Content prediction method, device, electronic device and storage medium

Abstract

The present disclosure relates to a content prediction method, apparatus, electronic device, storage medium, and computer program product. The method comprises: obtaining content to be predicted; obtaining initial release information for the content to be predicted; the initial release information being used to represent pre-set associated information for the release of the content to be predicted; and inputting the content to be predicted and the initial release information into a pre-trained content prediction model to obtain prediction information for the content to be predicted; the prediction information including at least content quality information and content conversion information. This method can improve content prediction efficiency.

Description

Technical Field

[0001] The present disclosure relates to the field of Internet technology, and in particular to a content prediction method, device, electronic device, storage medium, and computer program product. Background Art

[0002] With the development of Internet technology, recommendation systems will recommend content to accounts based on their needs.

[0003] In related technologies, before content recommendation, it is necessary to determine the predicted information of the content, such as the content quality. This predicted information is generally determined through manual analysis; however, determining the predicted information of the content through manual analysis is time-consuming, resulting in low content prediction efficiency. Summary of the Invention

[0004] The present disclosure provides a content prediction method, apparatus, electronic device, storage medium, and computer program product to at least address the problem of low content prediction efficiency in related technologies. The technical solutions of the present disclosure are as follows:

[0005] According to a first aspect of an embodiment of the present disclosure, a content prediction method is provided, including:

[0006] Obtain the content to be predicted;

[0007] Acquiring initial release information for the content to be predicted; the initial release information is used to represent preset associated information for releasing the content to be predicted;

[0008] The content to be predicted and the initial release information are input into a pre-trained content prediction model to obtain prediction information of the content to be predicted; the prediction information at least includes content quality information and content conversion information.

[0009] In an exemplary embodiment, inputting the content to be predicted and the initial release information into a pre-trained content prediction model to obtain prediction information of the content to be predicted includes:

[0010] Inputting the content to be predicted and the initial release information into a pre-trained content prediction model, performing content conversion prediction processing on the content to be predicted and the initial release information through the content prediction model to obtain a content operation probability of the content to be predicted;

[0011] The content conversion information is determined according to the content operation probability of the content to be predicted.

[0012] In an exemplary embodiment, inputting the content to be predicted and the initial release information into a pre-trained content prediction model to obtain prediction information of the content to be predicted further includes:

[0013] Inputting the content to be predicted and the initial release information into a pre-trained content prediction model, performing content quality prediction processing on the content to be predicted using the content prediction model to obtain quality information of the content to be predicted in various preset content quality dimensions;

[0014] The content quality information is determined according to the quality information of the content to be predicted in each preset content quality dimension.

[0015] In an exemplary embodiment, the initial publishing information includes an initial content publishing value and an initial recommended account set, and the prediction information also includes a target content publishing value and a target recommended account set; the method further includes:

[0016] The initial content publishing value and the initial recommended account set are updated through the content prediction model to obtain the target content publishing value and the target recommended account set.

[0017] In an exemplary embodiment, the prediction information further includes a first quality label and a second quality label, wherein the first quality label is a label that triggers an increase in a content quality score corresponding to the content quality information, and the second quality label is a label that triggers a decrease in the content quality score corresponding to the content quality information; the method further includes:

[0018] If the prediction information does not meet a preset condition, generating content update information according to the first quality label and the second quality label;

[0019] The content to be predicted is updated according to the content update information.

[0020] In an exemplary embodiment, the prediction information further includes at least one of related content, content classification labels, and content understanding labels of the content to be predicted; and the method further includes:

[0021] If the prediction information does not satisfy the preset condition, screening, based on the content classification label and the content understanding label, from a preset content library, content whose content similarity with the content to be predicted satisfies a first condition and whose content quality score corresponding to the content quality information satisfies a second condition, as associated content of the content to be predicted;

[0022] Filtering target content whose content quality score meets a third condition from the associated content;

[0023] The target content is published according to the content publishing information associated with the target content.

[0024] In an exemplary embodiment, the method further comprises:

[0025] In a case where the prediction information satisfies the preset condition, determining content publishing information associated with the content to be predicted according to the prediction information;

[0026] The content to be predicted is published according to the content publishing information associated with the content to be predicted.

[0027] In an exemplary embodiment, after inputting the content to be predicted and the initial release information into a pre-trained content prediction model to obtain prediction information of the content to be predicted, the method further includes:

[0028] In response to a re-prediction instruction for the content to be predicted, the content to be predicted and the initial release information are input into an updated content prediction model to obtain updated prediction information of the content to be predicted; the updated content prediction model is obtained by retraining the pre-trained content prediction model.

[0029] According to a second aspect of an embodiment of the present disclosure, there is provided a content prediction apparatus, comprising:

[0030] A content acquisition unit, configured to acquire content to be predicted;

[0031] An information acquisition unit is configured to acquire initial release information for the content to be predicted; the initial release information is used to represent preset associated information for releasing the content to be predicted;

[0032] The content prediction unit is configured to input the content to be predicted and the initial release information into a pre-trained content prediction model to obtain prediction information of the content to be predicted; the prediction information includes at least content quality information and content conversion information.

[0033] In an exemplary embodiment, the content prediction unit is further configured to input the content to be predicted and the initial release information into a pre-trained content prediction model, perform content conversion prediction processing on the content to be predicted and the initial release information through the content prediction model, and obtain the content operation probability of the content to be predicted; and determine the content conversion information based on the content operation probability of the content to be predicted.

[0034] In an exemplary embodiment, the content prediction unit is further configured to input the content to be predicted and the initial release information into a pre-trained content prediction model, perform content quality prediction processing on the content to be predicted through the content prediction model, and obtain quality information of the content to be predicted in each preset content quality dimension; determine the content quality information based on the quality information of the content to be predicted in each preset content quality dimension.

[0035] In an exemplary embodiment, the initial publishing information includes an initial content publishing value and an initial recommended account set, and the prediction information also includes a target content publishing value and a target recommended account set;

[0036] The apparatus further includes an information updating unit configured to update the initial content publishing value and the initial recommended account set using the content prediction model to obtain the target content publishing value and the target recommended account set.

[0037] In an exemplary embodiment, the prediction information further includes a first quality label and a second quality label, wherein the first quality label is a label that triggers an increase in a content quality score corresponding to the content quality information, and the second quality label is a label that triggers a decrease in the content quality score corresponding to the content quality information;

[0038] The apparatus further includes a content updating unit configured to generate content updating information according to the first quality label and the second quality label when the prediction information does not meet a preset condition; and update the content to be predicted according to the content updating information.

[0039] In an exemplary embodiment, the prediction information further includes at least one of related content, content classification labels, and content understanding labels of the content to be predicted;

[0040] The device also includes a first publishing unit, which is configured to execute, when the prediction information does not meet the preset condition, based on the content classification label and the content understanding label, screening out from a preset content library content whose content similarity with the content to be predicted meets a first condition and whose content quality score corresponding to the content quality information meets a second condition, as associated content of the content to be predicted; screening out from the associated content target content whose content quality score meets a third condition; and publishing the target content according to the content publishing information associated with the target content.

[0041] In an exemplary embodiment, the device also includes a second publishing unit, which is configured to determine content publishing information associated with the content to be predicted based on the prediction information when the prediction information meets the preset condition; and publish the content to be predicted according to the content publishing information associated with the content to be predicted.

[0042] In an exemplary embodiment, the device also includes a re-prediction unit, which is configured to execute a re-prediction instruction in response to the content to be predicted, input the content to be predicted and the initial release information into an updated content prediction model, and obtain updated prediction information of the content to be predicted; the updated content prediction model is obtained by re-training the pre-trained content prediction model.

[0043] According to a third aspect of an embodiment of the present disclosure, there is provided an electronic device, including:

[0044] processor;

[0045] a memory for storing instructions executable by the processor;

[0046] The processor is configured to execute the instructions to implement any of the above content prediction methods.

[0047] According to a fourth aspect of an embodiment of the present disclosure, a computer-readable storage medium is provided. When instructions in the computer-readable storage medium are executed by a processor of an electronic device, the electronic device is enabled to perform any of the content prediction methods described above.

[0048] According to a fifth aspect of an embodiment of the present disclosure, a computer program product is provided, wherein the computer program product includes instructions, and when the instructions are executed by a processor of an electronic device, the electronic device is capable of executing any of the content prediction methods described above.

[0049] The technical solutions provided by the embodiments of the present disclosure bring at least the following beneficial effects:

[0050] The method obtains the content to be predicted, then obtains initial release information for the content to be predicted; the initial release information is used to represent the pre-set associated information for the release of the content to be predicted; and finally, the content to be predicted and the initial release information are input into a pre-trained content prediction model to obtain prediction information for the content to be predicted; the prediction information includes at least content quality information and content conversion information. This achieves the goal of automatically obtaining prediction information for the content to be predicted based on the pre-trained content prediction model, eliminating the need for manual analysis. This simplifies the process of determining content prediction information, saves considerable time, and improves content prediction efficiency.

[0051] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] The accompanying drawings herein are incorporated into and constitute a part of the specification, illustrate embodiments consistent with the present disclosure, and together with the description are used to explain the principles of the present disclosure, and do not constitute an improper limitation of the present disclosure.

[0053] Figure 1 The figure is a flowchart of a content prediction method according to an exemplary embodiment.

[0054] Figure 2 The figure is a flowchart showing steps of publishing target content according to an exemplary embodiment.

[0055] Figure 3 is a flowchart showing another content prediction method according to an exemplary embodiment.

[0056] Figure 4 is a flowchart of yet another content prediction method according to an exemplary embodiment.

[0057] Figure 5 is a flowchart of yet another content prediction method according to an exemplary embodiment.

[0058] Figure 6 It is a block diagram of a content prediction device according to an exemplary embodiment.

[0059] Figure 7 It is a block diagram of an electronic device according to an exemplary embodiment. DETAILED DESCRIPTION

[0060] In order to enable ordinary persons in the art to better understand the technical solutions of the present disclosure, the technical solutions in the embodiments of the present disclosure will be clearly and completely described below with reference to the accompanying drawings.

[0061] It should be noted that the terms "first," "second," and the like in the specification and claims of the present disclosure and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or precedence. It should be understood that the numbers used in this manner are interchangeable where appropriate so that the embodiments of the present disclosure described herein can be implemented in an order other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present disclosure. Instead, they are merely examples of apparatus and methods consistent with certain aspects of the present disclosure as detailed in the appended claims.

[0062] It should also be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for display, data for analysis, etc.) involved in this disclosure are all information and data authorized by the user or fully authorized by all parties.

[0063] Figure 1 is a flow chart of a content prediction method according to an exemplary embodiment. Figure 1 As shown, the content prediction method is used in a terminal; it is understood that the method can also be applied to a server, or to a system including a terminal and a server, and implemented through interaction between the terminal and the server. In this exemplary embodiment, the method includes the following steps:

[0064] In step S110 , the content to be predicted is obtained.

[0065] The content to be predicted refers to the content for which prediction information needs to be determined. Specifically, it refers to advertising materials, such as videos and images. This can be uploaded by users or directly retrieved from a local database. It should be noted that before publishing content, it is necessary to evaluate the content to confirm whether it should be updated or published.

[0066] Specifically, in response to a content selection operation, the terminal determines the selected content as the content to be predicted; or, in response to a content upload operation, the terminal obtains the uploaded content as the content to be predicted. For example, a user selects content to be uploaded on a content prediction interface provided by the terminal and clicks the upload button, triggering the generation of a content upload request. In response to the content upload request, the terminal obtains the user-uploaded content as the content to be predicted, facilitating subsequent prediction processing of the content to be predicted.

[0067] In step S120 , initial release information for the content to be predicted is obtained; the initial release information is used to represent preset associated information for releasing the content to be predicted.

[0068] The initial release information refers to preset associated information for the release of the predicted content, such as the content release value, the recommended account set, the content release target, the content type, etc.

[0069] The content publishing value refers to the publishing cost of the content to be predicted, specifically the cost of publishing the content to be predicted to the accounts in the recommended account set. A recommended account set is a collection of multiple accounts, specifically accounts with the same account characteristics, and is used to represent the recommended content for the content to be predicted. Account characteristics represent account attributes, including age and region. In real-world scenarios, the recommended account set is determined by the recommended account set features, which represent the common account attributes of the accounts included in the recommended account set, including age and region.

[0070] The content publishing goal refers to the goal of publishing the content to be predicted, such as increasing click-through rate, increasing download rate, etc. The content type refers to the type of content to be predicted published to the accounts in the recommended account set.

[0071] Specifically, in response to the selection operation of the release information of the content to be predicted, the terminal obtains the selected release information, such as the content release value, the recommended account set, etc., and uses the selected release information as the initial release information for the content to be predicted.

[0072] For example, on the content prediction interface provided by the terminal, a user selects content publishing information such as content publishing value and recommended account set. After selecting the content publishing information, the user clicks the content prediction button, triggering the generation of a content prediction request. The terminal parses the content prediction request, confirms the content publishing information selected by the user, and uses it as the initial publishing information for the content to be predicted.

[0073] In step S130, the content to be predicted and the initial release information are input into a pre-trained content prediction model to obtain prediction information of the content to be predicted; the prediction information includes at least content quality information and content conversion information.

[0074] Among them, the pre-trained content prediction model is a model that can automatically output prediction information of content, such as a deep learning model, which is specifically used to predict content quality information and content conversion information of the content; the content prediction model includes at least a content quality prediction network and a content conversion prediction network.

[0075] The content quality information is used to measure the quality of the content to be predicted, such as clarity, subtitles, background music, etc., and can be represented by a content quality score. The higher the content quality score of the content to be predicted, the higher the content quality of the content to be predicted.

[0076] Content conversion information is used to measure the conversion effect of the predicted content after it is released, such as predicted click-through rate and predicted conversion rate, and can be represented by a content conversion score. The higher the content conversion score of the predicted content, the better the content conversion effect of the predicted content.

[0077] It should be noted that the prediction information also includes at least one of the related content, similarity, content level, content classification label, content understanding label, first quality label, and second quality label of the content to be predicted.

[0078] Specifically, the terminal inputs the content to be predicted and the initial release information into a pre-trained content prediction model. The content quality prediction network within the content prediction model performs content quality prediction on the content to be predicted, obtaining content quality information for the content to be predicted. The content conversion prediction network within the content prediction model, based on the initial content release information, performs content conversion prediction on the content to be predicted, obtaining content conversion information for the content to be predicted. Based on the content quality information and content conversion information for the content to be predicted, the terminal determines the predicted information for the content to be predicted. Furthermore, the terminal can display the predicted information for the content to be predicted on the content prediction interface for easy viewing by the user.

[0079] Furthermore, the pre-trained content prediction model is trained in the following manner: the terminal obtains sample data; the sample data includes sample content, sample release information for the sample content, actual content quality information corresponding to the sample content, and actual content conversion information; the sample content and sample release information are input into the neural network model to be trained to obtain the predicted content quality information and predicted content conversion information of the sample content; the loss value is obtained based on the difference between the predicted content quality information and the actual content quality information, and the difference between the predicted content conversion information and the actual content conversion information; the network parameters of the neural network model to be trained are adjusted based on the loss value; the neural network model after the network parameters are adjusted is repeatedly trained until the training end condition is met, and the trained neural network model that meets the training end condition is used as the pre-trained content prediction model.

[0080] Furthermore, the prediction information for the content to be predicted includes, in addition to content quality information and content conversion information, a content rating for the content to be predicted. After obtaining the content quality information and content conversion information for the content to be predicted, the terminal can also determine the content rating of the content to be predicted, such as "excellent," based on the content quality information and content conversion information. For example, the terminal performs a weighted summation of the content quality score corresponding to the content quality information and the content conversion score corresponding to the content conversion information to obtain the content score for the content to be predicted; the terminal then obtains the rating corresponding to the content score as the content rating for the content to be predicted.

[0081] Furthermore, the prediction information for the content to be predicted also includes the similarity of the content to be predicted. This similarity is used to measure the similarity between the content to be predicted and existing content, specifically to characterize the originality of the content to be predicted. After obtaining the content quality information and content conversion information for the content to be predicted, the terminal can also analyze and process the content characteristics of the content to be predicted using a content prediction model to determine the similarity between the content to be predicted and existing content. This similarity between the content to be predicted and existing content is then used to determine the originality of the content to be predicted.

[0082] In the above-mentioned content prediction method, the content to be predicted is obtained; then, initial release information for the content to be predicted is obtained; the initial release information is used to represent the preset associated information for the release of the content to be predicted; finally, the content to be predicted and the initial release information are input into a pre-trained content prediction model to obtain prediction information for the content to be predicted; the prediction information includes at least content quality information and content conversion information. This achieves the goal of automatically obtaining prediction information for the content to be predicted based on the pre-trained content prediction model, eliminating the need for manual analysis. This simplifies the process of determining content prediction information, saves considerable time, and improves content prediction efficiency.

[0083] In an exemplary embodiment, the above-mentioned step S130 inputs the content to be predicted and the initial release information into a pre-trained content prediction model to obtain prediction information of the content to be predicted, specifically including: inputting the content to be predicted and the initial release information into a pre-trained content prediction model, performing content conversion prediction processing on the content to be predicted and the initial release information through the content prediction model to obtain the content operation probability of the content to be predicted; and determining the content conversion information based on the content operation probability of the content to be predicted.

[0084] The content operation probability of the content to be predicted may refer to the content click probability, content conversion probability, etc.

[0085] Specifically, the terminal inputs the content to be predicted and the initial release information into a pre-trained content prediction model, extracts the content features of the content to be predicted and the information features of the initial release information through the content conversion prediction network in the content prediction model, and performs content conversion prediction processing on the content features of the content to be predicted and the information features of the initial release information to obtain the content operation probability of the content to be predicted; performs conversion processing on the content operation probability of the content to be predicted to obtain a content conversion score corresponding to the content operation probability as the content conversion score of the content to be predicted; and determines the content conversion score of the content to be predicted as the content conversion information of the content to be predicted.

[0086] For example, a pre-trained content prediction model is used to perform content conversion prediction on the content to be predicted and the initial release information to obtain the content click probability of the content to be predicted; based on the content click probability of the content to be predicted, the content conversion score of the content to be predicted is determined as the content conversion information of the content to be predicted.

[0087] Furthermore, the prediction information for the content to be predicted also includes a content classification label and a content understanding label for the content to be predicted. During the process of obtaining the content quality information and content conversion information for the content to be predicted, the terminal can also analyze and process the content features of the content to be predicted using a content prediction model to obtain a content classification label and a content understanding label for the content to be predicted. The content classification label represents the categorization of the content to be predicted, such as electronic products or cosmetics. The content understanding label represents the content information of the content to be predicted, such as musical effects and contrast.

[0088] The technical solution provided by the embodiments of the present disclosure automatically outputs content conversion information for the content to be predicted using a pre-trained content prediction model, eliminating the need for manual analysis, thereby improving the efficiency of determining content conversion information. Furthermore, outputting content conversion information for the content to be predicted using a pre-trained content prediction model avoids the drawbacks of manual analysis, which is prone to errors and results in low accuracy of the resulting content conversion information, further improving the accuracy of determining content conversion information.

[0089] In an exemplary embodiment, the above-mentioned step S130, in which the content to be predicted and the initial release information are input into a pre-trained content prediction model to obtain prediction information of the content to be predicted, also includes: inputting the content to be predicted and the initial release information into a pre-trained content prediction model, performing content quality prediction processing on the content to be predicted through the content prediction model, and obtaining quality information of the content to be predicted in each preset content quality dimension; and determining content quality information based on the quality information of the content to be predicted in each preset content quality dimension.

[0090] The preset content quality dimensions refer to various dimensions used to measure the quality of the content to be predicted, such as the content perception. The quality information of the preset content quality dimensions can also be represented by a content quality score.

[0091] Specifically, the terminal inputs the content to be predicted and the initial release information into a pre-trained content prediction model, extracts the content features of the content to be predicted through the content quality prediction network in the content prediction model, and performs content quality prediction processing on the content features of the content to be predicted to obtain the quality information of the content to be predicted in each preset content quality dimension; performs weighted sum processing on the content quality scores corresponding to the quality information of the content to be predicted in each preset content quality dimension to obtain the content quality score of the content to be predicted as the content quality information of the content to be predicted.

[0092] For example, the content quality scores corresponding to the quality information of the content to be predicted in the three preset content quality dimensions are A1, A2 and A3 respectively, and the corresponding weights are a1, a2 and a3 respectively, then the content quality information of the content to be predicted is A1×a1+A2×a2+A3×a3.

[0093] Furthermore, the prediction information of the content to be predicted also includes a first quality label and a second quality label of the content to be predicted. In the process of obtaining the content quality information and content conversion information of the content to be predicted, the terminal can also analyze and process the content features of the content to be predicted through the content prediction model to obtain the first quality label and the second quality label of the content to be predicted. Among them, the first quality label is a label that triggers an increase in the content quality score corresponding to the content quality information of the content to be predicted, specifically refers to a label that affects points, such as clear picture quality, no obvious occlusion, clear sound, etc.; in actual scenarios, the first quality label refers to the advantages of the content to be predicted. The second quality label is a label that triggers a decrease in the content quality score corresponding to the content quality information of the content to be predicted, specifically refers to a label that affects points, such as blurred picture quality, obvious occlusion, unclear sound, etc.; in actual scenarios, the second quality label refers to the shortcomings of the content to be predicted.

[0094] The technical solution provided by the embodiments of the present disclosure automatically outputs content quality information for the content to be predicted using a pre-trained content prediction model, eliminating the need for manual analysis and thus improving the efficiency of determining content quality information. Furthermore, outputting content quality information for the content to be predicted using a pre-trained content prediction model avoids the drawbacks of manual analysis, which is prone to errors and results in low accuracy of the resulting content quality information, further improving the accuracy of content quality information determination.

[0095] In an exemplary embodiment, the initial release information includes an initial content release value and an initial recommended account set, and the prediction information also includes a target content release value and a target recommended account set; the content prediction method provided by the present disclosure also includes the following content: through a content prediction model, the initial content release value and the initial recommended account set are updated to obtain the target content release value and the target recommended account set.

[0096] The initial content publishing value refers to the preset publishing cost of the content to be predicted, specifically the preset cost of publishing the content to be predicted to the accounts in the initial recommended account set. The initial recommended account set refers to a set composed of multiple accounts, specifically a set composed of accounts with the same account characteristics, and is used to represent the preset recommendation objects for the content to be predicted. The target content publishing value refers to the predicted publishing cost of the content to be predicted, specifically the predicted cost of publishing the content to be predicted to the accounts in the target recommended account set. The target recommended account set refers to a set composed of multiple accounts, specifically a set composed of accounts with the same account characteristics, and is used to represent the predicted recommendation objects for the content to be predicted.

[0097] Specifically, in the process of obtaining the content quality information and content conversion information of the content to be predicted, the terminal can also analyze and process the initial content release value and the initial recommended account set through the content prediction model to obtain the analysis results; based on the analysis results, the initial content release value and the initial recommended account set are updated to obtain the updated content release value and the updated recommended account set, which correspond to the target content release value and the target recommended account set.

[0098] The technical solution provided by the embodiments of the present disclosure updates the initial content release value and the initial recommended account set through a content prediction model to obtain a target content release value and a target recommended account set, thereby achieving the purpose of updating the initial content release value and the initial recommended account set, while avoiding the defect of low accuracy of the manually set initial content release value and the initial recommended account set, thereby improving the accuracy of determining the content release value and the recommended account set of the content to be predicted.

[0099] In an exemplary embodiment, the content prediction method provided by the present disclosure further includes the following: generating content update information based on the first quality label and the second quality label when the prediction information does not meet the preset conditions; and updating the predicted content based on the content update information.

[0100] Among them, the preset conditions refer to that the content quality score corresponding to the content quality information is greater than the content quality score threshold, the content conversion score corresponding to the content conversion information is greater than the content conversion score threshold, etc.

[0101] The content update information refers to a suggestion for updating the predicted content.

[0102] Specifically, the terminal compares the prediction information of the content to be predicted with the preset conditions. If the prediction information does not meet the preset conditions, the terminal generates an instruction through content update information, and generates content update information based on the first quality label and the second quality label of the content to be predicted; according to the content update information, the predicted content is updated to obtain the updated content.

[0103] For example, when the content quality score corresponding to the content quality information of the content to be predicted is less than or equal to the content quality score threshold, or when the content conversion score corresponding to the content conversion information of the content to be predicted is less than or equal to the content conversion score threshold, the terminal generates content update information based on the first quality label and the second quality label, and updates the content to be predicted based on the content update information to obtain updated content.

[0104] The technical solution provided by the embodiments of the present disclosure generates content update information based on the first quality label and the second quality label when the prediction information does not meet the preset conditions, and updates the content to be predicted based on the content update information, thereby achieving the purpose of updating the content to be predicted when the content to be predicted does not meet the requirements, and avoiding the defect that the prediction information of the content to be predicted does not meet the preset conditions, resulting in poor publishing effect of the content to be predicted.

[0105] In an exemplary embodiment, Figure 2 As shown, the content prediction method provided by the present disclosure also includes the step of publishing target content, which specifically includes the following steps:

[0106] In step S210, when the prediction information does not meet the preset conditions, based on the content classification label and the content understanding label, the content whose content similarity with the content to be predicted meets the first condition and the content quality score corresponding to the content quality information meets the second condition is screened out from the preset content library as the associated content of the content to be predicted.

[0107] In step S220, target content whose content quality score meets the third condition is screened out from the associated content.

[0108] In step S230 , the target content is published according to the content publishing information associated with the target content.

[0109] The first condition refers to that the content similarity is greater than a preset similarity; the content similarity between the content to be predicted and the content to be predicted satisfies the first condition, which means that the content similarity between the content to be predicted and the content to be predicted is greater than a preset similarity.

[0110] Among them, the second condition means that the content quality score is greater than a preset score; the content quality score corresponding to the content quality information satisfies the second condition means that the content quality score corresponding to the content quality information is greater than the preset score.

[0111] Among them, the third condition means that the content quality score is the largest; the content quality score meeting the third condition means that the content quality score is the largest.

[0112] The associated content of the content to be predicted refers to content whose content similarity with the content to be predicted is greater than a preset similarity and whose content quality score corresponding to the content quality information is greater than a preset score, specifically high-quality content.

[0113] The content publishing information associated with the target content refers to the content publishing plan of the target content.

[0114] Among them, the preset content library pre-stores a plurality of contents with high content quality scores, and each content carries corresponding content quality information, content classification label and content understanding label.

[0115] Specifically, the terminal compares the prediction information of the content to be predicted with the preset conditions. If the prediction information does not meet the preset conditions, the terminal counts the content similarity between the content in the preset content library and the content to be predicted based on the content classification label and content understanding label of the content to be predicted; filters out from the preset content library the content whose content similarity with the content to be predicted is greater than the preset similarity and whose content quality score corresponding to the content quality information is greater than the preset score, and uses it as the associated content of the content to be predicted; filters out the target content with the largest content quality score from the associated content; obtains the content publishing information associated with the target content, and recommends the target content to the corresponding account according to the content publishing information associated with the target content.

[0116] For example, after filtering out the target content with the highest content quality score from the associated content, the terminal obtains the content publishing information associated with the target content and sends the content publishing information associated with the target content to the content publishing system. The content publishing system recommends the target content to the corresponding account according to the content publishing information associated with the target content.

[0117] The technical solution provided by the embodiments of the present disclosure, when the prediction information does not meet the preset conditions, filters out the related content of the content to be predicted from the preset content library based on the content classification label and the content understanding label, thereby achieving the purpose of automatically recommending the related content of the content to be predicted; at the same time, filters out the target content with the highest content quality score from the related content of the content to be predicted, and publishes the target content according to the content publishing information associated with the target content, which is conducive to improving the content publishing effect and avoids the defect that the prediction information of the content to be predicted does not meet the preset conditions, resulting in a poor publishing effect of the content to be predicted.

[0118] In an exemplary embodiment, the content prediction method provided by the present disclosure further includes: determining content publishing information associated with the content to be predicted based on the prediction information when the prediction information meets a preset condition; and publishing the content to be predicted according to the content publishing information associated with the content to be predicted.

[0119] The content release information associated with the content to be predicted refers to a content release plan for the content to be predicted.

[0120] Specifically, the terminal compares the prediction information of the content to be predicted with the preset conditions. When the prediction information meets the preset conditions, the terminal obtains the content publishing information associated with the content to be predicted, and recommends the content to be predicted to the corresponding account according to the content publishing information associated with the content to be predicted.

[0121] For example, when the content quality score corresponding to the content quality information of the content to be predicted is greater than the content quality score threshold, or when the content conversion score corresponding to the content conversion information of the content to be predicted is greater than the content conversion score threshold, the terminal analyzes and processes the prediction information, obtains the content publishing information associated with the content to be predicted, and sends the content publishing information associated with the content to be predicted to the content publishing system, which recommends the content to be predicted to the corresponding account according to the content publishing information associated with the content to be predicted.

[0122] The technical solution provided by the embodiment of the present disclosure determines the content publishing information associated with the content to be predicted based on the prediction information when the prediction information meets the preset conditions, and publishes the content to be predicted according to the content publishing information associated with the content to be predicted, which is conducive to improving the content publishing effect and the content recommendation accuracy.

[0123] In an exemplary embodiment, after inputting the content to be predicted and the initial release information into a pre-trained content prediction model to obtain prediction information of the content to be predicted, it also includes: in response to a re-prediction instruction for the content to be predicted, inputting the content to be predicted and the initial release information into an updated content prediction model to obtain updated prediction information of the content to be predicted; the updated content prediction model is obtained by re-training the pre-trained content prediction model.

[0124] The re-prediction instruction is used to re-determine the prediction information of the content to be predicted. The updated content prediction model can also be used to predict content quality information and content conversion information of the content, specifically by re-training the pre-trained content prediction model using real-time information in the database.

[0125] The update prediction information of the content to be predicted includes at least the update content quality information and the update content conversion information of the content to be predicted.

[0126] Specifically, after a preset time period, the terminal responds to a re-prediction instruction for the content to be predicted, and inputs the content to be predicted and the initial release information into the updated content prediction model; performs content quality prediction processing on the content to be predicted again through the content quality prediction network in the updated content prediction model to obtain updated content quality information of the content to be predicted; performs content conversion prediction processing on the content to be predicted again based on the initial content release information through the content conversion prediction network in the updated content prediction model to obtain updated content conversion information of the content to be predicted; and determines updated prediction information of the content to be predicted based on the updated content quality information and updated content conversion information of the content to be predicted.

[0127] For example, after the content to be predicted has been uploaded for a period of time, if the user needs to re-determine the prediction information of the content to be predicted before publishing the content to be predicted, the user can click the re-detection button on the content prediction interface to trigger a re-prediction instruction for the content to be predicted; the terminal responds to the re-prediction instruction for the content to be predicted, and re-predicts the content to be predicted and the initial release information through the updated content prediction model to obtain updated prediction information for the content to be predicted.

[0128] The technical solution provided by the embodiment of the present disclosure utilizes an updated content prediction model to re-predict the content to be predicted and the initial release information, thereby obtaining updated prediction information of the content to be predicted, thereby achieving the purpose of re-predicting the content to be predicted and further improving the accuracy of determining the prediction information.

[0129] Figure 3 is a flow chart showing another content prediction method according to an exemplary embodiment. Figure 3 As shown, the content prediction method is used in a terminal and includes the following steps:

[0130] In step S310, the content to be predicted is obtained.

[0131] In step S320, initial release information for the content to be predicted is obtained; the initial release information is used to represent preset associated information for releasing the content to be predicted.

[0132] In step S330, the content to be predicted and the initial release information are input into a pre-trained content prediction model to obtain prediction information of the content to be predicted; the prediction information includes at least content quality information and content conversion information.

[0133] In step S340 , when the prediction information does not meet the preset condition, content update information is generated according to the first quality label and the second quality label; and the predicted content is updated according to the content update information.

[0134] In step S350, when the prediction information does not meet the preset conditions, based on the content classification label and the content understanding label, the content whose content similarity with the content to be predicted meets the first condition and whose content quality score corresponding to the content quality information meets the second condition is screened out from the preset content library as the associated content of the content to be predicted; the target content whose content quality score meets the third condition is screened out from the associated content; and the target content is published according to the content publishing information associated with the target content.

[0135] In step S360, when the prediction information satisfies a preset condition, content publishing information associated with the content to be predicted is determined according to the prediction information; and the content to be predicted is published according to the content publishing information associated with the content to be predicted.

[0136] The above-described content prediction method achieves the goal of automatically obtaining prediction information for the content to be predicted based on a pre-trained content prediction model, eliminating the need for manual analysis. This simplifies the process of determining content prediction information, saves significant time, and improves content prediction efficiency. Furthermore, by utilizing the pre-trained content prediction model to determine prediction information for the content to be predicted, it avoids the drawbacks of manual analysis, which is prone to errors and results in low accuracy of the content prediction information, further improving the accuracy of content prediction information determination.

[0137] To more clearly illustrate the content prediction method provided by the embodiments of the present disclosure, the following describes the content prediction method using a specific embodiment. In one exemplary embodiment, the system performs a predictive analysis on the content to be published based on the results of massive data analysis, and then proceeds to the next step, such as content publishing or content updating, based on the prediction results. Specifically, the following contents are included:

[0138] refer to Figure 4 , the user uploads the content to be predicted, and selects the corresponding optimization target in the operation box to perform the prediction processing before the content is released. After the content is uploaded successfully and the system completes the prediction, the content rating, content quality, conversion effect, similarity and other prediction dimension information corresponding to the content to be predicted can be viewed in the list on the content prediction interface. Specifically, after completing the content prediction, the user can push the content to be predicted to the content library to publish the content. After the content to be predicted has been uploaded for a period of time, if the user needs to re-determine the prediction information of the content to be predicted before the content is released, the user can click the re-detection button on the content prediction interface, and the system will again determine the prediction information of the content to be predicted based on the real-time information in the database. In addition, the user clicks the detailed information button on the content prediction interface to view further prediction information of the content to be predicted, such as content quality information, content conversion information, similarity information, associated high-quality content, etc.

[0139] refer to Figure 5 The user provides the content to be predicted and enters the content prediction tool to perform content prediction. The user selects the target industry for the content and performs predictions based on both conversion and content direction. The conversion direction prediction process is as follows: The user sets the content release target, the recommended account set, and the content release value. Based on this information, the system performs content conversion predictions, such as predicting the click probability of the content, ultimately producing the first output: content conversion information. Furthermore, the system also provides a recommended account set and content release value based on the user's set recommended account set and content release value, corresponding to the system's third output. The content direction prediction process is as follows: The content understanding model obtains content tag information and content perception information for the predicted content. Based on this information, the system determines the probability of the predicted content being broadcasted. Furthermore, the system determines the strengths, weaknesses, and improvement areas for the predicted content, corresponding to the system's second output. After content prediction is complete, the system provides users with high-quality content for comparison and selection, and offers a method for quickly pushing it to the content library. Finally, the user navigates to the content publishing platform, creates a content release information, and initiates content publishing.

[0140] The above-mentioned content prediction method uses a content prediction tool to perform prediction processing on the content to be predicted, which is beneficial to improving the content publishing effect and the accuracy of determining the content prediction information.

[0141] It should be understood that, although the various steps in the flowcharts involved in the various embodiments described above are displayed in sequence according to the instructions of the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and these steps can be executed in other orders. Moreover, at least a portion of the steps in the flowcharts involved in the various embodiments described above can include multiple steps or multiple stages, and these steps or stages are not necessarily executed and completed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a portion of steps or stages in other steps.

[0142] It can be understood that the same / similar parts between the various embodiments of the above method in this specification can be referred to each other, and each embodiment focuses on the differences from other embodiments. For related parts, please refer to the description of other method embodiments.

[0143] Based on the same inventive concept, an embodiment of the present disclosure further provides a content prediction device for implementing the above-mentioned content prediction method.

[0144] Figure 6 FIG. 1 is a block diagram of a content prediction device according to an exemplary embodiment. Figure 6 , the device includes a content acquisition unit 610, an information acquisition unit 620 and a content prediction unit 630.

[0145] The content acquisition unit 610 is configured to acquire content to be predicted.

[0146] The information acquisition unit 620 is configured to acquire initial release information for the content to be predicted; the initial release information is used to represent preset associated information for releasing the content to be predicted.

[0147] The content prediction unit 630 is configured to input the content to be predicted and the initial release information into a pre-trained content prediction model to obtain prediction information of the content to be predicted; the prediction information includes at least content quality information and content conversion information.

[0148] In an exemplary embodiment, the content prediction unit 630 is further configured to input the content to be predicted and the initial release information into a pre-trained content prediction model, perform content conversion prediction processing on the content to be predicted and the initial release information through the content prediction model, and obtain the content operation probability of the content to be predicted; and determine the content conversion information based on the content operation probability of the content to be predicted.

[0149] In an exemplary embodiment, the content prediction unit 630 is further configured to input the content to be predicted and the initial release information into a pre-trained content prediction model, perform content quality prediction processing on the content to be predicted through the content prediction model, and obtain quality information of the content to be predicted in each preset content quality dimension; determine the content quality information based on the quality information of the content to be predicted in each preset content quality dimension.

[0150] In an exemplary embodiment, the initial publishing information includes an initial content publishing value and an initial recommended account set, and the prediction information also includes a target content publishing value and a target recommended account set;

[0151] The content prediction device also includes an information updating unit configured to update the initial content publishing value and the initial recommended account set through the content prediction model to obtain the target content publishing value and the target recommended account set.

[0152] In an exemplary embodiment, the prediction information further includes a first quality label and a second quality label, wherein the first quality label is a label that triggers an increase in the content quality score corresponding to the content quality information, and the second quality label is a label that triggers a decrease in the content quality score corresponding to the content quality information;

[0153] The content prediction device also includes a content update unit configured to generate content update information based on the first quality label and the second quality label when the prediction information does not meet the preset conditions; and update the predicted content based on the content update information.

[0154] In an exemplary embodiment, the prediction information further includes at least one of related content of the content to be predicted, a content classification label, and a content understanding label;

[0155] The content prediction device also includes a first publishing unit, which is configured to execute, when the prediction information does not meet the preset conditions, based on the content classification label and the content understanding label, to filter out from the preset content library the content whose content similarity with the content to be predicted meets the first condition and whose content quality score corresponding to the content quality information meets the second condition, as the associated content of the content to be predicted; filter out from the associated content the target content whose content quality score meets the third condition; and publish the target content according to the content publishing information associated with the target content.

[0156] In an exemplary embodiment, the content prediction device also includes a second publishing unit, which is configured to determine content publishing information associated with the content to be predicted based on the prediction information when the prediction information meets a preset condition; and publish the content to be predicted according to the content publishing information associated with the content to be predicted.

[0157] In an exemplary embodiment, the content prediction device also includes a re-prediction unit, which is configured to execute a re-prediction instruction in response to the content to be predicted, input the content to be predicted and the initial release information into an updated content prediction model, and obtain updated prediction information of the content to be predicted; the updated content prediction model is obtained by re-training the pre-trained content prediction model.

[0158] Regarding the apparatus in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0159] Each module in the above-mentioned content prediction device can be implemented in whole or in part through software, hardware, or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in the computer device in the form of software, so that the processor can call and execute the corresponding operations of each module.

[0160] Figure 7 FIG2 is a block diagram of an electronic device 700 for implementing a content prediction method according to an exemplary embodiment. For example, the electronic device 700 may be a mobile phone, a computer, a digital broadcast terminal, a messaging device, a game console, a tablet device, a medical device, a fitness device, a personal digital assistant, etc.

[0161] Reference Figure 7 , the electronic device 700 may include one or more of the following components: a processing component 702 , a memory 704 , a power supply component 706 , a multimedia component 708 , an audio component 710 , an input / output (I / O) interface 712 , a sensor component 714 , and a communication component 716 .

[0162] The processing component 702 generally controls the overall operation of the electronic device 700, such as operations associated with display, phone calls, data communications, camera operation, and recording operations. The processing component 702 may include one or more processors 720 to execute instructions to perform all or part of the steps of the above-described method. In addition, the processing component 702 may include one or more modules to facilitate interaction between the processing component 702 and other components. For example, the processing component 702 may include a multimedia module to facilitate interaction between the multimedia component 708 and the processing component 702.

[0163] The memory 704 is configured to store various types of data to support operations on the electronic device 700. Examples of such data include instructions for any application or method operating on the electronic device 700, contact data, phone book data, messages, pictures, videos, etc. The memory 704 can be implemented by any type of volatile or non-volatile storage device, or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk, optical disk, or graphene memory.

[0164] The power supply component 706 provides power to the various components of the electronic device 700. The power supply component 706 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power to the electronic device 700.

[0165] The multimedia component 708 includes a screen that provides an output interface between the electronic device 700 and the user. In some embodiments, the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen can be implemented as a touch screen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touch, slide, and gestures on the touch panel. The touch sensor can not only sense the boundaries of the touch or slide action, but also detect the duration and pressure associated with the touch or slide operation. In some embodiments, the multimedia component 708 includes a front camera and / or a rear camera. When the electronic device 700 is in an operating mode, such as a shooting mode or a video mode, the front camera and / or the rear camera can receive external multimedia data. Each front camera and rear camera can be a fixed optical lens system or have a focal length and optical zoom capability.

[0166] The audio component 710 is configured to output and / or input audio signals. For example, the audio component 710 includes a microphone (MIC), which is configured to receive external audio signals when the electronic device 700 is in an operating mode, such as a call mode, a recording mode, and a voice recognition mode. The received audio signal can be further stored in the memory 704 or transmitted via the communication component 716. In some embodiments, the audio component 710 also includes a speaker for outputting audio signals.

[0167] I / O interface 712 provides an interface between processing component 702 and peripheral interface modules, such as a keyboard, click wheel, buttons, etc. These buttons may include but are not limited to: a home button, volume buttons, a start button, and a lock button.

[0168] The sensor assembly 714 includes one or more sensors for providing various aspects of status assessment for the electronic device 700. For example, the sensor assembly 714 can detect the open / closed state of the electronic device 700, the relative positioning of components, such as the display and keypad of the electronic device 700. The sensor assembly 714 can also detect changes in the position of the electronic device 700 or components of the electronic device 700, the presence or absence of user contact with the electronic device 700, the orientation or acceleration / deceleration of the electronic device 700, and temperature changes of the electronic device 700. The sensor assembly 714 may include a proximity sensor configured to detect the presence of nearby objects without any physical contact. The sensor assembly 714 may also include a light sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, the sensor assembly 714 may also include an accelerometer, a gyroscope sensor, a magnetic sensor, a pressure sensor, or a temperature sensor.

[0169] The communication component 716 is configured to facilitate wired or wireless communication between the electronic device 700 and other devices. The electronic device 700 can access a wireless network based on a communication standard, such as WiFi, an operator network (such as 2G, 3G, 4G or 5G), or a combination thereof. In an exemplary embodiment, the communication component 716 receives a broadcast signal or broadcast-related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component 716 also includes a near field communication (NFC) module to facilitate short-range communication. For example, the NFC module can be implemented based on radio frequency identification (RFID) technology, infrared data association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology and other technologies.

[0170] In an exemplary embodiment, the electronic device 700 may be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the above methods.

[0171] In an exemplary embodiment, a computer-readable storage medium including instructions is also provided, such as a memory 704 including instructions, and the instructions can be executed by the processor 720 of the electronic device 700 to perform the above method. For example, the computer-readable storage medium can be a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, etc.

[0172] In an exemplary embodiment, a computer program product is further provided. The computer program product includes instructions, and the instructions can be executed by the processor 720 of the electronic device 700 to implement the above method.

[0173] It should be noted that the above-mentioned devices, electronic devices, computer-readable storage media, computer program products, etc. can also include other implementation methods according to the description of the method embodiments. The specific implementation methods can refer to the description of the relevant method embodiments and will not be described one by one here.

[0174] Other embodiments of the present disclosure will readily occur to those skilled in the art after considering the specification and practicing the invention disclosed herein. This disclosure is intended to cover any variations, uses, or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or customary techniques in the art not disclosed herein. The description and examples are to be considered as exemplary only, with the true scope and spirit of the present disclosure being indicated by the claims.

[0175] It should be understood that the present disclosure is not limited to the exact structures that have been described above and shown in the drawings, and that various modifications and changes can be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.

Claims

1. A content prediction method, characterized in that: include: Obtain the content to be predicted; The content to be predicted is used to represent unpublished content; Obtaining initial release information for the content to be predicted; The initial release information is used to represent preset associated information for releasing the content to be predicted; The initial publishing information includes at least one of a content publishing value, a recommended account set, and a content publishing target; The recommendation account set is used to represent a set consisting of multiple recommendation objects of the content to be predicted; Inputting the content to be predicted and the initial release information into a pre-trained content prediction model to obtain prediction information of the content to be predicted; the prediction information includes at least content quality information and content conversion information; the content quality information is used to measure the quality of the content to be predicted itself; If the content quality score corresponding to the content quality information is less than or equal to the content quality score threshold, or if the content conversion score corresponding to the content conversion information is less than or equal to the content conversion score threshold, generating content update information based on the first quality tag that triggers an increase in the content quality score and the second quality tag that triggers a decrease in the content quality score in the prediction information; The content update information is used to represent suggestion information for updating the content to be predicted; Before publishing, the content to be predicted is updated according to the content update information to obtain updated content.

2. The content prediction method according to claim 1, characterized in that: The step of inputting the content to be predicted and the initial release information into a pre-trained content prediction model to obtain prediction information of the content to be predicted includes: Inputting the content to be predicted and the initial release information into a pre-trained content prediction model, performing content conversion prediction processing on the content to be predicted and the initial release information through the content prediction model to obtain a content operation probability of the content to be predicted; The content conversion information is determined according to the content operation probability of the content to be predicted.

3. The content prediction method according to claim 2, characterized in that: The step of inputting the content to be predicted and the initial release information into a pre-trained content prediction model to obtain prediction information of the content to be predicted further includes: Inputting the content to be predicted and the initial release information into a pre-trained content prediction model, performing content quality prediction processing on the content to be predicted using the content prediction model to obtain quality information of the content to be predicted in various preset content quality dimensions; The content quality information is determined according to the quality information of the content to be predicted in each preset content quality dimension.

4. The content prediction method according to claim 3, characterized in that: The initial release information includes an initial content release value and an initial recommended account set, and the prediction information also includes a target content release value and a target recommended account set; The method further comprises: The initial content publishing value and the initial recommended account set are updated through the content prediction model to obtain the target content publishing value and the target recommended account set.

5. The content prediction method according to claim 1, characterized in that: The prediction information also includes at least one of the associated content, content classification label, and content understanding label of the content to be predicted; and the method further includes: If the prediction information does not meet the preset conditions, screening, based on the content classification label and the content understanding label, from a preset content library, content whose content similarity with the content to be predicted meets a first condition and whose content quality score corresponding to the content quality information meets a second condition, as associated content of the content to be predicted; Filtering target content whose content quality score meets a third condition from the associated content; The target content is published according to the content publishing information associated with the target content.

6. The content prediction method according to claim 5, characterized in that: The method further comprises: In a case where the prediction information satisfies the preset condition, determining content publishing information associated with the content to be predicted according to the prediction information; The content to be predicted is published according to the content publishing information associated with the content to be predicted.

7. The content prediction method according to claim 1, characterized in that: After inputting the content to be predicted and the initial release information into a pre-trained content prediction model to obtain prediction information of the content to be predicted, the method further includes: In response to a re-prediction instruction for the content to be predicted, the content to be predicted and the initial release information are input into an updated content prediction model to obtain updated prediction information of the content to be predicted; the updated content prediction model is obtained by retraining the pre-trained content prediction model.

8. A content prediction device, characterized in that: include: A content acquisition unit, configured to acquire content to be predicted; The content to be predicted is used to represent unpublished content; an information acquisition unit, configured to acquire initial release information for the content to be predicted; The initial release information is used to represent preset associated information for releasing the content to be predicted; The initial publishing information includes at least one of a content publishing value, a recommended account set, and a content publishing target; The recommendation account set is used to represent a set consisting of multiple recommendation objects of the content to be predicted; a content prediction unit configured to input the content to be predicted and the initial release information into a pre-trained content prediction model to obtain prediction information of the content to be predicted; the prediction information includes at least content quality information and content conversion information; the content quality information is used to measure the quality of the content to be predicted; a content update unit configured to, when a content quality score corresponding to the content quality information is less than or equal to a content quality score threshold, or when a content conversion score corresponding to the content conversion information is less than or equal to a content conversion score threshold, generate content update information based on a first quality tag that triggers an increase in the content quality score and a second quality tag that triggers a decrease in the content quality score in the prediction information; The content update information is used to represent suggestion information for updating the content to be predicted; Before publishing, the content to be predicted is updated according to the content update information to obtain updated content.

9. The content prediction device according to claim 8, characterized in that The content prediction unit is further configured to input the content to be predicted and the initial release information into a pre-trained content prediction model, perform content conversion prediction processing on the content to be predicted and the initial release information through the content prediction model, and obtain the content operation probability of the content to be predicted; and determine the content conversion information based on the content operation probability of the content to be predicted.

10. The content prediction device according to claim 9, characterized in that The content prediction unit is further configured to input the content to be predicted and the initial release information into a pre-trained content prediction model, perform content quality prediction processing on the content to be predicted through the content prediction model, and obtain quality information of the content to be predicted in each preset content quality dimension; and determine the content quality information based on the quality information of the content to be predicted in each preset content quality dimension.

11. The content prediction device according to claim 10, wherein: The initial release information includes an initial content release value and an initial recommended account set, and the prediction information also includes a target content release value and a target recommended account set; The apparatus further includes an information updating unit configured to update the initial content publishing value and the initial recommended account set using the content prediction model to obtain the target content publishing value and the target recommended account set.

12. The content prediction device according to claim 8, wherein: The prediction information also includes at least one of the related content, content classification label, and content understanding label of the content to be predicted; The device also includes a first publishing unit, which is configured to execute, when the prediction information does not meet the preset conditions, based on the content classification label and the content understanding label, to filter out from a preset content library content whose content similarity with the content to be predicted meets a first condition and whose content quality score corresponding to the content quality information meets a second condition, as associated content of the content to be predicted; filter out from the associated content target content whose content quality score meets a third condition; and publish the target content according to the content publishing information associated with the target content.

13. The content prediction device according to claim 12, characterized in that: The device further includes a second publishing unit configured to determine content publishing information associated with the content to be predicted based on the prediction information if the prediction information meets the preset condition; and publish the content to be predicted according to the content publishing information associated with the content to be predicted.

14. The content prediction device according to claim 8, wherein The device also includes a re-prediction unit, which is configured to execute a re-prediction instruction in response to the content to be predicted, input the content to be predicted and the initial release information into an updated content prediction model, and obtain updated prediction information of the content to be predicted; the updated content prediction model is obtained by retraining the pre-trained content prediction model.

15. An electronic device, characterized in that: include: processor; a memory for storing instructions executable by the processor; The processor is configured to execute the instructions to implement the content prediction method according to any one of claims 1 to 7.

16. A computer-readable storage medium, characterized in that When the instructions in the computer-readable storage medium are executed by a processor of an electronic device, the electronic device is enabled to perform the content prediction method according to any one of claims 1 to 7.

17. A computer program product comprising instructions, characterized in that: When the instructions are executed by a processor of an electronic device, the electronic device is enabled to perform the content prediction method according to any one of claims 1 to 7.

Images