Display information delivery method and device, electronic equipment and storage medium

Abstract

The embodiment of the invention relates to a display information delivery method and device, electronic equipment and a storage medium. The method comprises the following steps: acquiring scene data and basic data corresponding to a target user; performing display value prediction on the scene data and the basic data by using a pre-trained value prediction model to obtain a display value score of the target user; and if the display value score meets the preset display condition, sending target display information to a terminal used by the target user. According to the embodiment of the invention, the scene related features are fully extracted from the scene data, and the model extracts richer features from data of more dimensions, so that the precision of predicting the display value of the user is improved, the display information can be more accurately put to the target user, and in addition, the user experience is improved. In addition, the value prediction model provided by the embodiment of the invention can also analyze scene data in different display scenes, the model can adapt to various display scenes, and the adaptability of the model is improved.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a method, device, electronic device, and storage medium for displaying information. Background Art

[0002] Typically, accurate user matching is crucial for information delivery. Current mainstream information delivery technologies use deep learning to find optimal results, thereby determining which users are the target audience for information delivery.

[0003] However, information delivery faces significant differences across different delivery scenarios. For example, when delivering information on different platforms, user groups, scenario traffic, and other factors naturally differ. Furthermore, from the perspective of the information distributor, delivery objectives are often multi-dimensional, such as revenue, engagement rate, and play duration. Therefore, simply determining target users based on personal data presents challenges such as insufficient data diversity and low model prediction accuracy. Therefore, how to simultaneously incorporate both user and scenario characteristics into a single model to improve model prediction accuracy is a pressing issue. Summary of the Invention

[0004] In view of this, in order to solve some or all of the above technical problems, the embodiments of the present application provide a method, device, electronic device and computer-readable storage medium for display information delivery.

[0005] In a first aspect, an embodiment of the present application provides a method for delivering display information, the method comprising: obtaining scenario data and basic data corresponding to a target user, wherein the scenario data represents data recorded for multiple users viewing display information in a target display scenario, and the basic data represents data recorded for the target user viewing display information; using a pre-trained value prediction model, performing display value prediction on the scenario data and the basic data to obtain a display value score for the target user, wherein the display value score represents the probability of the target user performing a preset operation after viewing the target display information; if the display value score meets the preset display conditions, sending the target display information to the terminal used by the target user.

[0006] In one possible implementation, the scenario data includes scenario basic data and scenario bidding data, wherein the scenario basic data represents data recorded for the behavior of multiple users viewing display information in a target display scenario, and the scenario bidding data represents the cost paid for users to view display information in the target display scenario.

[0007] In one possible implementation, a pre-trained value prediction model is used to predict the display value of the scene data and the basic data to obtain a display value score of the target user, including: using the value prediction model to predict the display value of the scene basic data, the scene bid data and the basic data to obtain a first display value score; obtaining preset initial scene basic data; using the value prediction model to predict the display value of the initial scene basic data, the scene bid data and the basic data to obtain a second display value score; and determining the display value score of the target user based on the difference between the first display value score and the second display value score.

[0008] In one possible implementation, a pre-trained value prediction model is used to perform display value prediction on scene data and basic data to obtain a display value score for the target user, including: utilizing a multi-task shared network included in the value prediction model to perform feature extraction on the basic data to obtain multi-task basic features; utilizing a scene feature extraction network included in the value prediction model to perform feature extraction on the scene data to obtain scene features; utilizing each of at least two value prediction networks included in the value prediction model to perform value prediction on the multi-task basic features and scene features to obtain at least two display value prediction values; and fusing the at least two display value prediction values to obtain a display value score.

[0009] In one possible implementation, the scene feature extraction network included in the value prediction model is used to extract features from the scene data to obtain scene features, including: merging basic data and scene data to obtain merged data; and inputting the merged data into the scene feature extraction network to obtain scene features.

[0010] In one possible embodiment, at least two value prediction networks include a revenue prediction network and a usage time prediction network; using each of the at least two value prediction networks included in the value prediction model, value prediction is performed on multi-task basic features and scene features to obtain at least two display value prediction values, including: using the revenue prediction network to predict the revenue value of multi-task basic features and scene features, and obtaining a revenue prediction value as the display value prediction value corresponding to the revenue prediction network, wherein the revenue prediction value represents the size of the revenue obtained from the target user by using the target display information within a first preset time period in the future; using the usage time prediction network to predict the usage time of the multi-task basic features and scene features to obtain the predicted usage time of the target user, wherein the predicted usage time represents the length of time the target user uses the object indicated by the target display information within a second preset time period in the future.

[0011] In one possible embodiment, target display information is sent to a terminal used by a target user, including: obtaining a preset candidate scene bid data sequence, and obtaining scene basic data; for each candidate scene bid data in the candidate scene bid data sequence, merging the candidate scene bid data and the scene basic data to obtain candidate scene data; using a value prediction model, performing display value prediction on the candidate scene data and the basic data to obtain a candidate revenue prediction value output by a revenue prediction network; determining a target revenue prediction value that meets a preset revenue condition from the obtained candidate revenue prediction value sequence, wherein the candidate revenue prediction values in the candidate revenue prediction value sequence correspond one-to-one to the candidate scene bid data in the candidate scene bid data sequence; determining the expenditure cost of the target display information based on the target candidate scene bid data corresponding to the target candidate revenue prediction value; and sending the target display information to the terminal used by the target user based on the expenditure cost.

[0012] In one possible implementation, a target revenue prediction value that meets preset revenue conditions is determined from the obtained candidate revenue prediction value sequence, including: for each candidate revenue prediction value in the candidate revenue prediction value sequence, if the candidate revenue prediction value is not the first candidate revenue prediction value, determining the revenue increase corresponding to the candidate revenue prediction value, wherein the revenue increase is the growth value between two adjacent candidate revenue prediction values; and from each obtained revenue increase, determining the candidate revenue prediction value corresponding to the maximum revenue increase as the target candidate revenue prediction value.

[0013] In a second aspect, an embodiment of the present application provides a display information delivery device, which includes: an acquisition module for acquiring scene data and basic data corresponding to a target user, wherein the scene data represents data recorded for multiple users viewing display information in a target display scene, and the basic data represents data recorded for the target user viewing display information; a prediction module for using a pre-trained value prediction model to perform display value prediction on the scene data and the basic data to obtain a display value score of the target user, wherein the display value score represents the probability of the target user performing a preset operation after viewing the target display information; a sending module for sending the target display information to the terminal used by the target user if the display value score meets the preset display conditions.

[0014] In a third aspect, an embodiment of the present application provides an electronic device, comprising: a memory for storing a computer program; a processor for executing the computer program stored in the memory, and when the computer program is executed, implementing the method of any embodiment of the method for displaying information in the first aspect of the present application.

[0015] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the method of any embodiment of the method for displaying information in the first aspect is implemented.

[0016] In a fifth aspect, an embodiment of the present application provides a computer program, which includes a computer-readable code. When the computer-readable code runs on a device, the processor in the device implements a method as described in any embodiment of the method for displaying information in the first aspect above.

[0017] The display information delivery method, device, electronic device and computer-readable storage medium provided in the embodiments of the present application obtain the scene data and basic data corresponding to the target user, and use a pre-trained value prediction model to predict the display value of the scene data and basic data to obtain the display value score of the target user. When the display value score meets the preset display conditions, the target display information is sent to the terminal used by the target user. The embodiments of the present application realize the joint analysis of basic data and scene data using the same model, fully extract scene-related features from the scene data, and the model extracts richer features from data of more dimensions, thereby improving the accuracy of predicting the user's display value, which helps to deliver display information to the target user more accurately. In addition, the value prediction model provided in the embodiments of the present application can also analyze scene data under different display scenarios. The model can adapt to various display scenarios, thereby improving the adaptability of the model. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0019] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0020] One or more embodiments are exemplarily illustrated by pictures in the corresponding drawings. These exemplifications do not constitute limitations on the embodiments. Elements with the same reference numerals in the drawings are represented as similar elements. Unless otherwise stated, the figures in the drawings do not constitute proportional limitations.

[0021] Figure 1 A flowchart of a method for delivering display information provided in an embodiment of the present application;

[0022] Figure 2A flowchart of another method for displaying information provided in an embodiment of the present application;

[0023] Figure 3 A flowchart of another method for delivering display information provided in an embodiment of the present application;

[0024] Figure 4 An exemplary architecture diagram of a value prediction model provided in an embodiment of the present application;

[0025] Figure 5 A flowchart of another method for delivering display information provided in an embodiment of the present application;

[0026] Figure 6 An exemplary architecture diagram of another value prediction model provided in an embodiment of the present application;

[0027] Figure 7 An exemplary architecture diagram of another value prediction model provided in an embodiment of the present application;

[0028] Figure 8 A flowchart of another method for delivering display information provided in an embodiment of the present application;

[0029] Figure 9 A flowchart of another method for delivering display information provided in an embodiment of the present application;

[0030] Figure 10 A flowchart of another method for delivering display information provided in an embodiment of the present application;

[0031] Figure 11 A schematic diagram of the structure of a display information delivery device provided in an embodiment of the present application;

[0032] Figure 12 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0033] Various exemplary embodiments of the present application will now be described in detail with reference to the accompanying drawings. It is apparent that the described embodiments are only a portion of the embodiments of the present application, rather than all of the embodiments. It should be noted that, unless otherwise specifically stated, the relative arrangement of components and steps, numerical expressions, and numerical values described in these embodiments do not limit the scope of the present application.

[0034] Those skilled in the art will understand that the terms "first" and "second" in the embodiments of the present application are only used to distinguish between different steps, devices, modules and other objects, and neither represent any specific technical meaning nor indicate the logical order between them.

[0035] It should also be understood that in this embodiment, “a plurality of” may refer to two or more than two, and “at least one” may refer to one, two or more than two.

[0036] It should also be understood that any component, data or structure mentioned in the embodiments of the present application can generally be understood as one or more, unless explicitly limited or otherwise indicated in the context.

[0037] In addition, the term "and / or" in this application is simply a description of the association relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent three situations: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this application generally indicates that the related objects are in an "or" relationship.

[0038] It should also be understood that the description of each embodiment in this application focuses on the differences between the embodiments, and the same or similar aspects can be referenced with each other. For the sake of brevity, they will not be described one by one.

[0039] The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the present disclosure, its application, or uses.

[0040] Technologies, circuits, and devices known to those of ordinary skill in the relevant art may not be discussed in detail, but where appropriate, the above-mentioned technologies, circuits, and devices should be considered part of the specification.

[0041] It should be noted that like reference numerals and letters refer to like items in the following figures, and therefore, once an item is defined in one figure, it need not be further discussed in subsequent figures.

[0042] It should be noted that, unless there is a conflict, the embodiments and features in the embodiments of this application can be combined with each other. To facilitate understanding of the embodiments of this application, the application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments. Obviously, the embodiments described are part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0043] In order to solve the technical problem of low accuracy of information delivery in the existing technology, this application provides a method for display information delivery, which can enable the model to extract richer features from data of more dimensions, thereby delivering display information to target users more accurately.

[0044] Figure 1A flow chart of a method for displaying information provided in an embodiment of the present application. This method can generally be applied to a server, by setting a value prediction model on the server, analyzing the basic data and scenario data, and then pushing display information to the user terminal. In addition, the execution subject of this method can be hardware or software. When the above-mentioned execution subject is hardware, the execution subject can be one or more of the above-mentioned electronic devices. For example, a single electronic device can execute this method, or a plurality of electronic devices can cooperate with each other to execute this method. When the above-mentioned execution subject is software, this method can be implemented as multiple software or software modules, or as a single software or software module. No specific limitation is made here.

[0045] like Figure 1 As shown, the method specifically includes:

[0046] Step 101: Acquire the scene data and basic data corresponding to the target user.

[0047] In some embodiments, the scenario data represents data recorded for multiple users viewing display information in a target display scenario. The target display scenario is a specific information service platform, such as an application or website that provides audio and video playback services, or an application or website that provides document, image, or shopping services.

[0048] In a target display scenario, multiple users may view the displayed information. When users view the information, various data generated by each viewing operation can be recorded to obtain scenario data. For example, information bid count (i.e., the number of times the information was bid to push the information to users), information exposure count (i.e., the number of times the information was displayed to multiple users), information click count (i.e., the number of times multiple users clicked on the pushed display information), and information bid (i.e., the value paid by users to click on the displayed information, pay for it, etc.).

[0049] The basic data described above represents data recorded for target users viewing display information. Optionally, the basic data may include static data (e.g., gender, age, education level, consumption level, etc.), statistical data (e.g., number of clicks in different scenarios, duration of audio and video viewing, amount of payments, etc.), and sequence data (e.g., sequence of video numbers and number of views in the last N days).

[0050] Step 102: Use the pre-trained value prediction model to predict the display value of the scene data and basic data to obtain the display value score of the target user.

[0051] In some embodiments, the display value score output by the value prediction model represents the probability that a target user will perform a preset action after viewing the target display information. Specifically, the higher the display value score, the greater the probability that the user will perform the preset action, and the higher the potential revenue for the provider of the target display information.

[0052] The target display information may include various types of information, such as advertising information, recommendation information, item description information, etc. The preset operations include various types of pre-set operations, such as click operations, purchase operations after clicks, etc.

[0053] The value prediction model is pre-trained using a machine learning algorithm. Specifically, a large number of training samples are collected in advance. Each training sample includes sample scenario data and sample base data obtained for the sample display information, as well as annotated revenue information, user usage time information, and other information. During model training, the sample scenario data and sample base data serve as the model inputs, while the annotated revenue information and user usage time information serve as the model's expected outputs. By adjusting the model parameters, the error between the model's actual output and the expected output is minimized, resulting in a trained value prediction model.

[0054] When performing value prediction, the trained model can integrate scenario data and basic data into the model. For example, the value prediction model can include a scenario feature extraction network that extracts scenario features from the scenario data. Using these features as weights, the model multiplies the scene features with the basic data to generate fused features, which are then used to predict the display value. The value prediction model can output predicted revenue information and predicted user usage time corresponding to the annotated information. This information is then calculated according to a preset value score calculation strategy to generate a display value score.

[0055] Step 103: If the display value score meets the preset display conditions, target display information is sent to the terminal used by the target user.

[0056] In some embodiments, the aforementioned display conditions are pre-set, representing criteria for determining whether the target user can generate revenue for the provider of the target display information. For example, a display value score threshold may be set; if the display value score is greater than or equal to the threshold, the target display information may be pushed to the target user. For another example, the display value scores corresponding to multiple users may be tallied and sorted in descending order of display value scores. If the target user is in the top N users, the target display information may be pushed to the target user.

[0057] The method for delivering display information provided by the embodiment of the present application obtains the scene data and basic data corresponding to the target user, and uses a pre-trained value prediction model to predict the display value of the scene data and basic data, thereby obtaining the display value score of the target user, and sending the target display information to the terminal used by the target user when the display value score meets the preset display conditions. The embodiment of the present application realizes the joint analysis of basic data and scene data using the same model, fully extracts scene-related features from the scene data, and the model extracts richer features from data of more dimensions, thereby improving the accuracy of predicting the display value of the user, and helping to deliver display information to the target user more accurately. In addition, the value prediction model provided by the embodiment of the present application can also analyze scene data under different display scenarios, and the model can adapt to various display scenarios, thereby improving the adaptability of the model.

[0058] In some optional implementations, the scenario data includes scenario basic data and scenario bidding data.

[0059] Scenario-based data refers to data recorded for multiple users viewing displayed information in a target display scenario. For example, in a target display scenario (e.g., an application service platform), information bids (i.e., the number of times an information bid was entered to push information to users), information exposures (i.e., the number of times an information was displayed to multiple users), information clicks (i.e., the number of times multiple users clicked on the pushed display information), information bids (i.e., the value paid to induce users to click on, pay for, etc. the displayed information), and pull-ups (i.e., the number of times a display information was clicked to jump to the target page).

[0060] Scenario bidding data represents the cost of enabling users to view the displayed information in the target display scenario. For example, scenario bidding data includes various types of fees paid by the display information provider, including paid bids (i.e., the fees paid by the display information provider to the media platform after the user views the display information and pays), wake-up bids (i.e., the fees paid by the display information provider to the media platform after the user views and clicks on the display information and jumps to the target page), etc.

[0061] Optionally, the above-mentioned scene basic data and scene bidding data can be directly merged into scene data, or the scene basic data and scene bidding data can be processed to a certain extent (such as filtering out invalid data, normalizing, etc.) and then merged to obtain scene data.

[0062] By acquiring scene basic data and scene bidding data, this embodiment can enable the scene data to represent scene features of more dimensions, thereby helping the model to more fully analyze the scene features during prediction and improve the accuracy of display value score prediction.

[0063] In some optional implementations, such as Figure 2 As shown, step 102 includes:

[0064] Step 1021: Use the value prediction model to predict the display value of the scene basic data, the scene bidding data, and the basic data to obtain a first display value score.

[0065] Specifically, the value prediction model can receive input scene basic data, scene bidding data and basic data, extract features from the three types of data, then fuse the extracted features, and finally predict the display value of the fused features. The first display value score obtained represents the probability that the target user will perform a preset operation after viewing the target display information under the state of real scene basic data.

[0066] Step 1022: Obtain preset initial scene basic data.

[0067] The initial scene basic data can be obtained by setting at least one value included in the scene basic data to an initial value (e.g., 0). For example, the number of competitions, the number of displays, the number of clicks, etc. can be set to 0.

[0068] Step 1023: Use the value prediction model to predict the display value of the initial scene basic data, the scene bid data, and the basic data to obtain a second display value score.

[0069] The value prediction model can receive input initial scene basic data, scene bidding data and basic data, and the output second display value score can represent the probability of the target user performing a preset operation after viewing the target display information under the state of the initial scene basic data.

[0070] Step 1024: Determine the target user's exhibition value score based on the difference between the first exhibition value score and the second exhibition value score.

[0071] The difference between the first and second display value scores represents the display value gain, that is, the growth of the display value score from the initial state, given the real scene basic data. This difference can be used as the display value score for the target user, or it can be converted (for example, normalized or converted to a percentage system) to obtain the display value score.

[0072] The execution process of this embodiment can be implemented based on the S-Learner paradigm, namely, by changing some input variables (in this embodiment, changing the scene base data), the model outputs the predicted values before and after the variables are changed, and a predicted gain value is calculated to represent the impact of the changed variables on the prediction. This embodiment sets the initial scene base data and determines the display value score based on the display value gain in both the initial scene base data and the actual scene base data. This allows the resulting display value score to more accurately match the actual display scene, thereby improving the accuracy of determining the user who receives the display information.

[0073] In some optional implementations, such as Figure 3 As shown, step 102 includes:

[0074] Step 1025 , using the multi-task shared network included in the value prediction model, extract features from the basic data to obtain multi-task basic features.

[0075] like Figure 4 , which shows the architecture of the value prediction model. This model architecture can be implemented based on the MMoE (Multi-gate Mixture-of-Experts) architecture. The multi-task shared network 401 can include a multi-layer deep neural network. The basic data is calculated by the multi-task shared network, allowing multiple networks performing specific prediction tasks to share the extracted multi-task basic features.

[0076] Step 1026: Utilize the scene feature extraction network included in the value prediction model to extract features from the scene data to obtain scene features.

[0077] like Figure 4 As shown, the scene feature extraction network 402 receives input scene data. The scene feature extraction network may include a multi-layer neural network, such as a ReLU activation function layer and a sigmoid activation function layer. The scene feature extraction network can extract scene-related features from the scene data and input the scene features as weights into the main network (i.e., the value prediction network described below), thereby enhancing the main network's ability to differentiate and learn scene-related information.

[0078] Step 1027 , using each of the at least two value prediction networks included in the value prediction model, perform value prediction on the multi-task basic features and the scene features to obtain at least two display value prediction values.

[0079] like Figure 4As shown, at least two value prediction networks 403 (including value prediction network 1-value prediction network N) respectively receive the input multi-task basic features and scenario features, respectively perform corresponding prediction tasks (such as revenue prediction task, play duration prediction task, play frequency prediction task, etc.), and output display value prediction value 1-display value prediction value N. When the value prediction model is implemented based on the MMoE architecture, the at least two value prediction networks are constructed in the form of task towers.

[0080] Step 1028: Fusing at least two display value prediction values to obtain an display value score.

[0081] Specifically, a weight value corresponding to each display value prediction value can be set, and the weighted sum of each display value prediction value can be performed to obtain a display value score; or on the basis of the weighted sum, further calculation can be performed according to a preset numerical conversion strategy (such as normalization, etc.) to obtain a display value score.

[0082] This embodiment incorporates scene features into multi-task prediction by setting a multi-task sharing network, a scene feature extraction network and at least two value prediction networks in the value prediction model, fully mines scene-related features from scene data, enhances the model's differentiated learning ability for scene-related information, and improves the accuracy of display value score prediction.

[0083] In some optional implementations, such as Figure 5 As shown, step 1026 includes:

[0084] Step 10261: merge the basic data and the scene data to obtain merged data.

[0085] The basic data and the scene data can be directly merged to obtain the merged data; or the basic data and the scene data can be processed to a certain extent (such as filtering out invalid data, normalizing, etc.) and then merged to obtain the scene data.

[0086] Step 10262: Input the merged data into the scene feature extraction network to obtain scene features.

[0087] like Figure 6 As shown, the scene feature extraction network receives the merged data and extracts scene features from the merged data.

[0088] This embodiment merges the basic data and the scene data, and then uses the scene feature extraction network to extract scene features from the merged data. This enables the scene feature extraction network to use the basic data as a reference dimension for scene feature extraction, thereby helping to make the extracted scene features more closely match the actual scene and improving the accuracy of value prediction.

[0089] In some optional implementations, such as Figure 7 As shown, the at least two value prediction networks include a revenue prediction network and a usage time prediction network.

[0090] As an example, if the value prediction model is implemented based on the MMoE architecture, the revenue prediction network and the usage time prediction network are implemented in the form of task towers.

[0091] like Figure 8 As shown, step 1027 includes:

[0092] In step 10271, the revenue prediction network is used to predict the revenue value of the multi-task basic features and the scene features, and the revenue prediction value is obtained as the display value prediction value corresponding to the revenue prediction network.

[0093] The revenue forecast value represents the revenue that the advertiser will earn from the target user by using the target display information within a first preset time period (e.g., the next 7 days). For example, if the target display information is an advertisement, the revenue forecast value represents the estimated revenue that the advertiser will earn from the target user by placing the advertisement.

[0094] Step 10272: Use the usage time prediction network to predict the usage time of the multi-task basic features and scene features to obtain the predicted usage time of the target user.

[0095] The predicted usage duration indicates the duration that the target user uses the object indicated by the target display information within a second preset duration in the future (eg, the same day).

[0096] The type of predicted usage duration can correspond to the actual type of the target display information. For example, if the target display information displays information about a video playback website, the predicted usage duration can be the time it takes the target user to access the video playback website and play the video. For another example, if the target display information displays a software download link, the predicted usage duration can be the time it takes the target user to use the software after downloading it.

[0097] like Figure 7 As shown in Figure 2, both the revenue prediction network and the usage time prediction network receive the input multi-task basic features and scene features. Both the revenue prediction network and the usage time prediction network include two layers of sub-networks (DNN Layer1, DNNLayer2). Correspondingly, the scene feature extraction network also includes two layers of sub-networks (Gate nu1, Gate nu2). Each layer of sub-network has Figure 7As shown in the corresponding relationship, the Gate nu1 sub-network simultaneously inputs the output scene features into the revenue prediction network and the usage time prediction network, and serves as the weights for the features output by DNN Layer 1 of the two networks. The Gate nu2 sub-network simultaneously inputs the output scene features into the revenue prediction network and the usage time prediction network, and serves as the weights for the features output by DNN Layer 2 of the two networks.

[0098] like Figure 7 As shown in FIG, the revenue prediction network and the usage time prediction network output the revenue prediction value and the predicted usage time respectively, which are fused and calculated to obtain the display value score.

[0099] This embodiment sets up a revenue prediction network and a usage time prediction network to achieve the use of a value prediction model to predict the revenue and usage time brought to the target user. By using revenue and time, the load generated after the target display information is delivered to the target user can be more accurately represented, thereby further improving the accuracy of the model's value prediction.

[0100] In some optional implementations, such as Figure 9 As shown, step 103 includes:

[0101] Step 1031: Obtain a preset candidate scenario bidding data sequence and obtain scenario basic data.

[0102] The scene basic data represents the data recorded for the behavior of multiple users viewing the display information in the target display scene. The scene bidding data represents the price paid for the user to view the display information in the target display scene. Figure 2 In combination with the corresponding embodiments, Figure 2 The scene basic data in the corresponding embodiment is the same as the scene basic data obtained in this step. For more information about the scene basic data, please refer to the above Figure 2 In addition, if this embodiment is consistent with the above Figure 2 In combination with the corresponding embodiments, Figure 2 The scene bid data in the corresponding embodiment has the same meaning as the candidate scene bid data in the candidate scene bid data sequence obtained in this step. For more information about scene basic data and scene bid data, please refer to the above Figure 2 Corresponding embodiment.

[0103] The above-mentioned candidate scenario bid data sequence can be a numerical sequence set in the order of the price paid from small to large. In one example, if the scenario bid data includes two types of bid data, namely paid bids and wake-up bids, the candidate scenario bid data sequence can include a paid bid sequence and a wake-up bid sequence. The paid bid sequence and the wake-up bid sequence can be set by setting a bid coefficient sequence. For example, the bid coefficient sequence is {0.5, 1, 1.5, 2}, and these coefficients are multiplied by the actual paid bid and wake-up bid respectively to obtain the paid bid sequence and the wake-up bid sequence.

[0104] Step 1032: For each candidate scene bid data in the candidate scene bid data sequence, the candidate scene bid data and the scene basic data are merged to obtain the candidate scene data; using the value prediction model, the candidate scene data and the basic data are displayed with value prediction to obtain the candidate revenue prediction value output by the revenue prediction network.

[0105] This step is performed on each candidate scene bidding data in a loop processing manner, that is, each loop processes one candidate scene bidding data in the sequence to obtain a candidate revenue prediction value. Therefore, the candidate scene bidding data described above is the candidate scene bidding data currently being processed.

[0106] Optionally, the scene basic data and the currently processed candidate scene bid data can be directly merged into candidate scene data, or the scene basic data and candidate scene bid data can be processed to a certain extent (such as filtering out invalid data, normalizing, etc.) and then merged to obtain candidate scene data.

[0107] The process of using the value prediction model to display value prediction is similar to the above Figure 8 The process described in the corresponding embodiment is the same and will not be repeated here.

[0108] Step 1033: Determine a target profit forecast value that meets a preset profit condition from the obtained candidate profit forecast value sequence.

[0109] Among them, the candidate profit prediction values in the candidate profit prediction value sequence correspond one-to-one to the candidate scenario bid data in the candidate scenario bid data sequence.

[0110] The revenue condition can be set according to actual needs. For example, the maximum value in the candidate revenue prediction value sequence can be used as the target candidate revenue prediction value.

[0111] Step 1034 : Determine the expenditure cost of the target display information based on the target candidate scenario bidding data corresponding to the target candidate revenue prediction value.

[0112] The aforementioned cost represents the price paid by the provider of the target display information to the provider of the target display scenario in order to deliver the target display information to the target user. For example, this embodiment can be applied to an RTA (Real-Time API) advertising delivery strategy, where the cost of the target display information represents the competitive value of the advertiser's payment to the advertising display platform.

[0113] The target candidate scene bid data may include at least one cost, and the sum of the costs may be used as the cost of the target display information. For example, if the target candidate scene bid data includes a paid bid and a wake-up bid, the sum of the two is the cost of the target display information.

[0114] Step 1035: Send target presentation information to the terminal used by the target user based on the expenditure cost.

[0115] After determining the cost, the provider of the target display information can pay a fee based on the cost and then send the target display information to the terminal used by the target user. Since the target user is a user participating in the bidding, the target user is more likely to perform the above-mentioned preset operation after viewing the target display information.

[0116] It should be understood that steps 1031 to 1035 require that after the value prediction model is used to make predictions in step 102, the value prediction model must be used again to make multiple value predictions on the candidate scenario bidding data sequence and the scenario basic data, so as to determine the expenditure cost that can bring the maximum benefit, thereby optimizing the expenditure cost.

[0117] This embodiment predicts the benefits corresponding to various expenditure costs by setting a candidate scenario bidding data sequence, and selects the expenditure cost that can bring the maximum benefit, so as to more accurately deliver target display information to target users and improve the targeted delivery of display information.

[0118] In some optional implementations, such as Figure 10 As shown, step 1033 includes:

[0119] Step 10331: For each candidate revenue prediction value in the candidate revenue prediction value sequence, if the candidate revenue prediction value is not the first candidate revenue prediction value, determine the revenue increase rate corresponding to the candidate revenue prediction value.

[0120] The revenue increase is the growth value between two adjacent candidate display value scores.

[0121] As an example, the candidate revenue forecast value sequence is represented as {y1, y2, y3, y4}, then the revenue increase corresponding to y2 is y2-y1, and the revenue increase corresponding to y3 is y3-y2.

[0122] Step 10332: Determine the candidate revenue prediction value corresponding to the maximum revenue increase from the obtained revenue increases as the target candidate revenue prediction value.

[0123] Continuing with the above example, let the candidate scenario bid data sequence be {x1, x2, x3, x4}. If y3 corresponds to the largest revenue increase, y3 is the target candidate revenue prediction value. Based on its corresponding candidate scenario bid data x3, the corresponding expenditure cost can be determined.

[0124] This embodiment determines the target candidate revenue prediction value with the largest revenue increase. Based on the target candidate revenue prediction value, the expenditure cost that maximizes the revenue increase can be obtained. The largest revenue increase can indicate that the target user selected under the expenditure cost has the greatest probability of performing the preset operation, thereby delivering target display information to the target user in a more targeted manner.

[0125] Figure 11 This is a schematic diagram of the structure of a display information delivery device provided in an embodiment of the present application. Specifically, it includes: an acquisition module 1101, which is used to obtain scenario data and basic data corresponding to a target user, wherein the scenario data represents data recorded for multiple users viewing display information in a target display scenario, and the basic data represents data recorded for a target user viewing display information; a prediction module 1102, which is used to use a pre-trained value prediction model to predict the display value of the scenario data and basic data to obtain a display value score for the target user, wherein the display value score represents the probability that the target user will perform a preset operation after viewing the target display information; and a sending module 1103, which is used to send the target display information to the terminal used by the target user if the display value score meets the preset display conditions.

[0126] In some optional implementations, the scenario data includes scenario basic data and scenario bidding data, wherein the scenario basic data represents data recorded for the behavior of multiple users viewing display information in a target display scenario, and the scenario bidding data represents the cost paid for users to view display information in a target display scenario.

[0127] In some optional implementations, the prediction module includes: a first prediction unit, used to use a value prediction model to predict the display value of the scene basic data, the scene bidding data and the basic data to obtain a first display value score; a first acquisition unit, used to obtain preset initial scene basic data; a second prediction unit, used to use a value prediction model to predict the display value of the initial scene basic data, the scene bidding data and the basic data to obtain a second display value score; a first determination unit, used to determine the display value score of the target user based on the difference between the first display value score and the second display value score.

[0128] In some optional implementations, the prediction module includes: a first extraction unit, used to utilize the multi-task shared network included in the value prediction model to perform feature extraction on the basic data to obtain multi-task basic features; a second extraction unit, used to utilize the scene feature extraction network included in the value prediction model to perform feature extraction on the scene data to obtain scene features; a third prediction unit, used to utilize each of the at least two value prediction networks included in the value prediction model to perform value prediction on the multi-task basic features and scene features to obtain at least two display value prediction values; a fusion unit, used to fuse the at least two display value prediction values to obtain a display value score.

[0129] In some optional implementations, the second extraction unit includes: a merging subunit, configured to merge the basic data and the scene data to obtain merged data; and an extraction subunit, configured to input the merged data into a scene feature extraction network to obtain scene features.

[0130] In some optional implementations, at least two value prediction networks include a revenue prediction network and a usage time prediction network; the third prediction unit includes: a first prediction subunit, used to use the revenue prediction network to predict the revenue value of multi-task basic features and scene features, and obtain the revenue prediction value as the display value prediction value corresponding to the revenue prediction network, wherein the revenue prediction value represents the size of the revenue obtained from the target user by using the target display information within a first preset time period in the future; a second prediction subunit, used to use the usage time prediction network to predict the usage time of multi-task basic features and scene features, and obtain the predicted usage time of the target user, wherein the predicted usage time represents the time the target user uses the object indicated by the target display information within a second preset time period in the future.

[0131] In some optional implementations, the sending module includes: a second acquisition unit, used to obtain a preset candidate scene bid data sequence, and obtain scene basic data; a fourth prediction unit, used to merge the candidate scene bid data and the scene basic data for each candidate scene bid data in the candidate scene bid data sequence to obtain candidate scene data; using a value prediction model, the candidate scene data and the basic data are displayed. The value prediction is performed to predict the candidate revenue prediction value output by the revenue prediction network; a second determination unit, used to determine the target revenue prediction value that meets the preset revenue conditions from the obtained candidate revenue prediction value sequence, wherein the candidate revenue prediction values in the candidate revenue prediction value sequence correspond one-to-one to the candidate scene bid data in the candidate scene bid data sequence; a third determination unit, used to determine the expenditure cost of the target display information based on the target candidate scene bid data corresponding to the target candidate revenue prediction value; and a sending unit, used to send the target display information to the terminal used by the target user based on the expenditure cost.

[0132] In some optional implementations, the second determination unit includes: a first determination subunit, used to determine the revenue increase corresponding to each candidate revenue prediction value in the candidate revenue prediction value sequence, if the candidate revenue prediction value is not the first candidate revenue prediction value, wherein the revenue increase is the growth value between two adjacent candidate revenue prediction values; a second determination subunit, used to determine the candidate revenue prediction value corresponding to the maximum revenue increase from the various revenue increases obtained as the target candidate revenue prediction value.

[0133] The display information delivery device provided in this embodiment can be as follows Figure 11 The display information delivery device shown in can execute all the steps of the above display information delivery methods, thereby achieving the technical effects of the above display information delivery methods. Please refer to the above related description for details. For the sake of brevity, it will not be repeated here.

[0134] Figure 12 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. Figure 12 The electronic device 1200 shown includes: at least one processor 1201, a memory 1202, at least one network interface 1204 and another user interface 1203. The various components in the electronic device 1200 are coupled together via a bus system 1205. It is understood that the bus system 1205 is used to achieve connection and communication between these components. In addition to including a data bus, the bus system 1205 also includes a power bus, a control bus, and a status signal bus. However, for the sake of clarity, the bus system 1205 is not described in detail. Figure 12 Various buses are labeled as bus system 1205.

[0135] The user interface 1203 may include a display, a keyboard, or a pointing device (eg, a mouse, a trackball, a touchpad, or a touch screen).

[0136] It is understood that the memory 1202 in the embodiment of the present application can be a volatile memory or a non-volatile memory, or can include both volatile and non-volatile memories. Among them, the non-volatile memory can be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory can be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of RAM are available, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate synchronous DRAM (DDRSDRAM), enhanced synchronous DRAM (ESDRAM), synchronous link DRAM (SLDRAM), and direct RAM bus random access memory (DRRAM). The memory 1202 described herein is intended to include, but is not limited to, these and any other suitable types of memory.

[0137] In some embodiments, the memory 1202 stores the following elements, executable units, or data structures, or a subset thereof, or an extended set thereof: an operating system 12021 and application programs 12022 .

[0138] Among them, the operating system 12021 includes various system programs, such as the framework layer, core library layer, and driver layer, which are used to implement various basic services and handle hardware-based tasks. Application 12022 includes various application programs, such as media players and browsers, which are used to implement various application services. The program that implements the method of the embodiment of the present application can be included in application 12022.

[0139] In this embodiment, by calling a program or instruction stored in the memory 1202, specifically, a program or instruction stored in the application 12022, the processor 1201 is configured to execute the method steps provided in each method embodiment, for example, including:

[0140] Obtain scenario data and basic data corresponding to the target user, where the scenario data represents data recorded for multiple users viewing display information in a target display scenario, and the basic data represents data recorded for the target user viewing display information; use a pre-trained value prediction model to predict the display value of the scenario data and basic data to obtain a display value score for the target user, where the display value score represents the probability that the target user will perform a preset operation after viewing the target display information; if the display value score meets the preset display conditions, send the target display information to the terminal used by the target user.

[0141] The methods disclosed in the above embodiments of the present application can be applied to or implemented by processor 1201. Processor 1201 may be an integrated circuit chip with signal processing capabilities. During implementation, each step of the above method can be completed by hardware integrated logic circuits in processor 1201 or by software instructions. The above processor 1201 may be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. The various methods, steps, and logic block diagrams disclosed in the embodiments of the present application can be implemented or executed. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the methods disclosed in the embodiments of the present application can be directly implemented and executed by a hardware decoding processor, or by a combination of hardware and software units in the decoding processor. The software units can be located in a storage medium mature in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, etc. The storage medium is located in the memory 1202 , and the processor 1201 reads the information in the memory 1202 and completes the steps of the above method in combination with its hardware.

[0142] It is understood that the embodiments described herein may be implemented using hardware, software, firmware, middleware, microcode, or a combination thereof. For hardware implementation, the processing unit may be implemented in 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), general-purpose processors, controllers, microcontrollers, microprocessors, or other electronic units or combinations thereof for performing the above-mentioned functions of the present application.

[0143] For software implementation, the techniques described above can be implemented by a unit that performs the functions described above. The software code can be stored in a memory and executed by a processor. The memory can be implemented in the processor or external to the processor.

[0144] The electronic device provided in this embodiment may be Figure 12 The electronic device shown in can execute all the steps of the above-mentioned display information delivery methods, thereby achieving the technical effects of the above-mentioned display information delivery methods. Please refer to the above related description for details. For the sake of brevity, it will not be repeated here.

[0145] The present application also provides a storage medium (computer-readable storage medium). The storage medium stores one or more programs. The storage medium may include volatile memory, such as random access memory; the memory may also include non-volatile memory, such as read-only memory, flash memory, hard disk, or solid-state drive; and the memory may also include a combination of the aforementioned types of memory.

[0146] When one or more programs in the storage medium can be executed by one or more processors, the above-mentioned method of display information delivery executed on the electronic device side can be implemented.

[0147] The processor is configured to execute a program stored in the memory to implement the following steps of the method for delivering display information executed on the electronic device side:

[0148] Obtain scenario data and basic data corresponding to the target user, where the scenario data represents data recorded for multiple users viewing display information in a target display scenario, and the basic data represents data recorded for the target user viewing display information; use a pre-trained value prediction model to predict the display value of the scenario data and basic data to obtain a display value score for the target user, where the display value score represents the probability that the target user will perform a preset operation after viewing the target display information; if the display value score meets the preset display conditions, send the target display information to the terminal used by the target user.

[0149] Professionals should also be further aware that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the above description has generally described the components and steps of each example according to their functions. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians can use different circuits to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0150] The steps of the circuits or algorithms described in conjunction with the embodiments disclosed herein may be implemented using hardware, a software module executed by a processor, or a combination of the two. The software module may be placed in a random access memory (RAM), a memory, a read-only memory (ROM), an electrically programmable ROM, an electrically erasable programmable ROM, a register, a hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.

[0151] It should be understood that the terms used herein are for the purpose of describing specific example embodiments only and are not intended to be limiting. Unless the context clearly indicates otherwise, the singular forms "one", "an" and "said" as used herein may also be meant to include plural forms. The terms "comprise", "include", "contain" and "have" are inclusive and therefore specify the presence of stated features, steps, operations, elements and / or parts, but do not exclude the presence or addition of one or more other features, steps, operations, elements, parts, and / or combinations thereof. The steps, processes, and operations described herein are not to be construed as necessarily requiring them to be performed in the specific order described or illustrated, unless the order of execution is clearly indicated. It should also be understood that additional or alternative steps may be used.

[0152] The foregoing is merely a list of specific embodiments of the present application, intended to enable those skilled in the art to understand or implement the present application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application is not limited to the embodiments shown herein, but is intended to conform to the broadest scope consistent with the principles and novel features of the present application.

Claims

1. A method for delivering display information, characterized in that: The method comprises: Obtaining scenario data and basic data corresponding to a target user, wherein the scenario data represents data of display information records viewed by multiple users in a target display scenario, and the basic data represents data of display information records viewed by the target user; Using a pre-trained value prediction model, the scene data and the basic data are subjected to display value prediction to obtain a display value score of the target user, wherein the display value score represents the probability of the target user performing a preset operation after viewing the target display information; If the display value score meets the preset display conditions, the target display information is sent to the terminal used by the target user.

2. The method according to claim 1, characterized in that The scenario data includes scenario basic data and scenario bidding data, wherein the scenario basic data represents data recorded for the behavior of multiple users viewing the display information under the target display scenario, and the scenario bidding data represents the cost paid for the user to view the display information under the target display scenario.

3. The method according to claim 2, characterized in that The method of using a pre-trained value prediction model to predict the display value of the scene data and the basic data to obtain the display value score of the target user includes: Using the value prediction model, performing display value prediction on the scene basic data, the scene bid data, and the basic data to obtain a first display value score; Get the preset initial scene basic data; Using the value prediction model, performing display value prediction on the initial scene basic data, the scene bid data, and the basic data to obtain a second display value score; An exhibition value score of the target user is determined based on a difference between the first exhibition value score and the second exhibition value score.

4. The method according to claim 1, wherein The method of using a pre-trained value prediction model to predict the display value of the scene data and the basic data to obtain the display value score of the target user includes: Using the multi-task shared network included in the value prediction model, feature extraction is performed on the basic data to obtain multi-task basic features; Utilizing the scene feature extraction network included in the value prediction model, feature extraction is performed on the scene data to obtain scene features; Using each of the at least two value prediction networks included in the value prediction model, perform value prediction on the multi-task basic features and the scene features to obtain at least two display value prediction values; The at least two display value prediction values are fused to obtain the display value score.

5. The method according to claim 4, characterized in that The method of extracting features from the scene data using the scene feature extraction network included in the value prediction model to obtain scene features includes: Merging the basic data and the scene data to obtain merged data; The combined data is input into the scene feature extraction network to obtain the scene features.

6. The method according to claim 4, characterized in that The at least two value prediction networks include a revenue prediction network and a usage time prediction network; The using each of the at least two value prediction networks included in the value prediction model to perform value prediction on the multi-task basic features and the scene features to obtain at least two display value prediction values includes: Using the revenue prediction network, predicting revenue values for the multi-task basic features and the scenario features, and obtaining a revenue prediction value as a display value prediction value corresponding to the revenue prediction network, wherein the revenue prediction value represents the amount of revenue obtained from the target user by using the target display information within a first preset time period in the future; The usage time prediction network is used to predict the usage time of the multi-task basic features and the scene features to obtain the predicted usage time of the target user, wherein the predicted usage time represents the time the target user uses the object indicated by the target display information within a second preset time period in the future.

7. The method according to claim 6, characterized in that The sending the target display information to the terminal used by the target user includes: Obtaining a preset candidate scene bid data sequence and scene basic data; For each candidate scene bid data in the candidate scene bid data sequence, merging the candidate scene bid data with the scene basic data to obtain candidate scene data; using the value prediction model, performing display value prediction on the candidate scene data and the basic data to obtain a candidate revenue prediction value output by the revenue prediction network; Determining a target profit prediction value that meets a preset profit condition from the obtained candidate profit prediction value sequence, wherein the candidate profit prediction values in the candidate profit prediction value sequence correspond one-to-one to the candidate scenario bid data in the candidate scenario bid data sequence; Determining the expenditure cost of the target display information based on the target candidate scenario bid data corresponding to the target candidate revenue prediction value; Based on the expenditure cost, the target display information is sent to a terminal used by the target user.

8. The method according to claim 7, characterized in that The step of determining a target revenue forecast value that meets a preset revenue condition from the obtained candidate revenue forecast value sequence includes: For each candidate revenue prediction value in the candidate revenue prediction value sequence, if the candidate revenue prediction value is not the first candidate revenue prediction value, determining a revenue increase corresponding to the candidate revenue prediction value, wherein the revenue increase is a growth value between two adjacent candidate revenue prediction values; From the obtained revenue increases, a candidate revenue prediction value corresponding to the maximum revenue increase is determined as the target candidate revenue prediction value.

9. A display information delivery device, characterized in that: The device comprises: An acquisition module, configured to acquire scenario data and basic data corresponding to a target user, wherein the scenario data represents data recorded when multiple users view display information in a target display scenario, and the basic data represents data recorded when the target user views display information; A prediction module, configured to use a pre-trained value prediction model to predict the display value of the scenario data and the basic data, and obtain a display value score of the target user, wherein the display value score represents the probability of the target user performing a preset operation after viewing the target display information; The sending module is configured to send the target display information to a terminal used by the target user if the display value score meets a preset display condition.

10. An electronic device, characterized in that: include: Memory for storing computer programs; The processor is configured to execute the computer program stored in the memory, and when the computer program is executed, the method for delivering display information according to any one of claims 1 to 8 is implemented.

11. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the display information delivery method described in any one of claims 1 to 8 is implemented.

Images