Medium information delivery method, computer equipment and storage medium

Abstract

The invention relates to a medium information delivery method, computer equipment and a storage medium. The method comprises the steps of obtaining multiple pieces of to-be-delivered medium information corresponding to different display modes in response to a medium information delivery request for a target account sent by a terminal of the target account; obtaining a first conversion virtual resource value of each piece of to-be-released medium information, obtaining a medium information exposure rate of each piece of to-be-released medium information for the target account, and obtaining a second conversion virtual resource value of each piece of to-be-released medium information for the target account based on the first conversion virtual resource value and the medium information exposure rate; and based on the second conversion virtual resource value, obtaining target medium information used for being put to a terminal of the target account for display. By adopting the method, the release of various kinds of media information can be realized by considering the exposure rates of the media information in different display modes for the specific target account, so that the individual difference of users is considered, and the accuracy degree of the release of the media information is improved.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a media information delivery method, computer equipment, storage medium, and computer program product. Background Art

[0002] With the development of computer technology, a technology for displaying media information on terminals has emerged. When users browse information through terminals, media information can be added to the information browsing page of the user terminal, thereby increasing the exposure rate of the media information.

[0003] In related technologies, media information can be broadly divided into two categories: media information that requires active user interaction before it can be displayed, and media information that can be displayed directly without user interaction. When delivering the above-mentioned multiple types of media information, it is usually adopted to display the priority types first and then provide a backup for the latter types, or to implement delivery by adopting a fixed type method after comparing the effects based on AB experiments.

[0004] However, the above-mentioned media information delivery method usually determines the media information to be displayed on the user terminal based on group indicators. This method ignores the differences between individual users, so the existing media information delivery is not accurate enough. Summary of the Invention

[0005] Based on this, it is necessary to provide a media information delivery method, device, computer equipment, computer-readable storage medium and computer program product that can improve the accuracy of media information delivery in response to the above technical problems.

[0006] In a first aspect, the present application provides a method for delivering media information, comprising:

[0007] In response to a media information delivery request for the target account sent by a terminal of the target account, a plurality of media information to be delivered corresponding to a plurality of different display modes is obtained; wherein each piece of the media information to be delivered corresponds to a different display mode;

[0008] Obtaining a first conversion virtual resource value for each of the multiple pieces of media information to be placed, obtaining a media information exposure rate for each of the multiple pieces of media information to be placed for the target account, and obtaining a second conversion virtual resource value for each of the pieces of media information to be placed for the target account based on the first conversion virtual resource value and the media information exposure rate of each piece of media information to be placed;

[0009] Based on the second converted virtual resource value of each piece of the media information to be delivered, target media information for delivery to the terminal of the target account for display is obtained from the multiple pieces of media information to be delivered.

[0010] In one embodiment, the media information to be placed includes: a first type of media information to be placed that is displayed on the terminal of any account only in response to a trigger operation performed by any account on a media information display entrance provided by any account to a terminal of any account; obtaining the media information exposure rate of each of the multiple media information to be placed for the target account includes: obtaining the predicted trigger rate of the media information display entrance provided by the target account to the terminal of the target account; and using the predicted trigger rate as the media information exposure rate of the first type of media information to be placed for the target account.

[0011] In one embodiment, obtaining the predicted trigger rate of the media information display entrance provided by the target account to the terminal of the target account includes: splicing the portrait features, historical behavior features and real-time behavior features of the target account to obtain the spliced ​​account features of the target account; inputting the spliced ​​account features of the target account into a pre-built display entrance trigger prediction model, and obtaining the predicted trigger rate of the media information display entrance provided by the target account to the terminal of the target account through the display entrance trigger prediction model.

[0012] In one embodiment, before inputting the spliced ​​account features of the target account into a pre-built display entry trigger prediction model, it also includes: obtaining sample account portrait features, sample historical behavior features, and sample real-time behavior features of the sample account; obtaining the historical trigger results of the sample account for the information display entry for displaying the first category of media information to be released; splicing the sample account portrait features, sample historical behavior features, and sample real-time behavior features to obtain the spliced ​​account features of the sample account, and using the spliced ​​account features of the sample account and the historical trigger results to train a classification model to obtain the display entry trigger prediction model.

[0013] In one embodiment, the media information to be placed includes: a second type of media information to be placed that is displayed directly on the terminal of any account without responding to a trigger operation performed by any account on a media information display entrance provided by the terminal of any account; obtaining the media information exposure rate of each of the multiple media information to be placed for the target account includes: obtaining the pre-set media information exposure rate for the second type of media information to be placed; and using the pre-set media information exposure rate as the media information exposure rate of the second type of media information to be placed for the target account.

[0014] In one embodiment, the obtaining of multiple pieces of media information to be placed corresponding to multiple different display modes includes: obtaining multiple pieces of candidate media information corresponding to each of the display modes, and obtaining the first conversion virtual resource value of each of the multiple pieces of candidate media information; respectively determining the maximum value of the first conversion virtual resource values ​​of the multiple pieces of candidate media information corresponding to each of the display modes, and using the candidate media information corresponding to each of the maximum values ​​as the media information to be placed for each of the display modes.

[0015] In one embodiment, based on the second conversion virtual resource value of each piece of the media information to be delivered, target media information for delivery to the terminal of the target account for display is obtained from the multiple pieces of media information to be delivered, including: obtaining the maximum value of the second conversion virtual resource value of each piece of the media information to be delivered; and using the media information to be delivered corresponding to the maximum value of the second conversion virtual resource value as the target media information for delivery to the terminal of the target account for display.

[0016] In a second aspect, the present application further provides a media information delivery device, comprising:

[0017] A first resource acquisition module is configured to, in response to a media information delivery request for a target account sent by a terminal of the target account, acquire a plurality of pieces of to-be-delivered media information corresponding to a plurality of different display modes; wherein each piece of the to-be-delivered media information corresponds to a different display mode;

[0018] a second resource acquisition module, configured to acquire a first conversion virtual resource value of each of the plurality of media information to be placed, acquire a media information exposure rate of each of the plurality of media information to be placed for the target account, and obtain a second conversion virtual resource value of each of the media information to be placed for the target account based on the first conversion virtual resource value and the media information exposure rate of each of the media information to be placed;

[0019] The media information delivery module is configured to obtain target media information for delivery to the terminal of the target account for display from the plurality of media information to be delivered based on the second converted virtual resource value of each piece of media information to be delivered.

[0020] In a third aspect, the present application further provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the method described in any one of the embodiments of the first aspect when executing the computer program.

[0021] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described in any one of the embodiments of the first aspect.

[0022] In a fifth aspect, the present application also provides a computer program product, comprising a computer program, which, when executed by a processor, implements the steps of the method described in any one of the embodiments of the first aspect.

[0023] The above-mentioned media information delivery method, device, computer equipment, storage medium and computer program product, when receiving a media information delivery request for a target account, collects the media information exposure rate of each media information to be delivered for the target account, so that the first conversion virtual resource value of the media information to be delivered and the media information exposure rate of the media information to be delivered for the target account can be used to obtain the second conversion virtual resource value of each media information to be delivered for the target account, and then use the second conversion virtual resource value to screen the target media information delivery. This method takes into account the exposure rate of media information with different display methods for a specific target account to achieve the delivery of multiple types of media information. Since the media information exposure rates of different target accounts represent the different browsing characteristics of different user individuals for media information, this method takes into account the differences between individual users, thereby improving the accuracy of media information delivery. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the following briefly introduces the drawings required for use in the embodiments or related technical descriptions. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0025] Figure 1 A diagram of an application environment of a media information delivery method according to an embodiment;

[0026] Figure 2 1 is a flow chart of a method for delivering media information in one embodiment;

[0027] Figure 3 A schematic diagram of a process for obtaining a predicted trigger rate in one embodiment;

[0028] Figure 4 A schematic diagram of a process for training a display entry trigger prediction model in one embodiment;

[0029] Figure 5 A schematic diagram of a process for obtaining information of media to be delivered in different display modes in one embodiment;

[0030] Figure 6 A schematic diagram of page interaction for triggering display of different types of advertisements in one embodiment;

[0031] Figure 7 A schematic diagram of page interaction for triggering display of different types of advertisements in another embodiment;

[0032] Figure 8 1 is a flow chart of a method for dynamically delivering multiple types of advertisements in one embodiment;

[0033] Figure 9 A schematic diagram of a process for constructing an entry click rate prediction model in one embodiment;

[0034] Figure 10 is a structural block diagram of a media information delivery device in one embodiment;

[0035] Figure 11 FIG. 1 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0036] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0037] The media information delivery method provided in the embodiment of the present application can be applied to Figure 1The application environment shown. The terminal 101 communicates with the server 102 via a network. Specifically, the terminal 101 of the target account can initiate a media information delivery request to the server 102 for delivering media information to the target account. After receiving the request, the server 102 can respond to the request, thereby collecting multiple pieces of media information to be delivered corresponding to different display methods, as well as the first conversion virtual resource value of each piece of media information to be delivered. The media information exposure rate of each piece of media information to be delivered to the target account can then be obtained. The first conversion virtual resource value and the media information exposure rate can then be used to calculate the second conversion virtual resource value of each piece of media information to be delivered to the target account. The second conversion virtual resource value can then be used to determine the target media information. The target media information can be delivered to the terminal 101 of the target account for display on the terminal 101. The terminal 101 can include, but is not limited to, various personal computers, laptops, smartphones, tablets, IoT devices, and portable wearable devices. The server 102 can be implemented as a standalone server or a server cluster consisting of multiple servers.

[0038] The following method is applied to Figure 1 The server 102 in FIG. 1 is used as an example to illustrate. Figure 2 As shown, an embodiment of the present application provides a method for delivering media information, including the following steps:

[0039] Step S201 : in response to a media information delivery request for a target account sent by the terminal 101 of the target account, a plurality of media information to be delivered corresponding to a plurality of different display modes are obtained; wherein each piece of media information to be delivered corresponds to a different display mode.

[0040] A media information delivery request to a target account refers to a request to deliver media information (e.g., an advertisement) to the target account. This request can be initiated by the target account's terminal 101 to the server 102 that manages media information. For example, when the target account needs to complete a task related to media information (e.g., browsing media information), the terminal 101 can initiate a media information delivery request to the server 102.

[0041] Pending media information refers to media information that is intended for delivery to a target account. There can be multiple pieces of pending media information, each corresponding to a different information display method. For example, a certain display method might require the target account to perform a certain triggering action before displaying the pending media information, or a certain display method might require the target account to meet a specific account level before displaying the pending media information.

[0042] The first conversion virtual resource value can be calculated in advance, and refers to the estimated virtual resource value that can be converted after the media information to be placed is placed and displayed. The estimated virtual resource value can be the ECPM corresponding to the media information to be placed, which can represent the virtual resource value that can be converted per thousand displays of the media information to be placed.

[0043] In this embodiment, when the server 102 receives a media information delivery request initiated by the terminal 101 of the target account, it may first collect multiple pieces of media information to be delivered corresponding to different display modes, and the first converted virtual resource value corresponding to each piece of media information to be delivered.

[0044] Step S202: obtain the first conversion virtual resource value of each piece of media information to be placed among the multiple pieces of media information to be placed, obtain the media information exposure rate of each piece of media information to be placed for the target account, and based on the first conversion virtual resource value of each piece of media information to be placed and the media information exposure rate, obtain the second conversion virtual resource value of each piece of media information to be placed for the target account.

[0045] The media information exposure rate refers to the probability that each piece of media information to be delivered will be displayed on the target account's terminal 101. In this embodiment, not all media information to be delivered will necessarily be displayed on the target account's terminal 101 after being delivered to the target account. For example, if the display condition of a certain piece of media information to be delivered requires that the target account meet a certain level condition before displaying it, but if the target account does not currently meet the level condition, then the media information to be delivered will not be displayed on the target account's terminal 101, and the media information exposure rate will be 0. Similarly, if the display condition of a certain piece of media information to be delivered requires that the target account complete a certain task before displaying it, then the media information exposure rate can be equivalent to the probability that the target account completes the task.

[0046] The second conversion virtual resource value refers to the virtual resource value that can be converted after the media information to be delivered is delivered to the target account, which is calculated based on the first conversion virtual resource of the media information to be delivered and the media information exposure rate. The second conversion virtual resource value can be obtained by calculating the product of the first conversion virtual resource value and the media information exposure rate.

[0047] Specifically, after obtaining the media information exposure rate of each media information to be delivered for the target account, the server 102 can also multiply the media information exposure rate of the same media information to be delivered by the first conversion virtual resource value to obtain the second conversion virtual resource value of the media information to be delivered for the target account.

[0048] Step S203 : Based on the second converted virtual resource value of each piece of media information to be delivered, target media information for delivery to a terminal of a target account for display is acquired from the plurality of pieces of media information to be delivered.

[0049] The target media information is the media information that is finally delivered to the terminal 101 of the target account for display among multiple media information to be delivered in different display modes. After obtaining the second conversion virtual resource value of each media information to be delivered for the target account, the server 102 can filter the target media information according to the above-mentioned second conversion virtual resource value. For example, the target media information can be filtered out from each media information to be delivered in order of the size of the second conversion virtual resource value, so as to deliver the target media information to the target account for display.

[0050] When the above-mentioned media information delivery method receives a media information delivery request for a target account, it collects the media information exposure rate of each media information to be delivered for the target account, so that the first conversion virtual resource value of the media information to be delivered and the media information exposure rate of the media information to be delivered for the target account can be used to obtain the second conversion virtual resource value of each media information to be delivered for the target account, and then the second conversion virtual resource value is used to screen the target media information delivery. This method takes into account the exposure rate of media information with different display methods for a specific target account to achieve the delivery of multiple types of media information. Since the media information exposure rates of different target accounts represent the different browsing characteristics of different user individuals for media information, this method takes into account the differences between individual users, thereby improving the accuracy of media information delivery.

[0051] In one embodiment, the media information to be delivered includes: a first type of media information to be delivered that is displayed on the terminal of any account only in response to a trigger operation performed by any account on the media information display entrance provided by any account to the terminal of any account; step S202 may further include: obtaining a predicted trigger rate of the media information display entrance provided by the target account to the terminal of the target account; and using the predicted trigger rate as the media information exposure rate of the first type of media information to be delivered for the target account.

[0052] The first category of media information to be delivered refers to media information to be delivered that is displayed in the first display mode. This media information to be delivered requires any account to perform a trigger operation on the media information display entrance provided by its account terminal before it will be displayed on the account terminal. In other words, if the first category of media information to be delivered is determined to be the target media information, then the server 102 will first deliver the information display entrance of the first category of media information to be delivered to the terminal 101 for display, rather than directly delivering the first category of media information to be delivered to the terminal 101 for display. Afterwards, if the target account performs a trigger operation on the information display entrance, the first category of media information to be delivered, which is the target media information, will be delivered to the terminal 101 of the target account for display. In this case, the media information exposure rate of the first category of media information to be delivered to the target account can be equivalent to the probability that the target account will perform a trigger operation on the displayed information display entrance, that is, the predicted trigger rate of the target account on the information display entrance.

[0053] Specifically, if the media information to be placed is the first category of media information to be placed, then the media information exposure rate of the media information to be placed can be the predicted trigger rate of the target account for the information display entrance of the first category of media information to be placed, that is, by obtaining the predicted trigger rate of the target account for the information display entrance, the predicted trigger rate can be used as the media information exposure rate of the first category of media information to be placed for the target account.

[0054] In this embodiment, the media information to be placed can be the first type of media information to be placed that is displayed only after the account performs a trigger operation on the information display entrance. The media information exposure rate of the first type of media information to be placed can be obtained by obtaining the predicted trigger rate of the target account for the above-mentioned information display entrance. In this way, the accuracy of obtaining the media information exposure rate corresponding to the first type of media information to be placed can be improved.

[0055] Further, if Figure 3 As shown, obtaining the predicted trigger rate of the target account for the information display portal may further include:

[0056] Step S301: splice the profile features, historical behavior features, and real-time behavior features of the target account to obtain the spliced ​​account features of the target account.

[0057] Account portrait features may refer to the account portrait information of the target account, such as the target account's gender, age, city, mobile phone brand used, etc. Historical behavioral features refer to the behavioral features of the target account in a historical time period, such as the behavioral features of the target account in a certain historical time period, such as the average daily content consumption duration or quantity in the past n days, the average daily task entry exposure times, click times, click-through rate, and click-through rate of the kth exposure in the past n days. Real-time behavioral features refer to the behavioral features of the target account in the current real-time time period, such as the target account's real-time content consumption duration or quantity on that day, the target account's real-time task entry exposure times, click times, click-through rate, and whether the kth exposure was clicked, etc.

[0058] Specifically, the server 102 can collect the account portrait features, historical behavior features and real-time behavior features of the target account, and then splice the account portrait features, historical behavior features and real-time behavior features of the target account to obtain the spliced ​​account features of the target account.

[0059] Step S302: Input the combined account features of the target account into a pre-built display portal trigger prediction model, and obtain the predicted trigger rate of the media information display portal provided by the target account to the terminal of the target account through the display portal trigger prediction model.

[0060] The display portal trigger prediction model is a pre-built prediction model used to predict the probability of an account triggering an information display portal. The model's input can be the concatenated account features of the target account. Since the first type of media information to be delivered is not displayed directly after the information display portal is displayed, but rather after the target account triggers the information display portal, the first type of media information to be delivered is not displayed directly. This means that the target account is unaware of the specific content of the first type of media information before it is displayed. Therefore, the display portal trigger prediction model does not require the features of the media information; only the features of the target account are needed.

[0061] Specifically, the server 102 may input the spliced ​​features of the target account into a pre-built display portal trigger prediction model, and the model may output the predicted trigger rate of the target account for the information display portal.

[0062] In this embodiment, the server 102 can also perform feature splicing on the collected account portrait features, historical behavior features, and real-time behavior features of the target account, thereby inputting the spliced ​​features of the target account into a pre-built display portal trigger prediction model, and the model outputs the predicted trigger rate of the target account for the information display portal. In this way, the accuracy of obtaining the predicted trigger rate can be improved.

[0063] like Figure 4 As shown, before step S302, a training process of the display entry trigger prediction model may also be included, specifically including:

[0064] Step S401: Obtain sample account portrait features, sample historical behavior features, and sample real-time behavior features of the sample account.

[0065] A sample account refers to the account corresponding to the sample data used to train the display entry trigger prediction model, while the sample account profile features, sample historical behavior features, and sample real-time behavior features are the account profile features, historical behavior features, and real-time behavior features of the sample account, respectively. In this embodiment, when training the entry trigger prediction model, it is first necessary to collect the sample account profile features, sample historical behavior features, and sample real-time behavior features of the sample account.

[0066] Step S402: Obtain historical triggering results of the sample account for the information display portal for displaying the first type of media information to be delivered.

[0067] Historical trigger results refer to the triggering results of the sample account for the information display portal of the media information displayed in the first display mode. These triggering results can include two types: one is the result of the media information information display portal being displayed and the sample account performing a triggering operation on the information display portal; the other is the result of the media information information display portal being displayed but the sample account not performing a triggering operation on the information display portal, which serve as positive and negative samples respectively. In this embodiment, server 102 can also collect historical triggering results of the sample account for the information display portal used to display the media information in the first display mode to implement training of the display portal trigger prediction model.

[0068] Step S403: The sample account portrait features, sample historical behavior features, and sample behavior features are spliced ​​together to obtain the spliced ​​account features of the sample account, and the spliced ​​account features of the sample account and the historical trigger results are used to train a classification model to obtain a display entry trigger prediction model.

[0069] Finally, the server 102 can splice the sample account portrait features, sample historical behavior features, and sample real-time behavior features to obtain the spliced ​​features of the sample account. Then, the server 102 can use the spliced ​​features of the sample account and the historical trigger results to train the classification model to obtain a display entry trigger prediction model. For example, the spliced ​​features of the sample account can be input into the classification model to output the predicted trigger result of the sample account for the information display entry of the media information of the first display method, and then use the difference between the predicted trigger result and the historical trigger result to train the classification model to obtain a display entry trigger prediction model.

[0070] In this embodiment, the display entry trigger prediction model can also be trained by collecting account portrait features, historical behavior features and real-time behavior features of sample accounts, as well as collecting historical trigger results of sample accounts for media information display entries. In this way, the accuracy of the display entry trigger prediction model in predicting the trigger rate of the information display entry can be improved.

[0071] In one embodiment, the media information to be delivered includes: a second type of media information to be delivered that is displayed directly on the terminal of any account without responding to a trigger operation performed by any account on a media information display portal provided by the terminal of any account; step S202 may further include: obtaining a pre-set media information exposure rate for the second type of media information to be delivered; and using the pre-set media information exposure rate as the media information exposure rate of the second type of media information to be delivered for the target account.

[0072] The second type of media information to be delivered refers to media information to be delivered that is displayed in the second display mode. This media information to be delivered can be directly displayed on the account's terminal without requiring the account to first perform a triggering operation before displaying it. That is, the exposure rate of the media information to be delivered will not be affected by the target account. Therefore, in this case, the server 102 can directly use the media information exposure rate pre-set for the second type of media information to be delivered as the media information exposure rate of the second type of media information to be delivered for the target account. For example, the media information exposure rate can be pre-set to 100%, indicating that the media information exposure rate can be 100% displayed on the target account's terminal 101.

[0073] In this embodiment, if the media information to be delivered is the second type of media information to be delivered that is directly displayed without the target account triggering an interactive operation, then the server 102 can directly use the media information exposure rate pre-set for the second type of media information to be delivered as the media information exposure rate of the second type of media information to be delivered for the target account. In this way, the efficiency of obtaining the media information exposure rate of the second type of media information to be delivered for the target account can be improved.

[0074] In one embodiment, Figure 5 As shown, step S201 may further include:

[0075] Step S501 : obtaining multiple pieces of candidate media information corresponding to each presentation mode, and obtaining a first converted virtual resource value of each piece of candidate media information in the multiple pieces of candidate media information.

[0076] Candidate media information can refer to any piece of media information in each display method. Since multiple pieces of media information to be placed correspond to different display methods, and a certain display method may contain multiple candidate media information, in this embodiment, it is necessary to first filter out one of the multiple candidate media information for each display method as the media information to be placed for that display method, thereby obtaining multiple pieces of media information to be placed in different display methods.

[0077] For example, the display method may include category A and category B, where category A includes candidate media information A1 and candidate media information A2, and category B includes candidate media information B1 and candidate media information B2. Then, it is necessary to first filter out the media information to be placed corresponding to category A from the candidate media information A1 and the candidate media information A2, and to filter out the media information to be placed corresponding to category B from the candidate media information B1 and the candidate media information B2, so as to obtain multiple media information to be placed corresponding to different display methods.

[0078] Specifically, the server 102 may first determine multiple candidate media information for each display mode, and respectively obtain a reference virtual resource that can be converted after each candidate media information is released, that is, a first converted virtual resource for each candidate media information.

[0079] Step S502 : determining the maximum value among the first converted virtual resource values ​​of the plurality of candidate media information corresponding to each display mode, and using the candidate media information corresponding to each maximum value as the media information to be delivered for each display mode.

[0080] After obtaining the first conversion virtual resource value of each candidate media information corresponding to each display method, the candidate media information corresponding to the maximum value of the first conversion virtual resource can be used as the media information to be placed for each display method. In this way, the media information to be placed for each display method can be obtained, and the above-mentioned media information to be placed can be used as multiple media information to be placed corresponding to different display methods.

[0081] In this embodiment, the candidate media information corresponding to the maximum value of the first conversion virtual resource can also be screened out from the multiple candidate media information for each display mode as the media information to be placed for each display mode, thereby obtaining multiple media information to be placed corresponding to different display modes. This method can ensure that the multiple media information to be placed for different display modes are the media information corresponding to the maximum value of the first conversion virtual resource in the corresponding display mode, thereby further improving the virtual resource conversion rate after the media information is placed.

[0082] In addition, step S203 may further include: obtaining the maximum value of the second conversion virtual resource value of each media information to be delivered; and using the media information to be delivered corresponding to the maximum value of the second conversion virtual resource value as the target media information for delivery to the terminal of the target account for display.

[0083] In this embodiment, the target media information can be the media information to be placed corresponding to the maximum value of the second conversion virtual resource. After obtaining the second conversion virtual resource value of each media information to be placed, the server 102 can also use the media information to be placed corresponding to the maximum value of the second conversion virtual resource as the target media information, thereby ensuring that the virtual resource conversion rate of the media information placed to the target account is maximized.

[0084] For example, the screening of target media information can be achieved through the following formula:

[0085]

[0086] in, The first converted virtual resource value representing the i-th piece of media information to be delivered may be the virtual resource converted from the average thousand exposures of the i-th piece of media information to be delivered. It represents the media information exposure rate of the target account for the media information to be delivered by the i-th media information. This represents the second converted virtual resource value of the i-th piece of media information to be delivered to the target account.

[0087] In this embodiment, the target media information may be the to-be-delivered media information corresponding to the maximum value of the second converted virtual resource value. This approach ensures that the virtual resource conversion rate of the media information delivered to the target account is maximized.

[0088] In one embodiment, a method for dynamically issuing multiple types of advertisements based on user behavior prediction is also provided. By real-time estimating the click-through rate (CTR) of the user's interactive entrance to the first type of advertisement and combining the ECPM of different types of advertisements, the advertisement type that maximizes conversion resources is dynamically selected.

[0089] Compared with the original fixed display strategy, this embodiment constructs a dynamic delivery strategy for multiple types of advertisements that maximizes conversion resources: when a user triggers the advertisement display opportunity, the system calculates the expected conversion resources of the first type of advertisement (real-time CTR estimate of the interactive portal × first type of advertisement ECPM) and the expected conversion resources of the second type of advertisement (second type of advertisement ECPM), and selects the advertisement type with higher conversion resources for dynamic delivery. This embodiment can effectively improve the resource conversion rate of advertisements without affecting the user experience by dynamically delivering multiple types of advertisements. Among them, the first type of advertisement refers to an advertisement that is triggered after the user actively clicks on the interactive portal. The advertisement is presented by first displaying the interactive portal, and when the user clicks on the interactive portal, the advertisement is displayed. The second type of advertisement refers to an advertisement that is directly displayed without user interaction. The advertisement is presented by directly displaying the advertisement itself. When the user clicks on the advertisement, the user enters the landing page of the advertisement.

[0090] Therefore, for the first type of ads, since there is no ad exposure at the interactive entrance, the ad exposure is triggered after the user clicks the entrance button, that is, the ad exposure rate is directly linked to the entrance click-through rate. The resource expectation for ad exposure should be calculated by combining the ad ECPM and the ad exposure rate (i.e., the entrance click-through rate). For the second type of ads, exposure will be obtained when the task entrance is displayed, that is, the ad exposure rate is 100%. The resource expectation for ad exposure is the ad ECPM. Figure 6 As shown, taking pop-up ads as an example, the first type of ads can be a pop-up window on the interface that first displays the interactive entrance, and when the user clicks the interactive entrance, the ad is displayed, while the second type of ads is to directly display the ad itself in the pop-up window, and when the user clicks the ad, the ad is displayed on the landing page. Figure 7 As shown, taking banner ads as an example, the first type of ads can be to display an interactive entrance in the form of a banner on the interface, and the ad is displayed when the user clicks on the interactive entrance, while the second type of ads is to directly display the ad itself in the form of a banner on the interface, and when the user clicks on the ad, he enters the ad landing page.

[0091] The specific implementation process of this embodiment can be as follows Figure 8 As shown, the following steps may be included:

[0092] Step 1: Get multiple types of ads and their ECPMs.

[0093] Unlike the task entry that is fixed to a single advertising type, in order to compare resources between multiple types of advertisements, each time an advertisement is requested from the advertising middle platform, multiple types of advertisements and their ECPMs will be pulled at the same time.

[0094] Step 2: Estimation of advertising exposure rate.

[0095] For the second type of ads, the ad exposure rate is 100%. For the first type of ads, the ad exposure rate is equivalent to the entrance click rate. The entrance click rate can be obtained through the entrance click rate prediction model. The construction process of the model can be as follows: Figure 9 As shown, the following processes may be included:

[0096] In terms of sample construction, PV-level records of task entrances exposed in target scenarios but not clicked to complete are collected as negative samples. At the same time, PV-level records of task entrances exposed in target scenarios and users clicked to complete are collected as positive samples.

[0097] In terms of feature construction, unlike traditional ad exposure and click scenarios, ad portal exposure does not reveal the underlying ad material, and users are unaware of the specific content of the ad before exposure. Therefore, the portal click-through rate prediction model does not require the inclusion of ad features, but rather a large number of user features. User features include relatively static user portrait features, relatively dynamic user historical behavior features, and user real-time behavior features. Among them:

[0098] User portrait features mainly include user gender, age, city, mobile phone brand, etc.

[0099] User historical behavior characteristics primarily include the user's average daily content consumption duration or volume over the past 1 / 3 / 7 / 14 / 28 / 84 days, the user's average daily task entry exposures, clicks, click-through rate, and the click-through rate of the kth exposure over the past 1 / 3 / 7 / 14 / 28 / 84 days. User historical behavior characteristics depict the user's average short-, medium-, and long-term content consumption, as well as the average task session volume and participation probability, and are crucial for estimating the user's task entry click-through rate.

[0100] Real-time user behavior characteristics primarily include the user's daily content consumption duration or quantity, the number of task entry exposures, clicks, click-through rate, and whether the kth exposure was clicked. While real-time user behavior characteristics statistically converge with historical user behavior characteristics, they may also exhibit real-time differences and are crucial for estimating the click-through rate of task entry.

[0101] Next, user profile features, historical user behavior features, and real-time user behavior features can be combined to train and validate a binary classification model (whether the task entry was clicked or not). This model can use either a classic tree model or a deep learning model. Finally, the model service is deployed to score all user requests, thereby deriving an estimated click-through rate for each user's task entry.

[0102] Step 3: Compare resources and return results.

[0103] Enter the ad eCPM and exposure rate p obtained in the previous stepexpo , output the ranking score corresponding to each style, sort the styles with the highest scores from large to small according to the ranking scores corresponding to each style, and return the results. The specific calculation formula is as follows:

[0104]

[0105] Through this embodiment, dynamic delivery of multiple types of advertisements can be achieved, which can effectively improve the resource conversion rate of advertisement exposure.

[0106] It should be understood that, although the various steps in the flowcharts involved in the various embodiments described above are displayed in sequence as indicated by 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 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.

[0107] Based on the same inventive concept, embodiments of the present application also provide a media information delivery device for implementing the aforementioned media information delivery method. The solution provided by this device is similar to the solution described in the aforementioned method. Therefore, the specific limitations of one or more of the following embodiments of the media information delivery device can be found in the aforementioned limitations of the media information delivery method and will not be further elaborated here.

[0108] In one embodiment, Figure 10 As shown, a media information delivery device is provided, comprising: a first resource acquisition module 1001, a second resource acquisition module 1002 and a media information delivery module 1003, wherein:

[0109] The first resource acquisition module 1001 is configured to, in response to a media information delivery request for a target account sent by a terminal of the target account, acquire a plurality of media information to be delivered corresponding to a plurality of different display modes; wherein each piece of media information to be delivered corresponds to a different display mode;

[0110] The second resource acquisition module 1002 is configured to acquire a first converted virtual resource value for each piece of media information to be delivered from the plurality of media information to be delivered, acquire a media information exposure rate for each piece of media information to be delivered from the plurality of media information to be delivered to the target account, and obtain a second converted virtual resource value for each piece of media information to be delivered to the target account based on the first converted virtual resource value and the media information exposure rate of each piece of media information to be delivered;

[0111] The media information delivery module 1003 is configured to obtain target media information for delivery to a terminal of a target account for display from multiple pieces of media information to be delivered based on the second converted virtual resource value of each piece of media information to be delivered.

[0112] Each module in the aforementioned media information delivery device may be implemented in whole or in part through software, hardware, or a combination thereof. Each module may be embedded in or independent of a processor within a computer device in the form of hardware, or may be stored in a computer device memory in the form of software, so that the processor can call and execute the corresponding operations of each module.

[0113] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 11 As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, abbreviated as I / O) and a communication interface. The processor, memory and input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The database of the computer device is used to store media information data. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a media information delivery method is implemented.

[0114] Those skilled in the art will understand that Figure 11 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0115] In one embodiment, a computer device is further provided, including a memory and a processor. The memory stores a computer program, and the processor implements the steps in the above method embodiments when executing the computer program.

[0116] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0117] In one embodiment, a computer program product is provided, including a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0118] It should 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 analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0119] Those skilled in the art will appreciate that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the above-mentioned embodiments. In particular, any reference to memory, database, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The databases involved in the various embodiments provided herein may include at least one of a relational database and a non-relational database. Non-relational databases may include, but are not limited to, distributed databases based on blockchains. The processors involved in the various embodiments provided herein may be, but are not limited to, general-purpose processors, central processing units (CPUs), graphics processing units (GPUs), digital signal processors (DSPs), programmable logic devices (PLDs), data processing logic devices based on quantum computing, and the like.

[0120] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0121] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present application. It should be noted that a person of ordinary skill in the art may make various modifications and improvements without departing from the spirit of the present application, and these modifications and improvements fall within the scope of protection of the present application. Therefore, the scope of protection of the present application shall be determined by the appended claims.

Claims

1. A method for delivering media information, characterized in that: The method comprises: In response to a media information delivery request for the target account sent by a terminal of the target account, a plurality of media information to be delivered corresponding to a plurality of different display modes is obtained; wherein each piece of the media information to be delivered corresponds to a different display mode; Obtaining a first conversion virtual resource value for each of the multiple pieces of media information to be placed, obtaining a media information exposure rate for each of the multiple pieces of media information to be placed for the target account, and obtaining a second conversion virtual resource value for each of the pieces of media information to be placed for the target account based on the first conversion virtual resource value and the media information exposure rate of each piece of media information to be placed; Based on the second converted virtual resource value of each piece of the media information to be delivered, target media information for delivery to the terminal of the target account for display is obtained from the multiple pieces of media information to be delivered.

2. The method according to claim 1, characterized in that The media information to be delivered includes: a first type of media information to be delivered that is displayed on the terminal of any account only in response to a triggering operation performed by any account on a media information display portal provided by the terminal of any account; The obtaining of the media information exposure rate of each of the plurality of media information to be delivered to the target account includes: Obtaining a predicted trigger rate of a media information display portal provided by the target account to a terminal of the target account; The predicted trigger rate is used as the media information exposure rate of the first type of media information to be delivered to the target account.

3. The method according to claim 2, characterized in that The obtaining of a predicted trigger rate of a media information display portal provided by the target account to the terminal of the target account includes: Splicing the target account's profile features, historical behavior features, and real-time behavior features to obtain spliced ​​account features of the target account; The spliced ​​account features of the target account are input into a pre-built display entry trigger prediction model, and the display entry trigger prediction model is used to obtain a predicted trigger rate of the media information display entry provided by the target account to the terminal of the target account.

4. The method according to claim 3, characterized in that Before inputting the combined account features of the target account into the pre-built display entry trigger prediction model, the method further includes: Obtain sample account profile features, sample historical behavior features, and sample real-time behavior features of the sample account; Obtaining historical triggering results of the sample account for an information display portal for displaying the first type of media information to be delivered; The sample account portrait features, sample historical behavior features, and sample behavior features are spliced ​​together to obtain the spliced ​​account features of the sample account, and the spliced ​​account features of the sample account and the historical trigger results are used to train a classification model to obtain the display entry trigger prediction model.

5. The method according to claim 1, wherein The media information to be delivered includes: second type of media information to be delivered that is directly displayed on the terminal of any account without responding to a triggering operation performed by any account on a media information display portal provided by the terminal of the any account; obtaining a media information exposure rate of each of the multiple media information to be delivered to the target account includes: Obtaining a pre-set media information exposure rate for the second type of media information to be delivered; The preset media information exposure rate is used as the media information exposure rate of the second type of media information to be delivered for the target account.

6. The method according to claim 1, characterized in that The obtaining of multiple pieces of to-be-delivered media information corresponding to multiple different display modes includes: Acquire multiple pieces of candidate media information corresponding to each of the display modes, and acquire a first converted virtual resource value of each piece of the multiple pieces of candidate media information; The maximum value among the first converted virtual resource values ​​of the plurality of candidate media information corresponding to each of the display modes is determined respectively, and the candidate media information corresponding to each of the maximum values ​​is used as the media information to be delivered for each of the display modes.

7. The method according to any one of claims 1 to 6, characterized in that The step of obtaining target media information for delivery to the terminal of the target account for display from the plurality of media information to be delivered based on the second converted virtual resource value of each piece of media information to be delivered includes: Obtaining a maximum value of the second converted virtual resource values ​​of each piece of the media information to be delivered; The to-be-delivered media information corresponding to the maximum value of the second converted virtual resource value is used as the target media information for delivery to the terminal of the target account for display.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images