Tag generation method and apparatus for content pushing, electronic device, and medium

Abstract

Embodiments of the present invention relate to a tag generation method for content pushing. The method comprises acquiring target content to be marked, the target content being marked with a first tag. The method further comprises sending, to a tagging queue corresponding to the first tag in a tagging platform, information indicating the target content, the tagging platform being used for generating a second tag for the target content, and the classification granularity of the second tag being finer than that of the first tag. The method further comprises acquiring from the tagging platform the second tag corresponding to the target content. In addition, the method further comprises associatively storing the second tag and the target content as marked target content.

Description

Tag generation method, device, electronic device and medium for content push

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims priority to the Chinese patent application filed with the State Intellectual Property Office of China on July 21, 2023, with application number 202310907823.1 and invention name “Label generation method, device, electronic device and medium for content push”, the entire contents of which are incorporated by reference into this application. Technical Field

[0003] The present disclosure generally relates to the field of big data technology, and more specifically, to a tag generation method, apparatus, electronic device, and medium for content push. Background Art

[0004] With the growth of content on Internet platforms and the increasing demand for personalized content among users, it is becoming increasingly important to provide personalized content to different users and match content with users.

[0005] Currently, user tagging is widely used in consumer-oriented (ToC) scenarios. By tagging users and content, users can be provided with content corresponding to the tag, enabling personalized content push. Content platforms can also categorize and display content based on tags to facilitate users' access to content in specific categories. In some cases, content generated in ToC scenarios is also provided to business-oriented (ToB) users.

[0006] Summary of the Invention

[0007] Embodiments of the present disclosure provide a tag generation method, apparatus, electronic device, and medium for content push.

[0008] According to a first aspect of the present disclosure, a tag generation method for content push is provided. The method includes obtaining target content to be annotated, and the target content is marked with a first tag. The method also includes sending information indicating the target content to a tagging queue corresponding to the first tag in a tagging platform, and the tagging platform is used to generate a second tag for the target content, and the second tag has a finer classification granularity than the first tag. The method also includes obtaining a second tag corresponding to the target content from the tagging platform. In addition, the method also includes storing the second tag in association with the target content as an annotated target content, and the annotated target content is the content in the target content to be annotated that is marked with the second tag.

[0009] In a second aspect of the present disclosure, a label generation device for content push is provided. The device includes a first acquisition module, which is configured to acquire target content to be labeled, and the target content is labeled with a first label. The device also includes a sending module, which is configured to send information indicating the target content to a labeling queue corresponding to the first label in a labeling platform, and the labeling platform is used to generate a second label for the target content, and the second label has a finer classification granularity than the first label. The device also includes a second acquisition module, which is configured to acquire a second label corresponding to the target content from the labeling platform. In addition, the device also includes a storage module, which is configured to store the second label in association with the target content as the labeled target content, and the labeled target content is the content in the target content to be labeled that is labeled with the second label.

[0010] In a third aspect of the present disclosure, an electronic device is provided, comprising a processor and a memory coupled to the processor, wherein the memory has instructions stored therein, and when the instructions are executed by the processor, the electronic device executes the method according to the first aspect.

[0011] In a fourth aspect of the present disclosure, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer-executable instructions, wherein the computer-executable instructions are executed by a processor to implement the method according to the first aspect.

[0012] This summary is intended to introduce a selection of concepts in a simplified form that are further described below in the detailed description. It is not intended to identify key features or essential features of the claimed subject matter, nor is it intended to limit the scope of the claimed subject matter. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] The above and other features, advantages and aspects of the embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. In the accompanying drawings, the same or similar reference numerals represent the same or similar elements, wherein:

[0014] FIG1A is a schematic diagram showing some exemplary application scenarios of an embodiment of the present disclosure;

[0015] FIG1B shows a schematic diagram of tags of content involved in an embodiment of the present disclosure;

[0016] FIG1C shows another schematic diagram of content labels involved in an embodiment of the present disclosure;

[0017] FIG2 illustrates a schematic diagram of an example environment in which various embodiments of the present disclosure may be implemented;

[0018] FIG3 shows a schematic flow chart of a tag generation method according to certain embodiments of the present disclosure;

[0019] FIG4 is a schematic diagram showing a process of determining tags for content according to certain embodiments of the present disclosure;

[0020] FIG5 shows a schematic diagram of a method for sending target content to a tagging platform according to certain embodiments of the present disclosure;

[0021] FIG6 is a schematic diagram illustrating a method for determining a second tag of target content and applying and analyzing the second tag according to certain embodiments of the present disclosure;

[0022] FIG7 is a schematic diagram showing a method of generating a second tag and analyzing the second tag according to certain embodiments of the present disclosure;

[0023] FIG8 shows a block diagram of a label generating apparatus according to certain embodiments of the present disclosure; and

[0024] FIG9 illustrates a block diagram of an electronic device according to certain embodiments of the present disclosure.

[0025] Throughout the drawings, the same or similar reference numbers denote the same or similar elements. DETAILED DESCRIPTION

[0026] It is understandable that the data involved in this technical solution (including but not limited to the data itself, the acquisition or use of the data) must comply with the requirements of relevant laws, regulations and relevant provisions.

[0027] The following describes embodiments of the present disclosure in more detail with reference to the accompanying drawings. Although certain embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as limited to the embodiments described herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are for illustrative purposes only and are not intended to limit the scope of protection of the present disclosure.

[0028] In the description of the embodiments of the present disclosure, the term "including" and similar terms should be understood as open inclusion, that is, "including but not limited to". The term "based on" should be understood as "based at least in part on". The term "one embodiment" or "the embodiment" should be understood as "at least one embodiment". The terms "first", "second", etc. can refer to different or the same objects, unless explicitly stated otherwise. Other explicit and implicit definitions may also be included below.

[0029] When generating or acquiring content from users, content platforms that provide users with audio, video, images, articles, and / or web pages can tag the content. This allows them to push the content to other users based on the tags, enabling personalized content delivery. Content platforms can also categorize and display content based on tags to facilitate user access to specific categories of content. Different types of users have different requirements for the granularity of content tag classification. For example, individual users may not require high granularity when acquiring content, while enterprise users may require higher granularity than individual users.

[0030] However, content on current content platforms typically has only one tag, corresponding to a specific application scenario, such as an individual user's application scenario. When content is provided to users with more refined classification requirements (such as enterprise users), the original tags of the content will not meet the user's demand for refined content. For example, users cannot directly obtain the required content with a finer classification granularity based on the original tags of the content, and users need to conduct further screening, which increases the cost of obtaining content.

[0031] To this end, an embodiment of the present disclosure provides a label generation method. In an embodiment of the present disclosure, a target content to be labeled is obtained, and the target content is labeled with a first label; information indicating the target content is sent to a labeling queue corresponding to the first label in a labeling platform, and the labeling platform is used to generate a second label for the target content, and the second label has a finer classification granularity than the first label; the second label corresponding to the target content is obtained from the labeling platform; and the second label is associated with the target content and stored as the labeled target content, and the labeled target content is the content in the target content to be labeled that is labeled with the second label. Through the embodiment of the present disclosure, a new label with a finer classification granularity can be determined for the labeled content, so that the content label can meet the needs of different application scenarios, thereby expanding the application scenarios of the content label. In addition, when determining the new label for the content, by sending the content to the labeling queue corresponding to the first label in the labeling platform, the content obtained by the labeling platform has been classified based on the first label, thereby improving the efficiency of generating new labels for the content. Each labeling queue will be processed by a labeler who is more familiar with the corresponding field of the queue or a labeling model that is more compatible with the queue, and the accuracy of the generated label will also be higher.

[0032] First, the application scenarios involved in the embodiments of the present disclosure are described with reference to Figures 1A, 1B, and 1C. Figure 1A is a schematic diagram of some exemplary application scenarios 100A of the embodiments of the present disclosure. As shown in Figure 1A, the application scenario 100A may include a labeling platform 120. The labeling platform 120 may be a user terminal, a mobile device, a computer, etc. It may also be a computing system, a single server, a distributed server, or a cloud-based server. The labeling platform 120 may also be implemented in software and / or hardware. The labeling platform 120 may include a labeling queue 121, a labeling queue 122, and a labeling queue 123. The labeling platform may determine fine-grained labels 130 for content 110 with coarse-grained labels through the labeling queues, thereby obtaining content 140 with fine-grained labels. For example, the labeling platform may provide the content in the labeling queue to the labeling personnel corresponding to the labeling queue and obtain the fine-grained labels 130 input by the labeling personnel. In some embodiments, each labeling queue of the labeling platform 120 corresponds to a label determination model, which can determine fine-grained labels 130 for the content in the labeling queue. It should be understood that the annotation platform 120 in FIG. 1A is only an example of an annotation platform in an embodiment of the present disclosure. The present disclosure does not limit the specific form of the annotation platform, nor does it limit the number of annotation queues in the annotation platform.

[0033] In the embodiments of the present disclosure, content may include, but is not limited to, video content, audio content, text content, image content, and graphic content. Content may be processed or transmitted in the form of data packets, and each content may be uniquely associated with an identifier (ID). A content tag may, for example, be a field indicating the category to which the content belongs, and such a field may be included in a data packet of the content.

[0034] In embodiments of the present disclosure, tags can be used to provide users with content of corresponding categories. For example, content can be pushed to users based on tags, or tags can be displayed to users through a menu bar to enable users to categorize and select content based on tags. As an example, tags are described below with reference to Figures 1B and 1C. In Figures 1B and 1C, tags indicating the category to which content belongs can be displayed through a menu bar on the user interface, allowing users to obtain desired content based on tags.

[0035] FIG1B is a schematic diagram of labels of content involved in an embodiment of the present disclosure. For example, the user interface shown in FIG1B may include a menu bar 150, which may display labels indicating general categories of content, such as "Science and Technology," "Film and Television Entertainment," "Health and Medical Care," "Education and Training," and "Fashion Trends." In some embodiments, labels may also indicate secondary categories of content. For example, FIG1B may further include a menu bar 160, which may display labels corresponding to secondary categories such as "Movies," "TV Series," "Animation," and "Variety Shows."

[0036] FIG1C is another schematic diagram of labels for content involved in an embodiment of the present disclosure. For example, the user interface shown in FIG1C may include a menu bar 170, which may display labels indicating the industry classification of the content, such as "Movie Library," "New Movie Trailers," "Movie Reviews," "Behind-the-Scenes Interviews," and "Box Office Trends." It should be understood that the labels in FIG1B and FIG1C are provided as examples for illustrative purposes only, and the present disclosure does not limit the specific content indicated by the labels. In some embodiments, the labels may also indicate the source of the content.

[0037] The labels in the menu bar 170 in FIG1C have a finer classification granularity than the labels in FIG1B . In the application scenario 100A shown in FIG1A , the coarse-grained labels of the content 110 may be, for example, labels as shown in FIG1B , and the fine-grained labels 130 may be, for example, labels as shown in the menu bar 170. Through the method provided by the embodiment of the present disclosure, labels with finer classification granularity as shown in the menu bar 170 can be determined for the content with the coarse-grained labels as shown in FIG1B . Based on the labels with finer classification granularity, finer-grained content can be displayed to the user, and more personalized content recommendations can be achieved. For example, if the user is more interested in movie reviews, the user can select a label "movie reviews" that is finer-grained than the label "movie" to quickly obtain the required content, thereby improving the user experience.

[0038] In some embodiments, each annotation queue in the annotation platform 120 can determine fine-grained labels for a category of content. For example, the annotation queue 121 can determine labels with finer classification granularity for content with the label "film and television entertainment", and the annotation queue 122 can determine labels with finer classification granularity for content with the label "health care", and so on.

[0039] FIG2 shows a schematic diagram of an example environment 200 in which multiple embodiments of the present disclosure can be implemented. As shown in FIG2 , the example environment 200 may include a content processing device 210 and an annotation platform 220. The content processing device 210 may be a computing system, a single server, a distributed server, or a cloud-based server. It may also be a user terminal, a mobile device, a computer, etc. The content processing device 210 may be implemented in software and / or hardware. The annotation platform may include multiple annotation queues, for example, an annotation queue 221 and an annotation queue 222, which may generate tags for content. Exemplarily, the annotation platform 220 may correspond to the annotation platform 120 in FIG1A .

[0040] Referring to Figure 2, in box 211, the content to be annotated is obtained, thereby obtaining the content to be annotated that is marked with a label. The content processing device 210 can obtain the content to be annotated that is marked with a label from a content platform or database. The content can be, for example, the content 110 shown in Figure 1A. The label marked with the content can be, for example, the label shown in Figure 1B. For ease of explanation, this label is referred to as the first label. In box 212, the content processing device 210 can process the obtained content to be annotated. For example, the content processing device 210 can obtain information such as the historical number of views, author, popularity, creation time, etc. of the content to be annotated, and can score the content to be annotated based on this information. In some embodiments, the content processing device 210 can sort the content to be annotated based on the score of the content to be annotated. In box 213, the content processing device 210 sends information indicating the content to be annotated to the annotation queue in the annotation platform 220. For example, the ID corresponding to the content to be annotated can be sent to the annotation platform. The annotation queue to which the information indicating the content to be annotated is sent can be related to the first label of the content to be annotated. In block 214, the content processing device 210 obtains the annotated content from the annotation platform 220, for example, obtaining a new label corresponding to the content to be annotated sent to the annotation platform 220. The new label can be, for example, the fine-grained label 130 in FIG. 1A or the label in the menu bar 170 shown in FIG. 1C. For ease of explanation, the new label is referred to as a second label. The second label has a finer classification granularity than the first label.

[0041] In an embodiment of the present disclosure, the number of second tags obtained by the content processing device 210 from the annotation platform 220 may be less than the number of contents to be annotated indicated by the information sent by the content processing device 210 to the annotation platform 220. That is, for a portion of the content, the annotation platform 220 will not generate a second tag for it. For example, for a piece of information in the annotation queue 222, the annotation platform can obtain the content to be annotated corresponding to the information from the content database based on the information and present it to the annotator corresponding to the annotation queue 222. If the annotator determines that the content corresponds to a second tag, the annotator can mark it with the second tag, so that the annotation platform 220 can generate a second tag corresponding to the annotated content and send it to the content processing device 210; if the annotator determines that the annotated content does not correspond to the second tag in the annotation queue 222 and does not mark it with the second tag, the annotation platform 220 cannot generate a second tag corresponding to the annotated content, and the content processing device 210 will not be able to obtain the second tag of the content.

[0042] In the example environment 200, the content processing device 210 sends information indicating the content to be annotated to the annotation platform, so that the annotation platform determines a label with finer classification granularity for the tagged content. It should be understood that the architecture and functions in the example environment 200 are described only for exemplary purposes and do not imply any limitation on the scope of the present disclosure. The embodiments of the present disclosure can also be applied to other environments with different structures and / or functions. For example, in some embodiments, the content processing device 210 and the annotation platform 220 can be implemented as a whole.

[0043] The following describes in detail the methods provided by the embodiments of the present disclosure in conjunction with Figures 3 to 7. For ease of understanding, the specific data and labels mentioned in the following description are exemplary and are not intended to limit the scope of protection of the present disclosure. It is understood that the embodiments described below may also include additional actions not shown and / or may omit the actions shown, and the scope of the present disclosure is not limited in this respect.

[0044] FIG3 illustrates a schematic flow chart of a tag generation method 300 according to certain embodiments of the present disclosure. Method 300 may be executed by a content processing device, which may be a computing system, a single server, a distributed server, or a cloud-based server. It may also be a user terminal, a mobile device, a computer, or the like. It may be implemented in software and / or hardware. For example, the content processing device may be the aforementioned content processing device 210. Referring to FIG3 , illustratively, method 300 may include blocks 302 to 308, which are described in detail below.

[0045] In box 302, the content processing device obtains the target content to be marked, and the target content is marked with a first label. The content processing device can, for example, obtain the content input by the user in real time from the content platform as the target content to be marked, or obtain the target content to be marked from a database storing the target content. In some embodiments, the target content can be one of audio, video, text, image or graphic content. In some embodiments, the target content can include multiple audio, video, text, image or graphic content. The first label of the target content can be created by the creator of the target content, or it can be determined by the content platform through a clustering algorithm, and the present disclosure does not limit this. In some embodiments, the data packet of the target content obtained by the content processing device can include a field corresponding to the first label, and the field can be included in the metadata of the target content, for example.

[0046] In some embodiments, the content processing device can also obtain the ID of the target content. In some embodiments, the content processing device can also obtain the author information of the target content. In some embodiments, the content processing device can also obtain the release time information of the target content. In some embodiments, the content processing device can also obtain the popularity value of the target content, which can be, for example, the number of views of the target content within a predefined time period, the number of likes of the target content within a predefined time period, and the number of comments on the target content within a predefined time period. It should be understood that the present disclosure does not specifically limit how to determine the popularity value of the target content. It should also be understood that the target content's ID, author information, release time information, and popularity value can be carried in the target content's data packet, thereby being obtained by the content processing device together with the target content, or can be obtained separately by the content processing device. The present disclosure does not limit the manner and order in which the above information is obtained.

[0047] In block 304, the content processing device sends information indicating the target content to a tagging queue corresponding to the first tag in the tagging platform. The tagging platform is used to generate a second tag for the target content, which has a finer classification granularity than the first tag. The information indicating the target content may be, for example, the ID of the target content or the name of the target content. The content processing device may obtain the information indicating the target content at the same time as obtaining the target content. After receiving the information indicating the target content, the tagging platform may retrieve the target content from a database storing the target content based on the information, and then generate a second tag for the target content.

[0048] In some embodiments, the content processing device may parse a data packet of the target content, obtain information indicating the target content and a field indicating the first tag from the data packet of the target content, and further determine the first tag of the target content.

[0049] In some embodiments, the annotation platform may include multiple annotation queues, each of which may be used to generate a second label for a type of content. The content processing device may determine the annotation queue corresponding to the target content based on the first label of the target content and send information indicating the target content to the annotation queue. Optionally, the annotation platform may be the annotation platform 120 in FIG. 1A or the annotation platform 220 in FIG. 2 .

[0050] In some embodiments, each annotation queue may correspond to a queue tag, and the content processing device may obtain the queue tag corresponding to the annotation queue, and based on the matching of the first tag and the queue tag, send the information indicating the target content to the annotation queue corresponding to the first tag. For example, in the aforementioned FIG1A, the annotation queue 121 in the annotation platform 120 may correspond to the queue tag "movie", and may generate a second tag related to the movie for the target content, the annotation queue 122 may correspond to the queue tag "series", and may generate a second tag related to the series for the target content, and the annotation queue 123 may correspond to the queue tag "animation", and may generate a second tag related to the animation for the target content. If the first tag of a target content is the same as the queue tag of a annotation queue, the content processing device may send the information indicating the target content to the annotation platform.

[0051] A tagging queue in the tagging platform is used to generate a second label for a type of target content. If the target content indicated by the information sent to the tagging queue does not belong to the category corresponding to the tagging queue, the tagging platform cannot generate a second label for the target content. The tagging platform will return the information indicating the target content to the content processing device or re-determine the tagging queue for the information indicating the target content. Therefore, sending the information indicating the target content to the tagging queue corresponding to the first label of the target content can make the category of the target content more relevant to the category of the tagging queue, thereby improving the efficiency of the tagging platform in generating the second label for the target content.

[0052] In some embodiments, the annotation platform includes an annotation queue that is also related to the popularity level of the target content. A content processing device can obtain the popularity value of the target content and, based on a predefined correspondence between the popularity value and the popularity level, determine the popularity level of the target content. The annotation platform can process the target content separately based on the popularity level of the target content. For example, for annotation queues with higher popularity levels, the annotation platform can allocate more resources to generate second tags for the target content in that queue. Optionally, the annotation platform's annotation queues can include a hotspot annotation queue and a normal annotation queue. The target content's popularity levels can include hotspot and normal. The content processing device can identify target content with a popularity value greater than or equal to a predetermined popularity threshold as hotspot target content, and target content with a popularity value less than the predetermined popularity threshold as normal target content. The content processing device can send information indicating hotspot target content to the hotspot annotation queue and information indicating normal target content to the normal annotation queue. Optionally, the target content can also include three popularity levels, and the annotation platform's annotation queues can be divided into three annotation queues corresponding to the target content's popularity levels. It should be understood that the present disclosure does not specifically limit the popularity level of the target content.

[0053] In some embodiments, the content processing device can determine the score of the target content based on at least one of the first tag, author, popularity, and release time of the target content, and determine the order of the target content based on the score, and send the information indicating the target content to the annotation queue of the annotation platform according to the order. For example, the content processing device can order the target content with high scores first and the content with low scores last, and send the information indicating the target content to the annotation queue based on the order of the target content. In some embodiments, the score of the target content determined by the content processing device can be related to the release time of the target content. The closer the release time, the higher the score of the target content. In this way, it can ensure that the latest content can be determined by the second tag in a timely manner, and then the content can be provided to the user device in a timely manner based on the second tag, so that the user can obtain the latest content when obtaining content based on the second tag, thereby improving the user experience.

[0054] Optionally, for different first tags, the scores of the corresponding target content are determined in different ways. For example, for target content with the first tag "Film and Television Entertainment", its score may be related to the author of the target content, while for target content with the first tag "Fashion Trends", its score may be related to the release time of the target content.

[0055] In some embodiments, the number of information indicating target content sent to the annotation queue per unit time is less than or equal to a predefined number threshold, for example, the number of information indicating target content sent to the annotation queue per day is less than or equal to 100. In this way, limiting the number of target content that the annotation queue needs to process can avoid the accumulation of tasks on the annotation platform and reduce the burden on the annotation platform. Optionally, the predefined number threshold can be related to the annotation queue and the current time. For example, between 7-19 o'clock every day, the number of information indicating target content sent to the ordinary annotation queue per hour is less than or equal to 300, and between 8-10 o'clock every day, the number of information indicating target content sent to the hotspot annotation queue per hour is less than or equal to 200.

[0056] In some embodiments, the content processing device may filter the acquired target content to be annotated based on predefined filtering rules. As one example, the content processing device may be configured with a block word library, which includes predefined block words. In the case where the acquired target content to be annotated contains block words in the block word library, the content processing device may filter the target content. As another example, the content processing device may be configured with a predefined first tag set, which includes multiple different first tags. In the case where the first tag of the acquired target content to be annotated is different from all the tags in the first tag set of the content processing device, the content processing device may filter the target content. Optionally, the target content to be annotated may be filtered multiple times until the number of remaining target content is less than or equal to a predefined number threshold.

[0057] In some embodiments, the content processing device can process the target content, generate sub-content related to the category of the target content, and send the sub-content to the annotation platform. In one example, if the target content is video content, the content processing device can capture multiple video frames from the video content and send the video frames to the annotation platform. In another example, if the target content is text content, the content processing device can extract a summary of the text and send the summary to the annotation platform. In this way, after receiving information indicating the target content, the annotation platform does not need to obtain the target content from the database, but can directly generate a second tag for the target content based on the sub-content, thereby improving the efficiency of the annotation platform.

[0058] In box 306, the content processing device obtains a second label corresponding to the target content from the annotation platform. The annotation platform can generate a second label based on the target content, and the classification granularity of the second label is finer than that of the first label. The annotation platform can send the second label together with the information indicating the target content to the content processing device. In an embodiment of the present disclosure, if the target content corresponding to the information in the annotation queue is marked with a second label by the annotation personnel or the preset annotation model, it means that the annotation platform has generated a second label related to the target content; if the target content corresponding to the information in the annotation queue is not marked with a second label after being processed by the annotation personnel or the preset annotation model, it means that the annotation platform cannot generate a second label related to the target content.

[0059] In some embodiments, the content processing device may also obtain information indicating the target content and information indicating that the second tag cannot be generated for the target content from the tagging platform. Optionally, the content processing device may re-determine the tagging queue corresponding to the target content based on this information. Optionally, the content processing device may remove the target content from the obtained target content to be tagged.

[0060] In block 308, the content processing device associates the second tag with the target content and stores it as the annotated target content. The annotated target content is the content in the target content to be annotated that is marked with the second tag. After obtaining the second tag corresponding to the target content from the annotation platform, the content processing device can tag the target content with the second tag to generate the annotated target content, for example, by adding a field to the data packet of the target content to indicate the second tag of the target content. The content processing device can store the annotated target content in a database or memory, which can be a local memory of the content processing device or an independent database or memory.

[0061] In some embodiments, a content processing device can provide a content platform with target content tagged with a second tag. For example, the content platform can be a server that provides content to user devices based on the second tag. The server can send a target content retrieval request message to the content processing device, which can include information indicating the second tag. After receiving the target content retrieval request, the content processing device can parse the content retrieval request message to determine the second tag indicated by the information carried in the content retrieval request message. The content processing device can then send the target content tagged with the second tag in the content retrieval request to the server. For example, if the second tag carried in the content retrieval request message from the server indicates "movie review," the content processing device can send the target content tagged with the second tag "movie review" to the server. Alternatively, the content processing device can retrieve the target content from a database storing tagged target content and send it to the server. Alternatively, the content processing device can directly send the tagged target content to the server after obtaining the second tag from the tagging platform and generating the tagged target. In other words, the content processing device can determine the second tag of the target content through the tagging platform in parallel with receiving the target content retrieval request from the server.

[0062] In some embodiments, in block 304, the content processing device determines the popularity level of the target content based on the popularity value of the target content. In block 308, the content processing device may tag the target content with the second tag and the popularity level, thereby generating the tagged target content. In this way, when the tagged target content is provided to the user, it can be sorted and displayed based on the popularity level, making it easier for the user to obtain the more popular content, thereby improving the user experience.

[0063] In some embodiments, after the content processing device determines the score of the target content in block 304, the content processing device may tag the target content with a second tag and a score in block 308, thereby generating tagged target content. Thus, when the tagged target content is provided to the user, it may be sorted and displayed based on the score, allowing the user to more easily access content recommended by the content platform, thereby improving the user experience.

[0064] Based on the above technical solution, the content processing device can obtain target content with a first tag and, through the tagging queue of the tagging platform, obtain another refined tag for the target content with a finer granularity than the first tag. In this way, the tags of the target content can meet the needs of more application scenarios. For example, in a B2B scenario, content can be pushed to B2B users based on the second tag to meet users' needs for refined content classification.

[0065] In some embodiments of the present disclosure, the content processing device may also analyze the second tag of the tagged target content, for example, it may evaluate the effectiveness of the second tag of the tagged target content. Exemplarily, the effectiveness of the second tag may be evaluated based on the conversion rate of the tag. In an embodiment of the present disclosure, the tag conversion rate is related to the number of views of the target content by users based on the tag. Optionally, the tag conversion rate of a tag may be positively correlated with the number of times the user clicks on the target content under the corresponding tag classification. Exemplarily, for the tag "movie review", its tag conversion rate may be positively correlated with the number of clicks of the user on the target content under the "movie review" classification. For the tagged target content marked with the second tag, the content processing device may obtain the first tag conversion rate corresponding to the first tag and the second tag conversion rate corresponding to the second tag within a predefined time period. The content processing device may compare the first tag conversion rate and the second tag conversion rate to determine the effectiveness of the second tag. In some embodiments, the effectiveness of the second tag of the target content can have two levels, high and low. When the first tag conversion rate is less than the second tag conversion rate, the content processing device can determine that the effectiveness of the second tag is high. When the first tag conversion rate is greater than or equal to the second tag conversion rate, the content processing device can determine that the effectiveness of the second tag is low.

[0066] In this way, the second tag of the target content can be evaluated. If the target content is accessed less frequently by users under the more fine-grained second tag system than under the first tag system, the second tag can be determined to be of low effectiveness. The content processing device can further send the result to the administrator to adjust the second tag.

[0067] The above description, combined with FIG3 , illustrates the method for tag generation provided by an embodiment of the present disclosure. Next, combined with FIG4 , the method provided by an embodiment of the present disclosure will be further explained from an interactive perspective. FIG4 is a schematic diagram of a process for determining tags for content in an embodiment of the present disclosure. In process 400 shown in FIG4 , in step 410, a content processing device 420 retrieves content to be tagged from a content repository 410 . The content to be tagged is tagged with a first tag. In step 402 , the content processing device 420 sends information indicating the content to be tagged to a tagging platform 430 . In step 403 , annotators 440 tag the content sent to the tagging platform with a second tag. In step 404 , the content processing device 420 retrieves the tagging results from the tagging platform 430 , such as obtaining the second tag tagged with the content to be tagged, or obtaining information indicating that the content cannot be tagged with the second tag. In step 405 , the content processing device 420 synchronizes the tagging results with the content platform, such as sending the review results to a content recommendation platform 450 , a content customization platform 460 , or a content management platform 470 . The content recommendation platform 450 can recommend content to users through a recommendation algorithm, the content customization platform 460 can further filter the content based on the user's settings and provide personalized content, and the content management platform 470 can provide content to users and accept user management operations on the content. In step 406, the content processing device 420 stores the annotation results in the database 480 to facilitate subsequent analysis and acquisition of the content. For example, the content platform can also obtain the second tag marked with the content from the database 480. Steps 405 and 406 can be performed simultaneously or successively. The present disclosure does not limit the order of steps 405 and 406. In step 407, the user 490 obtains content from the content platform based on the second tag.

[0068] Through process 400, content with a first tag is tagged with another, more refined tag with finer granularity. This allows the content platform to push content to users based on the second tag, and users to access the content based on the second tag. It should be understood that the detailed implementation of each step in process 400 has been described in the description of method 300 and will not be repeated here.

[0069] FIG5 shows a schematic diagram of a method 500 for sending target content to a tagging platform, provided by an embodiment of the present disclosure. The method 500 shown in FIG5 can be executed by a content processing device, which can be, for example, the content processing device 210 in FIG2 . Exemplarily, the method 500 may include blocks 501 to 506. In block 501, the content processing device obtains target content, which is marked with a first tag. The specific implementation method of block 501 can refer to the aforementioned step 302 and will not be repeated here.

[0070] In block 502, the content processing device filters the acquired target content for content that does not conform to predetermined rules. For example, the target content may be filtered based on whether it contains blocked words, or based on whether the first tag of the target content matches a tag in a tag set preconfigured by the content processing device, or based on other predefined rules. The target content may be filtered multiple times, and the filtering rules may be different each time.

[0071] In block 503, the content processing device determines a popularity level of the target content based on the popularity of the target content, for example, the target content may be determined as normal target content or hot target content. The popularity level may also include other predefined levels, and the popularity level of the target content may be determined based on predefined rules.

[0072] In block 504, the content processing device determines a score for the target content based on at least one of the target content's tag, author, release time, or popularity. For example, the score may be related to the target content's release time, with target content with a later release time receiving a lower score. The score may also be related to the target content's first tag, author, release time, and popularity. The content processing device may be configured with a scoring function that, based on the target content's first tag, author, release time, and popularity, can calculate the target content's score using the scoring function.

[0073] In block 505, the content processing device determines the order of the target content based on the target content's score, for example, the target content may be ordered from highest to lowest score. The content processing device may also determine a sequence number for the target content based on the target content's score, thereby implementing ordering of the target content.

[0074] In block 506, the content processing device sends information indicating the target content to a tagging queue in the tagging platform corresponding to the first tag and popularity level of the target content, according to the target content's ranking. For example, the target content ID can be sent to the corresponding tagging queue in ascending order of the target content's serial number. The information indicating the target content sent by the content processing device to the tagging queue can be less than or equal to a predefined quantity threshold, that is, only content ranked before the predetermined quantity threshold can be sent to the tagging platform. The tagging platform can generate a second tag for the target content through the tagging queue.

[0075] Based on the above technical solution, the content processing device can screen and filter the target content, determine the target content's annotation queue, and sort the target content before sending the information indicating the target content to the annotation platform. This can reduce the number of target content that needs to determine the second label, thereby reducing the burden on the annotation platform and improving the efficiency of the annotation platform in determining the second label for the content. Method 500 can be implemented in combination with the aforementioned method 300. For example, box 306 can be executed after box 506. It should be understood that method 500 is merely an illustration of an embodiment of the present disclosure and cannot be a limitation of the present disclosure. The embodiments of the present disclosure can also be implemented in other ways.

[0076] Figure 6 illustrates a schematic diagram of a method 600 for determining a second label for target content and applying and analyzing the second label, in certain embodiments of the present disclosure. The method 600 illustrated in Figure 6 can be performed by a content processing device, such as the content processing device 210 in Figure 2 . Referring to Figure 6 , the method 600 may include blocks 601 through 608. In block 601, the content processing device obtains target content to be annotated, which is labeled with a first label. In block 602, the content processing device sends information indicating the target content to an annotation queue corresponding to the first label in an annotation platform. The annotation platform generates a second label for the target content, the second label having a higher classification granularity than the first label. In block 603, the content processing device obtains the second label corresponding to the target content from the annotation platform. The specific implementation of block 601 may be performed with reference to block 302 in Figure 3 , the specific implementation of block 602 may be performed with reference to block 304 in Figure 3 , and the specific implementation of block 603 may be performed with reference to block 306 in Figure 3 , and will not be further described here.

[0077] In box 604, the content processing device can obtain a target content acquisition request from a server, which can be a server that provides target content to a user device based on a second tag. The target content acquisition request can include information indicating the second tag. In box 605, the content processing device can parse the target content acquisition request, for example, extracting the field corresponding to the second tag in the target content acquisition request to determine the second tag corresponding to the target content acquisition request. In box 606, the content processing device can determine the target content corresponding to the second tag based on the second tag corresponding to the target content acquisition request, and send it to the server, which is the server in the aforementioned box 604. Through boxes 604 to 606, the content processing device can provide the target content to the server based on the second tag. The second tag has a finer classification granularity, so it can meet the needs of different types of users.

[0078] In box 607, the content processing device obtains a first tag conversion rate associated with the first tag of the target content and a second tag conversion rate associated with the second tag of the target content. The tag conversion rate of the target content may be related to the number of views of the target content by users based on the tag, for example, it may be proportional to the number of views. In some embodiments, the content processing device may obtain the number of views of the target content by users based on the tag, and calculate the tag conversion rate accordingly. In box 608, the content processing device may determine the effectiveness of the second tag of the target content based on the first tag conversion rate and the second tag conversion rate. For example, when the first tag conversion rate is less than the second tag conversion rate, it may be determined that the effectiveness of the second tag is high. Based on the schemes of boxes 607 and 608, the evaluation and analysis of the second tag obtained from the annotation platform may be implemented, and the tag generation strategy may be modified accordingly to improve the effectiveness of the method for determining tags provided in the embodiments of the present disclosure.

[0079] It is understood that method 600 is merely an example of an embodiment of the present disclosure and does not limit the present disclosure. The embodiments of the present disclosure may also be implemented in other ways. Method 600 may be implemented in combination with the aforementioned method 300 or method 500. For example, block 604 may be executed after block 308, and block 603 may be executed after block 506.

[0080] The above, in conjunction with Figures 5 and 6, respectively illustrate the method for sending target content to the annotation platform and the method for applying and analyzing the second tag in an embodiment of the present disclosure. Next, in conjunction with Figure 7, the method for generating and analyzing the second tag in an embodiment of the present disclosure is schematically illustrated. The method in Figure 7 can be executed by a device such as a computer, a computing system, a single server, a distributed server, or a cloud-based server. It can be executed by a single device or by multiple devices separately. It can be implemented through software and / or hardware. Exemplarily, it can be executed by the content processing device 210. Referring to Figure 7, in box 701, content marked with a first tag is retrieved into a pool of content to be annotated. In box 702, the content in the pool of content to be annotated is filtered for the first time using filter 1. In box 703, the content filtered by filter 1 is filtered for the second time using filter 2, and this process is repeated until box 704, when the content is filtered for the nth time using filter n. Filters 1 to n can have different filtering rules, and the order in which they are filtered can vary, which is not limited by the present disclosure. For example, in some embodiments, filter 1 may filter content based on blocked words, filter 2 may filter content based on the first tag, and filter n may filter content based on the content's publication time. In some embodiments, filter 1 may filter content based on the content's publication time, filter 2 may filter content based on blocked words, and filter n may filter content based on the first tag. In block 705, the annotation queue to which the content will be sent is determined. In embodiments of the present disclosure, the annotation queue to which the content will be sent may correspond to the content's first tag. In block 706, an annotation task is generated for the annotation platform. This annotation task may include, for example, information indicating the content to be sent to the annotation platform, information indicating the annotation queue to which the content will be sent, and information such as the content's ranking. In block 707, the annotation platform receives the annotation task and generates a finer-grained tag for the content, with a finer granularity than the first tag. This generates annotated content based on the content to be annotated, the annotated content including the second tag. In block 708, the annotated content is scored, for example, based on the content's historical page views before being annotated. In block 709, the annotated content is stored in a memory. In block 710, the quality of the annotated label is analyzed, for example, based on the amount of content in the annotated queue and the amount of annotated content corresponding to the annotated queue, the quality of the annotated label is determined.

[0081] The above content describes the method in the embodiment of the present disclosure. Next, the device provided by the present disclosure is described in conjunction with Figures 8 and 9. Figure 8 shows a block diagram of a label generation device 800 according to certain embodiments of the present disclosure. As shown in Figure 8, the device 800 includes a first acquisition module 810, which is configured to obtain target content to be labeled, and the target content is labeled with a first label. The device 800 also includes a sending module 820, which is configured to send information indicating the target content to a labeling queue corresponding to the first label in the labeling platform, and the labeling platform is used to generate a second label for the target content, and the second label has a finer classification granularity than the first label. The device 800 also includes a second acquisition module 830, which is configured to obtain a second label corresponding to the target content from the labeling platform. In addition, the device 800 also includes a storage module 840, which is configured to store the second label in association with the target content as the labeled target content, and the labeled target content is the content marked with the second label in the target content to be labeled.

[0082] Figure 9 shows a block diagram of an electronic device 900 according to certain embodiments of the present disclosure, and the electronic device 900 may be the device or apparatus described in the embodiments of the present disclosure. As shown in Figure 9, the device 900 includes a central processing unit (CPU) and / or a graphics processing unit (GPU) 901, which can perform various appropriate actions and processes according to computer program instructions stored in a read-only memory (ROM) 902 or computer program instructions loaded from a storage unit 908 into a random access memory (RAM) 903. In RAM 903, various programs and data required for the operation of the electronic device 900 can also be stored. CPU / GPU 901, ROM 902, and RAM 903 are connected to each other via a bus 904. An input / output (I / O) interface 905 is also connected to the bus 904. Although not shown in Figure 9, the electronic device 900 may also include a coprocessor.

[0083] Multiple components in the device 900 are connected to the I / O interface 905, including an input unit 906, such as a keyboard, a mouse, etc.; an output unit 907, such as various types of displays, speakers, etc.; a storage unit 908, such as a magnetic disk, an optical disk, etc.; and a communication unit 909, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 909 allows the electronic device 900 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0084] The various methods or processes described above may be executed by the CPU / GPU 901. For example, in some embodiments, the methods may be implemented as a computer software program tangibly embodied in a machine-readable medium, such as the storage unit 908. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 900 via the ROM 902 and / or the communication unit 909. When the computer program is loaded into the RAM 903 and executed by the CPU / GPU 901, one or more steps or actions in the methods or processes described above may be performed.

[0085] In some embodiments, the methods and processes described above may be implemented as a computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for executing various aspects of the present disclosure.

[0086] Computer-readable storage medium can be a tangible device that can keep and store the instructions used by the instruction execution device.Computer-readable storage medium can be, for example, but not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device or any suitable combination thereof.More specific examples (non-exhaustive list) of computer-readable storage medium include: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a static random access memory (SRAM), a portable compact disc read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanical encoding device, for example, a punch card or a convex structure in a groove having instructions stored thereon, and any suitable combination thereof.Computer-readable storage medium used herein is not interpreted as a transient signal itself, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagated by waveguides or other transmission media (for example, light pulses by fiber optic cables), or electrical signals transmitted by wires.

[0087] The computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to each computing / processing device, or downloaded to an external computer or external storage device via a network, such as the Internet, a local area network, a wide area network, and / or a wireless network. The network can include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. The network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards the computer-readable program instructions to be stored in the computer-readable storage medium in each computing / processing device.

[0088] The computer program instructions for performing the disclosed operation can be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-related instructions, microcode, firmware instructions, state setting data or source code or the object code written in any combination of one or more programming languages, programming languages ​​include object-oriented programming languages, and conventional procedural programming languages.Computer-readable program instructions can be performed completely on a user's computer, partially on a user's computer, performed as an independent software package, partly on a user's computer and partly on a remote computer, or performed completely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer by any type of network-including local area network (LAN) or wide area network (WAN), or can be connected to an external computer (such as utilizing an internet service provider to connect by the internet). In certain embodiments, by utilizing the state information of computer-readable program instructions to carry out personalized customization electronic circuits, such as programmable logic circuits, field programmable gate arrays (FPGAs) or programmable logic arrays (PLA), this electronic circuit can perform computer-readable program instructions, thereby realizing various aspects of the present disclosure.

[0089] These computer-readable program instructions can be provided to a processing unit of a general-purpose computer, a special-purpose computer, or other programmable data processing device, thereby producing a machine, so that when these instructions are executed by the processing unit of the computer or other programmable data processing device, a device is generated that implements the functions / actions specified in one or more blocks in the flowchart and / or block diagram. These computer-readable program instructions can also be stored in a computer-readable storage medium, where these instructions cause the computer, programmable data processing device, and / or other device to operate in a specific manner. Thus, the computer-readable medium storing the instructions comprises an article of manufacture that includes instructions for implementing various aspects of the functions / actions specified in one or more blocks in the flowchart and / or block diagram.

[0090] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device, so that a series of operational steps are performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions executed on the computer, other programmable data processing apparatus, or other device to implement the functions / actions specified in one or more boxes in the flowchart and / or block diagram.

[0091] The flow charts and block diagrams in the accompanying drawings show the possible architecture, functions and operations of the devices, methods and computer program products according to multiple embodiments of the present disclosure. In this regard, each box in the flow chart or block diagram can represent a part of a module, program segment or instruction, and the part of the module, program segment or instruction contains one or more executable instructions for realizing the prescribed logical function. In some alternative implementations, the functions marked in the box can also occur in a sequence different from that marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flow chart, and the combination of the boxes in the block diagram and / or flow chart, can be implemented by a special hardware-based system that performs the prescribed function or action, or can be implemented by a combination of special hardware and computer instructions.

[0092] The embodiments of the present disclosure have been described above. The above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is selected to best explain the principles of the embodiments, practical applications, or technical improvements to existing technologies, or to enable other persons skilled in the art to understand the embodiments disclosed herein.

[0093] Some example implementations of the present disclosure are listed below.

[0094] Example 1. A tag generation method for content push, comprising:

[0095] Acquire target content to be annotated, where the target content is marked with a first tag;

[0096] Sending information indicating the target content to a labeling queue corresponding to the first label in a labeling platform, wherein the labeling platform is configured to generate a second label for the target content, wherein the second label has a finer classification granularity than the first label;

[0097] Acquire the second tag corresponding to the target content from the annotation platform; and

[0098] The second tag is associated with the target content and stored as the marked target content. The marked target content is the content marked with the second tag in the target content to be marked.

[0099] Example 2. The method of Example 1, further comprising:

[0100] Determine a score for the target content based on at least one of the tag, author, popularity, and release time of the target content; and

[0101] The step of sending information indicating the target content to the marking queue includes:

[0102] Determining a ranking of the target content based on the score;

[0103] The information indicating the target content is sent to the marking queue according to the order.

[0104] Example 3. The method of Example 1 or 2, wherein storing the second tag in association with the target content as the annotated target content comprises:

[0105] The second tag, the score of the target content, and the target content are associated and stored as the labeled target content.

[0106] Example 4. The method according to any one of Examples 1-3, wherein the amount of information indicating the target content sent to the annotation queue within a unit time is less than or equal to a predefined threshold amount.

[0107] Example 5. The method of any one of Examples 1-4, wherein the amount of the target content to be annotated obtained is greater than the predefined amount threshold, and before sending the information indicating the target content to the annotation queue of the annotation platform, the method further comprises:

[0108] Based on a predefined filtering model, the target content to be marked is filtered so that the quantity of the target content is less than or equal to the predefined quantity threshold.

[0109] Example 6. The method of any one of Examples 1-5, wherein the annotation queue corresponding to the first tag includes multiple annotation queues corresponding to multiple popularity levels, and wherein sending the information indicating the target content to the annotation queue corresponding to the first tag includes:

[0110] Determining a popularity level of the target content based on the popularity of the target content;

[0111] Information indicating the target content is sent to a marking queue corresponding to the first tag and the popularity level of the target content.

[0112] Example 7. The method of any one of Examples 1-6, wherein the annotation queue includes a hotspot annotation queue and a common annotation queue, and wherein sending the target content to the annotation queue corresponding to the first tag comprises:

[0113] Determining the target content whose popularity is greater than or equal to a predetermined popularity threshold as hot target content;

[0114] Sending information indicating the hot target content to the hot spot annotation queue corresponding to the first tag of the hot target content;

[0115] Determining the target content whose popularity is less than the predetermined popularity threshold as ordinary target content; and

[0116] Information indicating the common target content is sent to the common annotation queue corresponding to the first tag of the common target content.

[0117] Example 8. The method of any one of Examples 1-7, wherein storing the second tag in association with the target content as the annotated target content comprises:

[0118] The second tag, the popularity level of the target content, and the target content are associated and stored as the marked target content.

[0119] Example 9. The method of any one of Examples 1-8, further comprising:

[0120] Obtaining a target content acquisition request from a server, wherein the server is configured to provide the target content to a user device based on the second tag;

[0121] Parsing the target content acquisition request to determine the second tag corresponding to the target content acquisition request;

[0122] The target content associated with the second tag corresponding to the target content acquisition request is sent to the server.

[0123] Example 10. The method of any one of Examples 1-9, further comprising:

[0124] Obtaining a first tag conversion rate and a second tag conversion rate of the marked target content, where the first tag conversion rate is related to the number of views of the target content by users based on the first tag, and the second tag conversion rate is related to the number of views of the target content by users based on the second tag;

[0125] The validity of the second tag of the target content is determined based on the first tag conversion rate and the second tag conversion rate.

[0126] Example 11. The method of any one of Examples 1-10, wherein the target content comprises at least one of video content, text content, or graphic content.

[0127] Example 12. A label generation device for content push, comprising:

[0128] A first acquisition module is configured to acquire target content to be annotated, where the target content is marked with a first tag;

[0129] a sending module configured to send information indicating the target content to a tagging queue corresponding to the first tag in a tagging platform, wherein the tagging platform is configured to generate a second tag for the target content, wherein the second tag has a finer classification granularity than the first tag;

[0130] A second acquisition module is configured to acquire the second tag corresponding to the target content from the annotation platform; and

[0131] The storage module is configured to store the second tag in association with the target content as annotated target content, where the annotated target content is content in the target content to be annotated that is marked with the second tag.

[0132] Example 13. The apparatus of Example 12, further comprising:

[0133] a scoring module configured to determine a score of the target content based on at least one of the tag, author, popularity, and release time of the target content; and

[0134] The sending module further includes:

[0135] a ranking submodule, configured to determine a ranking of the target content based on the score;

[0136] The first sending submodule is configured to send the information indicating the target content to the marking queue according to the order.

[0137] Example 14. The apparatus of Example 12 or 13, wherein the storage module further comprises:

[0138] The first storage submodule is configured to store the second tag, the score of the target content, and the target content in association with each other as the labeled target content.

[0139] Example 15. The apparatus according to any one of Examples 12-14, wherein the amount of information indicating the target content sent to the annotation queue within a unit time is less than or equal to a predefined amount threshold.

[0140] Example 16. The apparatus according to any one of Examples 12-15, wherein the number of target contents to be annotated acquired by the first acquisition module is greater than the predefined number threshold, and the apparatus further comprises:

[0141] The filtering module is configured to filter the target content to be annotated based on a predefined filtering model before the sending module sends the information indicating the target content to the annotation queue of the annotation platform, so that the amount of the target content is less than or equal to the predefined quantity threshold.

[0142] Example 17. The apparatus of any one of Examples 12-16, wherein the annotation queue corresponding to the first tag includes multiple annotation queues corresponding to multiple heat levels, and the sending module further includes:

[0143] a popularity level determination module, configured to determine the popularity level of the target content based on the popularity of the target content;

[0144] The second sending submodule is configured to send information indicating the target content to a marking queue corresponding to the first tag and the popularity level of the target content.

[0145] Example 18. The apparatus of any one of Examples 12-17, wherein the annotation queue includes a hotspot annotation queue and a common annotation queue, and the sending module further includes:

[0146] a hot content determination module, configured to determine the target content whose popularity is greater than or equal to a predetermined popularity threshold as hot target content;

[0147] A hot content sending module is configured to send information indicating the hot target content to the hot mark queue corresponding to the first tag of the hot target content;

[0148] a common content determination module configured to determine the target content whose popularity is less than the predetermined popularity threshold as common target content; and

[0149] The common content sending module is configured to send information indicating the common target content to the common annotation queue corresponding to the first tag of the common target content.

[0150] Example 19. The apparatus of any of Examples 12-18, wherein the storage module further comprises:

[0151] The second storage submodule is configured to store the second tag, the popularity level of the target content, and the target content in association with each other as the labeled target content.

[0152] Example 20. The apparatus of any of Examples 12-19, further comprising:

[0153] a third acquisition module, configured to acquire a target content acquisition request from a server, wherein the server is configured to provide the target content to the user device based on the second tag;

[0154] a parsing module configured to parse the target content acquisition request to determine the second tag corresponding to the target content acquisition request;

[0155] The sending module is configured to send the target content associated with the second tag corresponding to the target content acquisition request to the server.

[0156] Example 21. The apparatus of any of Examples 12-20, further comprising:

[0157] a fourth acquisition module configured to acquire a first tag conversion rate and a second tag conversion rate of the annotated target content, wherein the first tag conversion rate is related to a number of views of the target content by users based on the first tag, and the second tag conversion rate is related to a number of views of the target content by users based on the second tag;

[0158] The analysis module is configured to determine the effectiveness of the second tag of the target content based on the first tag conversion rate and the second tag conversion rate.

[0159] Example 22. The apparatus of any one of Examples 12-21, wherein the target content comprises at least one of video content, text content, or graphic content.

[0160] Example 23. An electronic device comprising:

[0161] processor; and

[0162] A memory coupled to the processor, the memory having instructions stored therein, wherein when the instructions are executed by the processor, the electronic device performs actions, the actions comprising:

[0163] Acquire target content to be annotated, where the target content is marked with a first tag;

[0164] Sending information indicating the target content to a labeling queue corresponding to the first label in a labeling platform, wherein the labeling platform is configured to generate a second label for the target content, wherein the second label has a finer classification granularity than the first label;

[0165] Acquire the second tag corresponding to the target content from the annotation platform; and

[0166] The second tag is associated with the target content and stored as the marked target content. The marked target content is the content marked with the second tag in the target content to be marked.

[0167] Example 24. The electronic device of Example 23, wherein the actions further comprise:

[0168] Determine a score for the target content based on at least one of the tag, author, popularity, and release time of the target content; and

[0169] The step of sending information indicating the target content to the marking queue includes:

[0170] Determining a ranking of the target content based on the score;

[0171] The information indicating the target content is sent to the marking queue according to the order.

[0172] Example 25. The electronic device of Example 23 or 24, wherein storing the second tag in association with the target content as the annotated target content comprises:

[0173] The second tag, the score of the target content, and the target content are associated and stored as the labeled target content.

[0174] Example 26. An electronic device according to any one of Examples 23-25, wherein the amount of information indicating the target content sent to the annotation queue within a unit time is less than or equal to a predefined threshold amount.

[0175] Example 27. The electronic device of any one of Examples 23-26, wherein the acquired amount of the target content to be annotated is greater than the predefined amount threshold, and the action further comprises:

[0176] Before sending the information indicating the target content to the annotation queue of the annotation platform, the target content to be annotated is filtered based on a predefined filtering model so that the amount of the target content is less than or equal to the predefined amount threshold.

[0177] Example 28. The electronic device of any of Examples 23-27, wherein the annotation queue corresponding to the first tag includes multiple annotation queues corresponding to multiple heat levels, and wherein sending the information indicating the target content to the annotation queue corresponding to the first tag includes:

[0178] Determining a popularity level of the target content based on the popularity of the target content;

[0179] Information indicating the target content is sent to a marking queue corresponding to the first tag and the popularity level of the target content.

[0180] Example 29. The electronic device of any of Examples 23-28, wherein the annotation queue includes a hotspot annotation queue and a common annotation queue, and wherein sending the target content to the annotation queue corresponding to the first tag comprises:

[0181] Determining the target content whose popularity is greater than or equal to a predetermined popularity threshold as hot target content;

[0182] Sending information indicating the hot target content to the hot spot annotation queue corresponding to the first tag of the hot target content;

[0183] Determining the target content whose popularity is less than the predetermined popularity threshold as ordinary target content; and

[0184] Information indicating the common target content is sent to the common annotation queue corresponding to the first tag of the common target content.

[0185] Example 30. The electronic device of any of Examples 23-29, wherein storing the second tag in association with the target content as the annotated target content comprises:

[0186] The second tag, the popularity level of the target content, and the target content are associated and stored as the marked target content.

[0187] Example 31. The electronic device of any of Examples 23-30, wherein the actions further comprise:

[0188] Obtaining a target content acquisition request from a server, wherein the server is configured to provide the target content to a user device based on the second tag;

[0189] Parsing the target content acquisition request to determine the second tag corresponding to the target content acquisition request;

[0190] The target content associated with the second tag corresponding to the target content acquisition request is sent to the server.

[0191] Example 32. The electronic device of any of Examples 23-31, wherein the actions further comprise:

[0192] Obtaining a first tag conversion rate and a second tag conversion rate of the marked target content, where the first tag conversion rate is related to the number of views of the target content by users based on the first tag, and the second tag conversion rate is related to the number of views of the target content by users based on the second tag;

[0193] The validity of the second tag of the target content is determined based on the first tag conversion rate and the second tag conversion rate.

[0194] Example 33. An electronic device according to any one of Examples 23-32, wherein the target content includes at least one of video content, text content, or graphic content.

[0195] Example 34. A computer-readable storage medium having computer-executable instructions stored thereon, wherein the computer-executable instructions are executed by a processor to implement the method according to any one of Examples 1 to 11.

[0196] Example 35. A computer program product tangibly stored on a computer-readable medium and comprising computer-executable instructions that, when executed by a device, cause the device to perform the method of any one of Examples 1 to 11.

[0197] Although the present disclosure has been described in language specific to structural features and / or methodological logical acts, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or acts described above. Rather, the specific features and acts described above are merely example forms of implementing the claims.

Claims

1. A tag generation method for content push, comprising: Acquire target content to be annotated, where the target content is marked with a first tag; Sending information indicating the target content to a labeling queue corresponding to the first label in a labeling platform, wherein the labeling platform is used to generate a second label for the target content, wherein the second label has a finer classification granularity than the first label; Acquire the second tag corresponding to the target content from the annotation platform; as well as The second tag is associated with the target content and stored as the marked target content, where the marked target content is the content marked with the second tag in the target content to be marked.

2. The method according to claim 1, further comprising: Determine a score of the target content based on at least one of a tag, an author, popularity, and a release time of the target content; as well as The step of sending information indicating the target content to the marking queue includes: Based on the score, determining a ranking of the target content; as well as The information indicating the target content is sent to the marking queue according to the order.

3. The method according to claim 2, wherein storing the second tag in association with the target content as the annotated target content comprises: The second tag, the score of the target content, and the target content are stored in association with each other as the labeled target content. 4 . The method according to claim 1 , wherein the amount of information indicating the target content sent to the marking queue within a unit time is less than or equal to a predefined amount threshold.

5. The method according to claim 4, wherein the amount of the target content to be annotated obtained is greater than the predefined amount threshold, and before sending the information indicating the target content to the annotation queue of the annotation platform, the method further comprises: Based on a predefined filtering model, the target content to be marked is filtered so that the quantity of the target content is less than or equal to the predefined quantity threshold.

6. The method according to claim 1, wherein the annotation queue corresponding to the first tag includes a plurality of annotation queues corresponding to a plurality of heat levels, and wherein sending the information indicating the target content to the annotation queue corresponding to the first tag comprises: Determining a popularity level of the target content based on the popularity of the target content; Information indicating the target content is sent to a marking queue corresponding to the first tag and the popularity level of the target content.

7. The method according to claim 6, wherein the annotation queue comprises a hotspot annotation queue and a common annotation queue, and wherein sending the target content to the annotation queue corresponding to the first tag comprises: Determine the target content whose heat is greater than or equal to a predetermined heat threshold as hot target content; Sending information indicating the hot target content to the hot spot annotation queue corresponding to the first tag of the hot target content; Determining the target content whose popularity is less than the predetermined popularity threshold as common target content; as well as Information indicating the common target content is sent to the common annotation queue corresponding to the first tag of the common target content.

8. The method according to claim 6, wherein storing the second tag in association with the target content as the annotated target content comprises: The second tag, the popularity level of the target content, and the target content are stored in association with each other as the marked target content.

9. The method according to claim 1, further comprising: Acquire a target content acquisition request from a server, wherein the server is used to provide the target content to a user device based on the second tag; Parsing the target content acquisition request to determine the second tag corresponding to the target content acquisition request; as well as The target content associated with the second tag corresponding to the target content acquisition request is sent to the server.

10. The method according to claim 1, further comprising: Acquire a first tag conversion rate and a second tag conversion rate of the marked target content, wherein the first tag conversion rate is related to the number of views of the target content by users based on the first tag, and the second tag conversion rate is related to the number of views of the target content by users based on the second tag; as well as Based on the first tag conversion rate and the second tag conversion rate, the validity of the second tag of the target content is determined. The method according to claim 1 , wherein the target content comprises at least one of video content, text content or graphic content.

12. A label generation device for content push, comprising: A first acquisition module is configured to acquire target content to be annotated, wherein the target content is marked with a first tag; a sending module configured to send information indicating the target content to a labeling queue corresponding to the first label in a labeling platform, wherein the labeling platform is used to generate a second label for the target content, wherein the second label has a finer classification granularity than the first label; A second acquisition module is configured to acquire the second tag corresponding to the target content from the annotation platform; as well as The storage module is configured to store the second tag in association with the target content as annotated target content, where the annotated target content is content in the target content to be annotated that is marked with the second tag.

13. An electronic device, comprising: processor; as well as A memory coupled to the processor, the memory having instructions stored therein, wherein when the instructions are executed by the processor, the electronic device executes the method according to any one of claims 1 to 11.

14. A computer-readable storage medium having computer-executable instructions stored thereon, wherein the computer-executable instructions are executed by a processor to implement the method according to any one of claims 1 to 11.

Images