candidate tag through the quality inspection model" further includes:

[0028] The quality inspection model compares the target text with the second candidate label. If the target text and the second candidate label correspond, the quality inspection passes.

[0029] In a specific implementation of the above-described text content tag generation method, the method further includes:

[0030] If the quality inspection passes, the second alternative label is added to the label library.

[0031] When the above technical solution is adopted, the second alternative tag that passes the quality inspection can be added to the tag library, so that the tags in the tag library can cover more content and the annotation of text content can be more accurate.

[0032] In a specific implementation of the above text content tag generation method, the step of "determining whether the recommendation tag has been modified" further includes:

[0033] If the recommended tag is not modified, then the recommended tag will be used as the tag for the target text.

[0034] In a specific implementation of the above-mentioned text content tag generation method, the step of "generating and displaying recommended tags based on the target text using a preset tagging model" further includes:

[0035] During the target text editing process, the first candidate label is selected synchronously in the backend using the labeling model;

[0036] Once the target text has been edited, the first alternative label is displayed.

[0037] With the above technical solution, the tagging model can simultaneously tag the target text during editing, resulting in faster tagging speed and quicker display of recommended tags after editing is completed.

[0038] The present invention also discloses a text content tag generation device, comprising:

[0039] The acquisition module is configured to acquire the target text to be tagged;

[0040] The recommendation module is configured to generate and display recommendation tags based on the target text using a preset tagging model.

[0041] A judgment module is configured to determine whether the recommendation tag has been modified.

[0042] The quality inspection module is configured to perform quality inspection on the modified label if the recommended label is modified.

[0043] The annotation module is configured to use the modified tags as tags for the target text if the quality inspection passes.

[0044] By employing the above technical solution, the tagging model can generate and display recommended tags. Users can select recommended tags as tags for target text, avoiding the cold start problem of tag data. Furthermore, if a user modifies a recommended tag, the modified tag undergoes quality inspection. By using the modified tag as the tag for the target text, the target text can be annotated with the new, quality-inspected tag, thus avoiding the cold start problem of tag data caused by a lack of suitable tags.

[0045] The present invention also discloses an electronic device, comprising:

[0046] processor;

[0047] Memory, used to store programs.

[0048] When the program is executed by the processor, the processor implements the text content tag generation method described above.

[0049] The present invention also discloses a computer-readable medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the above-described text content tag generation method.

[0050] Solution 1. A method for generating text content tags, characterized in that the method includes:

[0051] Obtain the target text to be labeled;

[0052] Based on the target text, recommended tags are generated and displayed using a preset tagging model;

[0053] Determine whether the recommended tags have been modified;

[0054] If the recommended tags are modified, the modified tags will be subject to quality inspection.

[0055] If the quality inspection passes, the modified tag will be used as the tag of the target text, and the correspondence between the modified tag and the target text will be stored in the database.

[0056] Solution 2. The text content tag generation method according to Solution 1, characterized in that the step of "generating recommendation tags and displaying the recommendation tags based on the target text using a preset tagging model" further includes:

[0057] Based on the target text, select one or more first candidate tags from a preset tag library;

[0058] One or more of the first candidate tags are displayed to the user as the recommended tags.

[0059] Solution 3. The text content tag generation method according to Solution 2, characterized in that the step of "if the recommended tag is modified, then perform quality inspection on the modified tag" further includes:

[0060] Determine whether the modified tag belongs to the tag library;

[0061] If the modified tag belongs to the tag library, the modified tag is inspected using a preset quality inspection model.

[0062] Solution 4. The text content tag generation method according to Solution 3, characterized in that the step of "if the recommended tag is modified, then the modified tag is subjected to quality inspection" further includes:

[0063] If the modified tag does not belong to the tag library, then based on the modified tag, a second alternative tag is recommended through a preset machine recommendation model;

[0064] Determine whether the second alternative label is adopted;

[0065] Based on the assessment results, a second quality inspection will be conducted.

[0066] Solution 5. The text content tag generation method according to Solution 4, characterized in that the step of "performing quality inspection again based on the judgment result" further includes:

[0067] If the second candidate label is adopted, the second candidate label is subjected to quality inspection through the quality inspection model;

[0068] If the second alternative label is not adopted, the modified label will be subject to manual quality inspection.

[0069] Solution 6. The text content tag generation method according to Solution 5, characterized in that the step of "performing quality inspection on the second candidate tag through the quality inspection model" further includes:

[0070] The quality inspection model compares the target text with the second candidate label. If the target text and the second candidate label correspond, the quality inspection passes.

[0071] Solution 7. The text content tag generation method according to Solution 6, characterized in that the method further includes:

[0072] If the quality inspection passes, the second alternative label is added to the label library.

[0073] Solution 8. The text content tag generation method according to Solution 2, characterized in that, after the step of "determining whether the recommendation tag has been modified", it further includes:

[0074] If the recommended tag is not modified, then the recommended tag will be used as the tag for the target text.

[0075] Solution 9. The text content tag generation method according to Solution 2, characterized in that the step of "generating recommendation tags and displaying the recommendation tags based on the target text using a preset tagging model" further includes:

[0076] During the target text editing process, the first candidate label is selected synchronously in the backend using the labeling model;

[0077] Once the target text has been edited, the first alternative label is displayed.

[0078] Solution 10. A text content tag generation device, characterized in that it comprises:

[0079] The acquisition module is configured to acquire the target text to be tagged;

[0080] The recommendation module is configured to generate and display recommendation tags based on the target text using a preset tagging model.

[0081] A judgment module is configured to determine whether the recommendation tag has been modified.

[0082] The quality inspection module is configured to perform quality inspection on the modified label if the recommended label is modified.

[0083] The annotation module is configured to use the modified tags as tags for the target text if the quality inspection passes.

[0084] Solution 11. An electronic device, characterized in that it comprises:

[0085] processor;

[0086] Memory, used to store programs.

[0087] When the program is executed by the processor, the processor implements the text content tag generation method as described in any one of schemes 1-9.

[0088] Scheme 12. A computer-readable medium having a computer program stored thereon, characterized in that, when the computer program is executed by a processor, it implements the text content tag generation method as described in any one of Schemes 1-9. Attached Figure Description

[0089] The preferred embodiments of the present invention are described below with reference to the accompanying drawings, in which:

[0090] Figure 1 This is a flowchart of the main steps in the text content tag generation method;

[0091] Figure 2 This is a flowchart of the expansion step S104 in the text content tag generation method.

[0092] Figure 3 This is a flowchart of one possible implementation of a text content tag generation method;

[0093] Figure 4 This is a schematic diagram of the main modules of the label generation device.

[0094] List of reference numerals in the attached figures: 300-Label generation device; 301-Acquisition module; 302-Recommendation module; 303-Judgment module; 304-Quality inspection module; 305-Annotation module. Detailed Implementation

[0095] Preferred embodiments of this application are described below with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of this application and are not intended to limit the scope of protection of this application. Those skilled in the art can make adjustments as needed to adapt to specific application scenarios.

[0096] It should be noted that the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance. Furthermore, "multiple" in this application means at least two.

[0097] First refer to Figure 1This paper introduces the method for generating text content tags in this application.

[0098] like Figure 1 As shown, to address the cold start problem of existing tags and tag data, this invention provides a text content tag generation method, which includes:

[0099] S101, Obtain the target text to be tagged; for example, the user inputs text on a smart terminal device, and the text is obtained at the same time as the user inputs. Alternatively, the text can be obtained after the user has finished inputting. The smart terminal device can be a mobile phone, tablet computer, laptop computer, desktop computer, or other terminal device with input functionality.

[0100] S102, Based on the target text, recommended tags are generated and displayed using a pre-defined tagging model. In this application, the tagging model is used to represent the correspondence between the recommended tags and the target text. For example, the preferred pre-defined tagging model is a trained large language model. The target text is input into the tagging model, which generates recommended tags that match the target text and displays them on the user's smart terminal device. Of course, the pre-defined tagging model can also be a machine learning model. The trained large language model could be Chat-GPT, Wenxin Yiyan, or Pangu, etc.

[0101] S103, determine whether the recommended tags have been modified; for example, during the interaction, the user can modify the displayed preset tags or directly select recommended tags to use. After modification or selection, determine whether the recommended tags have been modified in order to facilitate subsequent operations.

[0102] S104, If the recommended tag is modified, perform quality checks on the modified tag; for example, if the recommended tag is modified, perform quality checks on whether the target text and the modified tag match, and output the quality check results.

[0103] S105, if the quality inspection passes, the modified tag is used as the tag for the target text, and the correspondence between the modified tag and the target text is stored in the database. For example, if the target text and the modified tag match after the quality inspection, the modified tag is used as the tag for the target text, and the correspondence between the modified tag and the target text is stored in the database, further ensuring that the target text can use the tag that best matches the target text.

[0104] In this way, the tagging model can generate and display recommended tags, which users can choose as tags for target text, avoiding the cold start problem of tag data. Furthermore, if a user modifies a recommended tag, the modified tag is quality checked. This quality check uses the modified tag as the tag for the target text. The target text can then be labeled with the new, quality-checked tag, and the correspondence between the modified tag and the target text is stored in the database. This facilitates labeling similar target text when it appears, while also avoiding the cold start problem of tag data caused by the lack of suitable tags.

[0105] In a preferred embodiment, the steps of "generating and displaying recommendation tags based on target text using a preset tagging model" include:

[0106] Based on the target text, one or more first candidate tags are selected from a preset tag library; for example, the target text entered by the user is "I made braised pork today, and it tasted amazing. It made me miss the food from my childhood"; the preset tag library includes tags such as: food, health, lifestyle, home, finance, sports, music, emotions, entertainment, film and television, fashion, and games; through the tagging model, "food" and "lifestyle" are selected from the tag library as the first candidate tags for the target text.

[0107] One or more primary candidate tags are displayed to the user as recommended tags. For example, after determining "food" and "lifestyle" as primary candidate tags, the "food" and "lifestyle" tags are displayed on the user's smart device and can be selected by the user.

[0108] After showing the recommended tags to the user, it determines whether the recommended tags are adopted. For example, the user can adopt one or two of the recommended tags, or the user can choose not to adopt the recommended tags and choose to create a new tag or modify the existing tag.

[0109] If a user selects one or more recommended tags, the target text is tagged. If the user does not adopt recommended tags and does not create new tags, the target text is not tagged. Thus, tagged target text can be recommended to people interested in the content associated with those tags, thereby increasing the target text's exposure. Untagged target text is randomly recommended to other users, reducing its exposure.

[0110] It should be noted that the above steps are preferred, and those skilled in the art can optimize them according to specific application scenarios. For example, the number of first candidate tags can be one, two, three, or more. Furthermore, a user can select one, two, three, or more first candidate tags as tags for the target text.

[0111] Furthermore, the step of "generating and displaying recommendation tags based on the target text using a pre-defined tagging model" also includes:

[0112] During the target text editing process, a tagging model synchronously selects the first candidate tag on the backend. For example, when the user pauses for more than 1 second while inputting text, the tagging model performs a tagging operation on the backend. Similarly, when the user inputs text, the tagging model performs a tagging operation on the backend after each punctuation mark or Enter key.

[0113] After the target text is edited, a tagging model generates recommended tags and displays the first candidate tags. For example, after the user finishes editing the text and clicks the "Done" button, the tagging model generates recommended tags that match the target text and displays them. Similarly, after the user finishes editing the text and clicks the "Tags" icon, the tagging model generates recommended tags that match the target text and displays them.

[0114] like Figure 2 As shown, in a preferred embodiment, the step of "if the recommendation label is modified, then perform quality inspection on the modified label" further includes:

[0115] S1041, determine whether the modified tag belongs to the tag library; for example, if the user modifies the preset tag by changing "food" to "sports", and the "sports" tag is in the tag library, then the tag belongs to the tag library. As another example, if the user modifies the preset tag by changing "food" to "psychology", and the "psychology" tag is not in the tag library, then the tag does not belong to the tag library.

[0116] If the modified tag belongs to the tag library, the modified tag is quality checked using a pre-defined quality inspection model. In this application, the quality inspection model is used to characterize the matching relationship between the target text and the tag. For example, the preferred quality inspection model is a trained large language model. If "food" is changed to "sports", the tag "sports" and the target text "Today I cooked a braised pork belly myself, and it tasted amazing. I miss the food from my childhood" are quality checked. Specifically, the target text and the modified tag are input into the quality inspection model, which determines whether the modified tag matches the target text. If they match, the quality inspection passes; otherwise, it fails.

[0117] S1042, if the modified tag does not belong to the tag library, then based on the modified tag, a second alternative tag is recommended through a preset machine recommendation model. In this application, the machine recommendation model is used to represent the correspondence between the modified tag and the second alternative tag. For example, the preferred machine recommendation model is a trained large language model. If "delicious food" is changed to "braised pork," and the modified tag "braised pork" does not belong to the tag library, then the machine recommendation model recommends a second alternative tag based on the modified tag, such as "culinary skills."

[0118] It should be noted that the above is a preferred implementation of this step, and those skilled in the art can adjust the steps according to specific application scenarios. For example, the step could also be: if the modified tag does not conform to the tag specifications, then a second alternative tag is recommended by a preset machine recommendation model before quality inspection. Tag specifications can be two characters or one word, etc. Alternatively, the step could be: if the modified tag does not belong to the tag library and does not conform to the tag specifications, then a second alternative tag is recommended by a preset machine recommendation model before quality inspection; if the modified tag does not belong to the tag library but conforms to the tag specifications, then quality inspection is performed directly. Furthermore, modifying tags is not limited to changing tags that appear based on user-selected recommended tags, but also includes tags created by the user.

[0119] It should be further explained that the step of "if the recommendation label is modified, then perform quality inspection on the modified label" can also be performed manually instead of using a quality inspection model to inspect the modified label.

[0120] S1043, determine whether the second alternative label is adopted; for example, obtain whether the user selects the second alternative label on the smart terminal device. If the user selects the second alternative label, the second alternative label is adopted; if the user does not select the second alternative label and re-edits the label, return to S1041.

[0121] S1044, Based on the judgment result, a quality inspection is carried out again.

[0122] Specifically, if the second candidate tag is adopted, a quality control model is used to check it. Specifically, the quality control model compares the target text with the second candidate tag. If the target text and the second candidate tag match, the quality control passes. For example, if the second candidate tag "cooking skills" is adopted, the quality control model determines whether the second candidate tag "cooking skills" matches the target text. If the target text and the second candidate tag match, the quality control passes.

[0123] If the second alternative tag is not adopted, the modified tag will undergo manual quality inspection. For example, if the user does not adopt "cooking skills" as a tag and insists on choosing "braised pork" as the tag for the target text, the modified tag "braised pork" will undergo manual quality inspection.

[0124] Furthermore, if the quality inspection passes, the second candidate tag is added to the tag library. The correspondence between the modified tag and the target text is also stored in the database. For example, if the second candidate tag "cooking skills" is adopted, and the quality inspection model determines that the second candidate tag "cooking skills" matches the target text, then the second candidate tag "cooking skills" is added to the tag library. Simultaneously, the correspondence between the modified tag and the target text is stored in the database. As another example, if the modified tag "braised pork" is adopted, and the quality inspection model determines that the modified tag "braised pork" matches the target text, then the modified tag "braised pork" is added to the tag library. In this way, second candidate tags that pass the quality inspection can be added to the tag library, allowing the tags in the tag library to cover more content and enabling more accurate annotation of text content.

[0125] It should be noted that the step of "if the recommended label is modified, then perform quality inspection on the modified label" can also be performed manually instead of using a quality inspection model to inspect the modified label.

[0126] It should be further noted that the modified label, the second alternative label, and the newly created label in this embodiment all belong to the modified label in claim 1.

[0127] The following is combined Figure 3 This paper describes one possible implementation process of this application.

[0128] like Figure 3 As shown, the process of a text content tag generation method according to an embodiment of the present invention includes the following steps:

[0129] S201, Obtain the target text to be labeled;

[0130] S202, the labeling model selects two first candidate labels from the preset label library;

[0131] S203, display the two first alternative tags as recommended tags to the user;

[0132] S204, determine whether the recommended tag has been adopted; if yes, proceed to S205; otherwise, proceed to S206.

[0133] S205, determine whether the recommendation tag has been modified; if yes, execute S206; otherwise, execute S213.

[0134] S206, determine whether the modified tag belongs to the tag library; if yes, execute S209; otherwise, execute S207.

[0135] S207, the machine recommendation model recommends the second alternative label;

[0136] S208, determine whether the second alternative label is adopted; if yes, execute S209; otherwise, execute S210.

[0137] S209, the label is inspected using a quality inspection model;

[0138] S210, Manual quality inspection of altered labels;

[0139] S211, determine whether the quality inspection has passed; if it has passed, proceed to 212; otherwise, proceed to 214.

[0140] S212, Add the tag to the tag library;

[0141] S213, use this tag as the tag for the target text;

[0142] S214, The target text does not use tags.

[0143] It should be noted that although the detailed steps of the method of this application have been described above, those skilled in the art can combine, split, and rearrange the above steps without departing from the basic principles of this application. Such modified technical solutions do not change the basic concept of this application and therefore fall within the protection scope of this application. For example, the order of S212 and S213 can be interchanged. Furthermore, S205 can be omitted. If the recommended label in S204 is adopted, then S213 is executed; if the recommended label is not adopted in S204, but rather modified or newly created, then S206 is executed.

[0144] Based on the same inventive concept, such as Figure 4 As shown, the present invention also discloses a text content tag generation device 300, comprising:

[0145] The acquisition module 301 is configured to acquire the target text to be labeled;

[0146] The recommendation module 302 is configured to generate and display recommendation tags based on the target text using a preset tagging model.

[0147] The judgment module 303 is configured to determine whether the recommendation tag has been modified.

[0148] The quality inspection module 304 is configured to perform quality inspection on the modified label if the recommendation label is modified.

[0149] The annotation module 305 is configured to use the modified label as the label of the target text if the quality inspection passes.

[0150] By employing the above technical solution, the tagging model can generate and display recommended tags. Users can select recommended tags as tags for target text, avoiding the cold start problem of tag data. Furthermore, if a user modifies a recommended tag, the modified tag undergoes quality inspection. By using the modified tag as the tag for the target text, the target text can be annotated with the new, quality-inspected tag, thus avoiding the cold start problem of tag data caused by a lack of suitable tags.

[0151] It should be noted that the text content tag generation device provided in the above embodiments is only illustrated by the division of the above functional modules (such as acquisition module 301, recommendation module 302, judgment module 303, quality inspection module 304, annotation module 305, etc.). In practical applications, the above functional modules can be completed by different functional units as needed, that is, the functional modules in this embodiment can be further decomposed or combined. For example, the functional modules in the above embodiments can be merged into one functional module, or further divided into multiple sub-modules to complete all or part of the functions described above. The names of the functional modules involved in this embodiment are only for differentiation and are not considered as an improper limitation of this application.

[0152] Those skilled in the art will understand that all or part of the processes in the method of the above embodiment of the present invention can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable file, or some intermediate form. The computer-readable storage medium can include any entity or device capable of carrying computer program code, media, USB flash drive, portable hard drive, magnetic disk, optical disk, computer memory, read-only memory, random access memory, electrical carrier signals, telecommunication signals, and software distribution media, etc. It should be noted that the content included in the computer-readable storage medium can be appropriately added or removed according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer-readable storage medium does not include electrical carrier signals and telecommunication signals.

[0153] The various component embodiments of the present invention can be implemented in hardware, or as software modules running on one or more processors, or a combination thereof. Those skilled in the art will understand that microprocessors or digital signal processors (DSPs) can be used in practice to implement some or all of the functions of some or all of the components in the server or client according to embodiments of the present invention. The present invention can also be implemented as a device or apparatus program (e.g., a PC program and PC program products) for performing part or all of the methods described herein. Such programs implementing the present invention can be stored on a PC-readable medium, or can be in the form of one or more signals. Such signals can be downloaded from an Internet website, provided on a carrier signal, or provided in any other form.

[0154] Based on the same inventive concept, this invention also discloses an electronic device, including: a processor; and a memory for storing a program, which, when executed by the processor, causes the processor to implement the above-described text content tag generation method. For ease of explanation, only the parts related to the embodiments of this invention are shown; for specific technical details not disclosed, please refer to the method section of the embodiments of this invention. This electronic device can be any of various devices capable of implementing the above-described method.

[0155] Based on the same inventive concept, this invention also discloses a computer-readable medium storing a computer program that, when executed by a processor, implements the aforementioned method. For ease of explanation, only the parts relevant to the embodiments of this invention are shown; for specific technical details not disclosed, please refer to the method section of the embodiments of this invention. This computer-readable storage medium can be a storage device comprising various electronic devices. Optionally, in the embodiments of this invention, the computer-readable storage medium is a non-transitory computer-readable storage medium.

[0156] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after such changes or substitutions will all fall within the scope of protection of the present invention.

Claims

1. A method for generating a text content label, characterized by, The method comprises: acquiring target text to be tagged; generating a recommended label based on the target text through a preset tagging model and displaying the recommended label; determining whether the recommended label is modified; if the recommended label is modified, performing quality inspection on the modified label; if the quality inspection passes, taking the modified label as the label of the target text and storing the corresponding relationship between the modified label and the target text in a database.

2. The text content label generation method according to claim 1, characterized by, The step of "generating a recommended label based on the target text through a preset tagging model and displaying the recommended label" further comprises: selecting one or more first alternative labels from a preset label library based on the target text; displaying the one or more first alternative labels as the recommended label to a user.

3. The text content label generation method according to claim 2, characterized by, The step of "if the recommended label is modified, performing quality inspection on the modified label" further comprises: determining whether the modified label belongs to the label library; if the modified label belongs to the label library, performing quality inspection on the modified label through a preset quality inspection model.

4. The text content label generation method according to claim 3, characterized by, The step of "if the recommended label is modified, performing quality inspection on the modified label" further comprises: if the modified label does not belong to the label library, recommending a second alternative label based on the modified label through a preset machine recommendation model; determining whether the second alternative label is adopted; based on the determination result, performing quality inspection again.

5. The text content tag generation method according to claim 4, characterized by, The step of "based on the determination result, performing quality inspection again" further comprises: if the second alternative label is adopted, performing quality inspection on the second alternative label through the quality inspection model; if the second alternative label is not adopted, performing artificial quality inspection on the modified label.

6. The text content tag generation method according to claim 5, characterized by, The step of "performing quality inspection on the second alternative label through the quality inspection model" further comprises: comparing the target text and the second alternative label through the quality inspection model, and if the target text and the second alternative label correspond, the quality inspection passes.

7. The text content tag generation method according to claim 6, characterized by, The method further comprises: if the quality inspection passes, adding the second alternative label to the label library.

8. The text content tag generation method according to claim 2, characterized by, The step of "determining whether the recommended label is modified" further comprises: if the recommended label is not modified, taking the recommended label as the label of the target text.

9. The text content tag generation method according to claim 2, characterized by, The step of "generating a recommended label based on the target text through a preset tagging model and displaying the recommended label" further comprises: during editing of the target text, synchronously selecting a first alternative label through the tagging model in the backend; after editing of the target text is completed, displaying the first alternative label.

10. A text content label generation apparatus characterized by comprising: comprises: an acquisition module configured to acquire target text to be tagged; a recommendation module configured to generate a recommended label based on the target text through a preset tagging model and display the recommended label; a determination module configured to determine whether the recommended label is modified; a quality inspection module configured to perform quality inspection on a modified label if the recommended label is modified; a labeling module configured to take the modified label as the label of the target text if the quality inspection passes.

11. An electronic device, comprising: comprises: a processor; a memory for storing a program, when the program is executed by the processor, the processor implements the text content label generation method as claimed in any one of claims 1-9.

12. A computer readable medium having stored thereon a computer program, characterized in that, the computer program is executed by the processor to implement the text content label generation method as claimed in any one of claims 1-9.

Images