Method, apparatus and storage medium for sending presentation information

Abstract

This application discloses a method, apparatus, device, and storage medium for sending display information, belonging to the field of computer application technology. The method includes: obtaining at least one object attribute tag for a target display object; determining a first set of account attribute tags corresponding to the at least one object attribute tag; determining a first set of reference accounts based on the first set of account attribute tags; determining the conversion rate corresponding to each account attribute tag in the first set of reference accounts based on at least one account attribute tag, at least one object attribute tag, and a conversion prediction model for each account in the first set of reference accounts; determining a first set of accounts to be displayed based on the conversion rate corresponding to each account attribute tag in the first set of reference accounts; and sending display information of the target display object to the accounts in the first set of accounts to be displayed. Using this application can improve the conversion rate of the target display object.

Description

Technical Field

[0001] This application relates to the field of computer application technology, and in particular to a method, apparatus, device, and storage medium for transmitting display information. Background Technology

[0002] Currently, when users use applications, they are often pushed various display information. This information is generally displayed through windows, and users can enter the page of the corresponding display object by clicking on the display information window. For example, the display object may be a product, and the corresponding display information may be the product's pictures, videos, text descriptions, etc. Another example is that the display object may be an inquiry, and the corresponding display information may be the title or summary of the inquiry. Yet another example is that the display object may be a video, and the corresponding display information may be the video's title, clips, etc.

[0003] Related technologies typically include a push notification mechanism that sends information to users and tracks the conversion rate of this mechanism to evaluate its effectiveness. Generally, if a user clicks on a pushed notification to access the displayed content's page and completes a specified action on that page, the conversion is considered successful. The conversion rate is the percentage of successfully converted messages out of the total number of notifications pushed under a given mechanism.

[0004] In related technologies, the push mechanism commonly used in applications involves the application's backend server matching the display object with the object attribute tags and the account attribute tags of the account, and then pushing the display information of the display object to the accounts in this account set.

[0005] The above solution simply selects an account for the displayed object based on the relationship between object attribute tags and account attribute tags, and pushes the displayed information. The method of selecting accounts is rather crude, resulting in a low conversion rate. Summary of the Invention

[0006] This application provides a method for sending display information, which can improve the conversion rate of the target display object.

[0007] Firstly, a method for sending display information is provided, the method comprising:

[0008] Retrieve at least one object attribute label of the target display object;

[0009] Determine a first set of account attribute tags corresponding to the at least one object attribute tag, and determine a first set of reference accounts based on the first set of account attribute tags;

[0010] Based on at least one account attribute tag for each account in the first reference account set, the at least one object attribute tag, and the conversion prediction model, the conversion rate corresponding to each account attribute tag in the first reference account set is determined.

[0011] Based on the conversion rate corresponding to each account attribute tag in the first reference account set, a first set of accounts to be displayed is determined, and the display information of the target display object is sent to the accounts in the first set of accounts to be displayed.

[0012] In one possible design, determining the first reference account set based on the first account attribute tag set includes:

[0013] The set of accounts corresponding to each account attribute tag in the first account attribute tag set is determined as the first reference account set.

[0014] In one possible design, determining the first reference account set based on the first account attribute tag set includes:

[0015] A first preset number of accounts are randomly selected from the accounts corresponding to the first set of account attribute tags to form a second set of accounts to be displayed.

[0016] Send the display information of the target display object to each account in the second set of accounts to be displayed, and obtain the operation information of each account in the second set of accounts to be displayed on the target display object;

[0017] Based on the operation information and conversion rate threshold of each account in the second set of accounts to be displayed to the target display object, a second set of account attribute tags is determined;

[0018] The first reference account set is determined based on the second account attribute tag set.

[0019] In one possible design, determining the second set of account attribute tags based on the operation information and conversion rate threshold of each account in the second set of accounts to be displayed to the target display object includes:

[0020] Based on the operation information of each account in the second set of accounts to be displayed to the target display object and at least one account attribute tag, the conversion rate corresponding to each account attribute tag in the second set of accounts to be displayed is determined;

[0021] In each account attribute tag corresponding to the second set of accounts to be displayed, obtain the account attribute tags whose conversion rate is greater than or equal to the conversion rate threshold, and form the second set of account attribute tags.

[0022] In one possible design, determining the first reference account set based on the second account attribute tag set includes:

[0023] Based on the conversion rate corresponding to each account attribute tag in the second set of account attribute tags, determine the weighted score corresponding to each account attribute tag;

[0024] Based on the weighted score corresponding to each account attribute tag, determine the total weighted score for each account that has at least one account attribute tag from the second account attribute tag set;

[0025] A second preset number of accounts are selected in descending order of total weighted score to form the first reference account set.

[0026] In one possible design, determining the conversion rate corresponding to each account attribute tag in the first reference account set based on at least one account attribute tag for each account in the first reference account set, the at least one object attribute tag, and the conversion prediction model includes:

[0027] Based on at least one account attribute label, at least one object attribute label, and the conversion prediction model for each account in the first reference account set, conversion prediction information for each account in the first reference account set is obtained.

[0028] Based on the conversion prediction information of each account in the first reference account set and at least one account attribute tag, the conversion rate corresponding to each account attribute tag in the first reference account set is determined.

[0029] In one possible design, determining the first set of accounts to be displayed based on the conversion rate corresponding to each account attribute tag in the first set of reference accounts includes:

[0030] In each account attribute tag corresponding to the first reference account set, obtain the account attribute tags whose conversion rate is greater than or equal to the conversion rate threshold, and form a third account attribute tag set.

[0031] A third preset number of accounts are selected from the accounts corresponding to the third account attribute tag set to form the first set of accounts to be displayed.

[0032] In one possible design, after sending the display information of the target display object to the accounts in the first set of accounts to be displayed, the method further includes:

[0033] Obtain the operation information of the accounts in the first set of accounts to be displayed on the target display object;

[0034] Based on the operation information and conversion rate threshold of the accounts in the first set of accounts to be displayed to the target display object, a fourth set of account attribute tags is determined;

[0035] The second reference account set is determined based on the fourth account attribute tag set;

[0036] Based on at least one account attribute tag, at least one object attribute tag, and the conversion prediction model for each account in the second reference account set, the conversion rate corresponding to each account attribute tag in the second reference account set is determined.

[0037] Based on the conversion rate corresponding to each account attribute tag in the second reference account set, a third set of accounts to be displayed is determined, and the display information of the target display object is sent to the accounts in the third set of accounts to be displayed.

[0038] Secondly, a device for sending display information is provided, the device comprising:

[0039] The first acquisition module is used to acquire at least one object attribute label of the target display object;

[0040] The first determining module is used to determine a first set of account attribute tags corresponding to the at least one object attribute tag, and to determine a first set of reference accounts based on the first set of account attribute tags.

[0041] The second determining module is used to determine the conversion rate corresponding to each account attribute tag of the first reference account set based on at least one account attribute tag of each account in the first reference account set, the at least one object attribute tag and the conversion prediction model.

[0042] The first sending module is used to determine the first set of accounts to be displayed based on the conversion rate corresponding to each account attribute tag in the first set of reference accounts, and to send the display information of the target display object to the accounts in the first set of accounts to be displayed.

[0043] In one possible design, the first determining module is configured to:

[0044] The set of accounts corresponding to each account attribute tag in the first account attribute tag set is determined as the first reference account set.

[0045] In one possible design, the first determining module is configured to:

[0046] A first preset number of accounts are randomly selected from the accounts corresponding to the first set of account attribute tags to form a second set of accounts to be displayed.

[0047] Send the display information of the target display object to each account in the second set of accounts to be displayed, and obtain the operation information of each account in the second set of accounts to be displayed on the target display object;

[0048] Based on the operation information and conversion rate threshold of each account in the second set of accounts to be displayed to the target display object, a second set of account attribute tags is determined;

[0049] The first reference account set is determined based on the second account attribute tag set.

[0050] In one possible design, the first determining module is configured to:

[0051] Based on the operation information of each account in the second set of accounts to be displayed to the target display object and at least one account attribute tag, the conversion rate corresponding to each account attribute tag in the second set of accounts to be displayed is determined;

[0052] In each account attribute tag corresponding to the second set of accounts to be displayed, obtain the account attribute tags whose conversion rate is greater than or equal to the conversion rate threshold, and form the second set of account attribute tags.

[0053] In one possible design, the first determining module is configured to:

[0054] Based on the conversion rate corresponding to each account attribute tag in the second set of account attribute tags, determine the weighted score corresponding to each account attribute tag;

[0055] Based on the weighted score corresponding to each account attribute tag, determine the total weighted score for each account that has at least one account attribute tag from the second account attribute tag set;

[0056] A second preset number of accounts are selected in descending order of total weighted score to form the first reference account set.

[0057] In one possible design, the second determining module is used for:

[0058] Based on at least one account attribute label, at least one object attribute label, and the conversion prediction model for each account in the first reference account set, conversion prediction information for each account in the first reference account set is obtained.

[0059] Based on the conversion prediction information of each account in the first reference account set and at least one account attribute tag, the conversion rate corresponding to each account attribute tag in the first reference account set is determined.

[0060] In one possible design, the first transmitting module is used for:

[0061] In each account attribute tag corresponding to the first reference account set, obtain the account attribute tags whose conversion rate is greater than or equal to the conversion rate threshold, and form a third account attribute tag set.

[0062] A third preset number of accounts are selected from the accounts corresponding to the third account attribute tag set to form the first set of accounts to be displayed.

[0063] In one possible design, the device further includes:

[0064] The second acquisition module is used to acquire operation information of the accounts in the first set of accounts to be displayed on the target display object;

[0065] The third determining module is used to determine the fourth set of account attribute tags based on the operation information and conversion rate threshold of the accounts in the first set of accounts to be displayed to the target display object;

[0066] The fourth determining module is used to determine the second reference account set based on the fourth account attribute tag set;

[0067] The fifth determining module is used to determine the conversion rate corresponding to each account attribute tag of the second reference account set based on at least one account attribute tag of each account in the second reference account set, the at least one object attribute tag, and the conversion prediction model.

[0068] The second sending module is used to determine the third set of accounts to be displayed based on the conversion rate corresponding to each account attribute tag in the second set of reference accounts, and to send the display information of the target display object to the accounts in the third set of accounts to be displayed.

[0069] Thirdly, a computer device is provided, the computer device including a processor and a memory, the memory storing at least one instruction, the instruction being loaded and executed by the processor to perform the operation of the method for sending display information.

[0070] Fourthly, a computer-readable storage medium is provided, wherein at least one instruction is stored in the storage medium, the instruction being loaded and executed by a processor to implement the operation performed by the method of sending display information.

[0071] The beneficial effects of the technical solution provided in this application are as follows: The solution mentioned in this application can first determine a first set of account attribute tags based on at least one object attribute tag of the target display object, and then determine a first set of reference accounts based on the correspondence between account attribute tags and accounts. Further refinement is then performed on the first set of reference accounts. First, based on the conversion prediction model, the conversion rate corresponding to each account attribute tag in the first set of reference accounts is predicted. Then, based on the conversion rate corresponding to the account attribute tag, the account to be displayed is determined, and the display information of the target display object is pushed to it. In this way, accounts corresponding to account attribute tags with higher conversion rates can be selected as the push targets, which can better improve the probability of successful conversion for each account, thereby improving the conversion rate of the target display object. Attached Figure Description

[0072] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0073] Figure 1 This is a flowchart illustrating a method for sending display information provided in an embodiment of this application;

[0074] Figure 2 This is a schematic diagram of an information display page provided in an embodiment of this application;

[0075] Figure 3 This is a schematic diagram of a details page of a target display object provided in an embodiment of this application;

[0076] Figure 4 This is a schematic diagram of the download page of application A provided in an embodiment of this application;

[0077] Figure 5 This is a schematic diagram of a transformation page for a target display object provided in an embodiment of this application;

[0078] Figure 6 This is a schematic diagram of an information aggregation page provided in an embodiment of this application;

[0079] Figure 7 This is a schematic diagram of a page displaying information delivery details provided in an embodiment of this application;

[0080] Figure 8 This is a schematic diagram illustrating the determination of a first set of reference accounts provided in an embodiment of this application;

[0081] Figure 9This is a schematic diagram illustrating an embodiment of the present application for determining conversion prediction information for an account;

[0082] Figure 10 This is a flowchart illustrating a method for sending display information provided in an embodiment of this application;

[0083] Figure 11 This is a schematic diagram of the structure of a device for sending display information provided in an embodiment of this application;

[0084] Figure 12 This is a structural block diagram of a server provided in an embodiment of this application. Detailed Implementation

[0085] To make the objectives, technical solutions, and advantages of this application clearer, the embodiments of this application will be described in further detail below with reference to the accompanying drawings.

[0086] This application provides a method for sending display information, which can be implemented by a server. The server can be a single server or a server cluster consisting of multiple servers.

[0087] A server can be the backend server of an application that provides users with various displayed information. A server may include a processor, memory, communication components, etc., with the processor connected to the memory and communication components respectively.

[0088] The processor can be a CPU (Central Processing Unit). The processor can be used to read instructions and process data, such as obtaining the object attribute tags of the target display object, determining the first set of account attribute tags corresponding to the object attribute tags, determining the first set of reference accounts, determining the conversion rate corresponding to each account attribute tag in the first set of reference accounts, determining the first set of accounts to be displayed, and so on.

[0089] Memory may include ROM (Read-Only Memory), RAM (Random Access Memory), CD-ROM (Compact Disc Read-Only Memory), disks, optical data storage devices, etc. Memory can be used for data storage, such as storing at least one object attribute tag of the acquired target display object, storing data during the process of determining a first set of account attribute tags, storing data during the process of determining a first set of reference accounts, storing data during the process of determining the conversion rate corresponding to each account attribute tag in the first set of reference accounts, storing data during the process of determining a first set of accounts to be displayed, and so on.

[0090] The communication component can be a wired network connector, a WiFi (Wireless Fidelity) module, a Bluetooth module, a cellular network communication module, etc. The communication component can be used to receive and send signals, for example, the transmission of information when acquiring at least one object attribute tag of a target display object, the transmission of information when sending display information of a target display object to accounts in a first set of accounts to be displayed, and so on.

[0091] The application in this embodiment can be an application that provides information display to users, such as a browser, shopping application, video application, novel application, etc. The backend server of the application can recommend suitable information, products, videos, etc. to users based on the user's account attribute tags.

[0092] Figure 1 This is a flowchart illustrating a method for sending display information according to an embodiment of this application. See also... Figure 1 This embodiment includes:

[0093] 101. Obtain at least one object attribute tag of the target display object.

[0094] In implementation, the target display object can be information, products, stores, videos, etc. Once the target display object to be recommended is determined, the object attribute tags of the target display object can be obtained based on the displayed content. A target display object can have one or more object attribute tags. For example, if the target display object is a children's English tutoring class, and the displayed information is an advertisement for that children's English tutoring class, then the object attribute tags of the target display object could be children, English, etc.

[0095] 102. Determine the first set of account attribute tags corresponding to at least one object attribute tag, and determine the first set of reference accounts based on the first set of account attribute tags.

[0096] In implementation, after determining at least one object attribute tag for the target display object, at least one account attribute tag corresponding to each target display object's object attribute tag can be determined based on the pre-stored correspondence between object attribute tags and account attribute tags. This determines at least one account attribute tag corresponding to at least one object attribute tag of the target display object. These account attribute tags are then deduplicated and combined into a set, which is the first account attribute tag set. For example, if the target display object's object attribute tags are "child" and "English," and the corresponding account attribute tags for "child" are "married" and "25-45 years old," and for "English" are "foreign company" and "translator," then the first account attribute tag set corresponding to at least one object attribute tag of the target display object is [married, 25-45 years old, foreign company, translator].

[0097] Generally, when an application's backend server detects the creation of a new user, it matches various account attribute tags to the user's account based on the user's historical operation information. The server stores the correspondence between these accounts and account attribute tags, so that each account attribute tag corresponds to one or more accounts. Accounts corresponding to different account attribute tags can include the same account or different accounts. After determining the first set of account attribute tags, at least one account corresponding to each account attribute tag in the first set of account attribute tags can be determined based on the pre-stored correspondence between account attribute tags and accounts. These accounts are then deduplicated and combined into a set, which is the first reference account set.

[0098] 103. Based on at least one account attribute label, at least one object attribute label, and the conversion prediction model for each account in the first reference account set, determine the conversion rate corresponding to each account attribute label in the first reference account set.

[0099] After determining the first set of reference accounts, predictions can be made for each account in the first set to predict the conversion prediction information for the target display object. The corresponding processing can be as follows: Using at least one account attribute tag from any account in the first set and at least one object attribute tag from the target display object as input to the conversion prediction model, the conversion prediction information for that account regarding the target display object can be obtained. This process is repeated for each account in the first set to obtain the conversion prediction information for each account in the first set regarding the target display object, thus predicting whether each account in the first set will convert successfully or fail. For example, if the target display object is a product, a successful conversion means purchasing the product, while a failed conversion means not purchasing the product. If the target display object is a piece of news, a successful conversion means that a user of that account clicks into the news page and browses for a duration greater than or equal to a first preset duration, while a failed conversion means that the user of that account does not click into the news page, or clicks into the news page but browses for a duration less than the first preset duration. The above processing determines whether each account in the first reference account set has a successful or unsuccessful conversion. Then, based on the account attribute tags corresponding to these accounts, the conversion rate for each account attribute tag is calculated. The conversion rate for each account attribute tag represents the proportion of accounts that successfully converted within that attribute tag. For example, if the first reference account set contains 5 accounts with three corresponding account attribute tags (A, B, and C), and the conversion prediction information for accounts 1, 3, and 5 is successful, while the prediction information for accounts 2 and 4 is unsuccessful, the account attribute tags for account 1 are A and B, for account 2 is C, for account 3 is A, for account 4 are B and C, for account 5 is B, and there are two accounts corresponding to the account attribute tag A (account 1...). If both accounts (the first and third accounts) convert successfully, the conversion rate for account attribute label A is 2 / 2 = 100%. If there are three accounts (the first, fourth, and fifth accounts) corresponding to account attribute label B, and two of these three accounts (the first and fifth accounts) convert successfully, the conversion rate for account attribute label B is 2 / 3 = 66.67%. If there are two accounts (the second and fourth accounts) corresponding to account attribute label C, and both of these accounts fail to convert, the conversion rate for account attribute label C is 0%.

[0100] 104. Based on the conversion rate corresponding to each account attribute tag in the first reference account set, determine the first set of accounts to be displayed, and send the display information of the target display object to the accounts in the first set of accounts to be displayed.

[0101] In implementation, based on the conversion rate of each account attribute tag corresponding to the obtained first reference account set, the account attribute tags can be arranged in descending order of conversion rate. Then, the account attribute tags are filtered according to preset rules, and the accounts corresponding to the filtered tags are grouped into a set to obtain the first set of accounts to be displayed. The accounts in this first set of accounts to be displayed are the accounts whose information will be pushed to them. Therefore, after determining the first set of accounts to be displayed, the display information of the target object is sent to the terminal where each account in the first set of accounts to be displayed is located. After receiving the display information of the target object, the terminal will display it to the user for viewing.

[0102] like Figure 2 As shown, Application A is a consumer application targeting a trial lesson of a children's English tutoring class. The advertiser of the trial lesson chooses to place it on a browser application. When it is determined to send it to a user's account, the advertisement for the trial lesson (i.e., the target audience's display information) is sent to the terminal where that account is located. When the user uses the browser application, they can see this advertisement at the top of the recommended information on the browser application's main page. The user can click the "Try Now" button, which will trigger the display of the advertisement. Figure 3 The children's English details page is shown. Users can then click the "Claim Now" button. At this point, two possibilities arise: one is that application A is not downloaded on the user's device, in which case clicking the "Claim Now" button will trigger a display like... Figure 4 The download page for application A, as shown, will trigger an event when a user downloads application A and then clicks the "Claim Now" button on the children's English details page. Figure 5 The page shown is for claiming a free trial lesson for children's English tutoring classes in application A. Users can claim a free trial lesson on this page. Alternatively, if application A is already downloaded on the user's device, clicking the "Claim Now" button will directly trigger the opening of the page. Figure 5 The image shows the page for claiming a trial lesson for a children's English tutoring class in application A. Once a user claims the trial lesson on this page, the account conversion is considered successful.

[0103] like Figure 6 As shown, advertisers can view the aggregation page for their ads in the browser application's data center. This aggregation page displays ad cards for trial lessons of children's English tutoring classes. Clicking on these ad cards will trigger the display of... Figure 7The page shown is the details page for the trial lesson of the children's English tutoring class. The details page displays real-time information such as current impressions, clicks, sales, conversion rate, and average order value. Advertisers can view the specific performance of their campaigns on this page.

[0104] Optionally, in step 102, after determining the first set of account attribute tags, it is necessary to determine the first set of reference accounts based on the account attribute tags included in the first set of account attribute tags. There are multiple methods for determining the first set of reference accounts, and the following are two of them:

[0105] The first type

[0106] The set of accounts corresponding to each account attribute label in the first account attribute label set is determined as the first reference account set.

[0107] In implementation, after obtaining the corresponding first set of account attribute tags based on at least one object attribute tag of the target display object, the accounts corresponding to each account attribute tag in the first set of account attribute tags can be directly obtained and combined into a set to obtain the first reference account set. It can be understood that the first reference account set is the set after deduplication of the set of accounts corresponding to each account attribute tag in the first set of account attribute tags; that is, the accounts in the first reference account set are all different.

[0108] The second type

[0109] A first preset number of accounts are randomly selected from the accounts corresponding to the first set of account attribute tags to form a second set of accounts to be displayed. Display information of the target object is sent to each account in the second set of accounts to be displayed, and operation information of each account in the second set of accounts to be displayed on the target object is obtained. Based on the operation information of each account in the second set of accounts to be displayed on the target object and the conversion rate threshold, a second set of account attribute tags is determined. A first set of reference accounts is determined based on the second set of account attribute tags.

[0110] In implementation, after obtaining the corresponding first account attribute tag set based on at least one object attribute tag of the target display object, a trial recommendation can be performed first to observe the effect of the trial recommendation. For example... Figure 8As shown, after determining the first set of account attribute tags, we can first obtain the accounts corresponding to each account attribute tag in the first set of account attribute tags, and then perform deduplication on these accounts to obtain an account set. From this account set, we randomly select a first preset number of accounts. This set of randomly selected first preset number of accounts constitutes the second set of accounts to be displayed. Then, we send the display information of the target display object to each account in the second set of accounts to be displayed for trial recommendation. When the recommendation duration reaches the second preset duration, we obtain the operation information of each account in the second set of accounts to be displayed regarding the target display object. The operation information can be whether the account successfully converted the target display object or failed to convert it within the second preset duration. For example, if the target display object is a product, the operation information could be placing an order for the product; if the target display object is a piece of news, the operation information could be the browsing time of the news content. Based on the operation information of each account to the target display object, it is determined whether each account has successfully converted. Then, the conversion rate corresponding to each account attribute tag of the second account to be displayed can be calculated. Based on the preset conversion rate threshold, these account attribute tags are filtered to obtain the second account attribute tag set. Then, based on the account corresponding to each account attribute tag in the second account attribute tag set, the first reference account set is obtained.

[0111] Based on the trial recommendations, a second set of account attribute tags with higher conversion rates is determined. Then, a first set of reference accounts is determined based on the second set of account attribute tags. This makes the accounts in the first set of reference accounts more likely to convert successfully, thereby increasing the likelihood of successful conversion for the accounts in the first set of accounts to be displayed determined based on the first set of reference accounts.

[0112] Optionally, the process of filtering the account attribute tags corresponding to the second set of accounts to be displayed based on a preset conversion rate threshold can be as follows:

[0113] Based on the operation information of each account in the second set of accounts to be displayed to the target display object and at least one account attribute tag, the conversion rate corresponding to each account attribute tag in the second set of accounts to be displayed is determined. From each account attribute tag in the second set of accounts to be displayed, the account attribute tags with a conversion rate greater than or equal to a conversion rate threshold are obtained and formed into a second set of account attribute tags.

[0114] In implementation, after obtaining the operation information of each account in the second set of accounts to be displayed regarding the target display object, the conversion rate corresponding to each account attribute tag in the second set of accounts to be displayed can be obtained based on the operation information of each account and at least one account attribute tag of that account. The method for calculating the conversion rate corresponding to each account attribute tag has been described in step 103 and will not be repeated here. Among the account attribute tags corresponding to each account in the second set of accounts to be displayed, the account attribute tags with a conversion rate greater than or equal to the conversion rate threshold are grouped into a set, which is the second account attribute tag set.

[0115] Optionally, there are several ways to determine the first set of reference accounts based on the second set of account attribute tags. The following are two possible methods:

[0116] The first type

[0117] A second preset number of accounts are randomly selected from the accounts corresponding to the second set of account attribute tags to form the first set of reference accounts.

[0118] In implementation, the accounts in the set corresponding to each account attribute tag in the second account attribute tag set are deduplicated. Then, a second preset number of accounts are randomly selected from the deduplicated set to form the first reference account set.

[0119] The second type

[0120] Based on the conversion rate corresponding to each account attribute tag in the second account attribute tag set, a weighted score is determined for each account attribute tag. Based on the weighted score for each account attribute tag, a total weighted score is determined for each account having at least one account attribute tag from the second account attribute tag set. A second preset number of accounts are selected in descending order of total weighted score to form a first reference account set.

[0121] In implementation, after determining the conversion rate corresponding to each account attribute tag in the second set of account attribute tags, a weighted score can be determined based on the conversion rate of each account attribute tag. The conversion rate of an account attribute tag can be directly determined as its weighted score. For example, if the conversion rate of account attribute tag E is 30% and the conversion rate of account attribute tag F is 65%, then the weighted score of account attribute tag E is 30 and the weighted score of account attribute tag F is 65. The weighted score corresponding to each account attribute tag in the second set of account attribute tags is determined using the above method.

[0122] For each account in the second set of account attribute labels, calculate its corresponding total weighted score. This involves processing each account as follows: determine the account attribute labels belonging to the second set of account attribute labels; determine the weighted score for each of these account attribute labels belonging to the second set; and then sum these weighted scores to obtain the total weighted score for that account. For example, if the account attribute labels belonging to the second set of account attribute labels are X and Y, and the weighted score for account attribute label X is 32 and the weighted score for account attribute label Y is 55, then the total weighted score for that account is 32 + 55 = 87.

[0123] Then, the accounts corresponding to the second account attribute tag set are arranged in descending order of their total weighted scores, and the first second preset number of accounts are selected to form a set, which is the first reference account set.

[0124] Optionally, the second preset number may be equal to or different from the first preset number, and can be set according to specific circumstances. This application embodiment does not limit this.

[0125] In this embodiment of the application, the corresponding processing for determining the conversion rate corresponding to each account attribute tag of the first reference account set based on at least one account attribute tag, at least one object attribute tag, and the conversion prediction model for each account in the first reference account set can be as follows:

[0126] Based on at least one account attribute tag, at least one object attribute tag, and a conversion prediction model for each account in the first reference account set, conversion prediction information for each account in the first reference account set is obtained. Based on the conversion prediction information and at least one account attribute tag for each account in the first reference account set, the conversion rate corresponding to each account attribute tag in the first reference account set is determined.

[0127] In implementation, the following operations can be performed on each account in the first set of reference accounts, such as... Figure 9As shown, a sorting order is determined for all account attribute tags and a sorting order for all object attribute tags. Then, in all the sequentially sorted account attribute tags, the account attribute tags that the account possesses are defined as 1, and the account attribute tags that the account does not possess are defined as 0, thus determining the first tag vector corresponding to an account. Similarly, in all the sequentially sorted object attribute tags, the object attribute tags that the target display object possesses are defined as 1, and the object attribute tags that it does not possess are defined as 0, thus determining the second tag vector of a target display object. The first tag vector of the account and the second tag vector of the target display object are input into the conversion prediction model to obtain the conversion prediction information for that account. The conversion prediction information for each account in the first reference account set is obtained in the above manner.

[0128] Account conversion prediction information can take two forms. The first is a conversion success or failure prediction. After inputting an account's attribute tag sequence and the target display object's object attribute tag sequence into the conversion prediction model, the model will output 0 or 1. An output of 0 indicates a conversion failure, and an output of 1 indicates a successful conversion. The second is a conversion probability value. After inputting an account's attribute tag sequence and the target display object's object attribute tag sequence into the conversion prediction model, the model will output a probability value. When this conversion probability value is greater than or equal to a preset conversion probability threshold, the account has successfully converted; when the conversion probability value is less than the preset threshold, the account has failed to convert.

[0129] Based on the conversion prediction information for each account in the first reference account set predicted by the conversion prediction model, it is determined whether the account has successfully converted or failed. Then, based on at least one account attribute tag for each account, the conversion rate corresponding to each account attribute tag in the first reference account set is determined.

[0130] In step 104, based on the conversion rate corresponding to each account attribute tag of the first reference account, the corresponding processing for the first set of accounts to be displayed can be determined as follows:

[0131] For each account attribute tag in the first reference account set, obtain the account attribute tags whose conversion rate is greater than or equal to the conversion rate threshold, and form a third account attribute tag set. Select a third preset number of accounts from the accounts corresponding to the third account attribute tag set to form a first set of accounts to be displayed.

[0132] In implementation, for each account attribute tag in the first reference account set, account attribute tags with conversion rates greater than or equal to a preset conversion rate threshold are grouped into a set, which is the third account attribute tag set. The account attribute tags in the third account attribute tag set are all account attribute tags with relatively high predicted conversion rates. Then, based on a preset third number, a third preset number of accounts are randomly selected from the accounts corresponding to the third account attribute tag set to form the first set of accounts to be displayed.

[0133] Optionally, after determining the third set of account attribute tags, the weighted score corresponding to each account attribute tag in the third set of account attribute tags can be determined based on the conversion rate corresponding to each account attribute tag in the third set of account attribute tags. Then, the total weighted score of the accounts with at least one account attribute tag in the third set of account attribute tags can be calculated. Then, the first three preset number of accounts can be obtained in descending order of total weighted score to form the first set of accounts to be displayed.

[0134] Optionally, the third preset number can be the same as the first preset number, or the same as the second preset number, or the same as both the first and second preset numbers, or different from both the first and second preset numbers. It can be set according to specific circumstances, and this application embodiment does not limit it.

[0135] Optionally, after step 104, this embodiment of the application can further select the set of accounts to be displayed in the next push of information. The recommended accounts can be filtered in real time based on the operation information of each account obtained after recommending to the first account to be displayed. Figure 10 As shown, the corresponding steps can be as follows:

[0136] 1001. Obtain the operation information of the accounts in the first set of accounts to be displayed on the target display object.

[0137] In implementation, after a second preset time has elapsed since the display information of the target object was sent to accounts in the first set of accounts to be displayed, the operation information of the accounts in the first set of accounts to be displayed on the target object is obtained. For example, if the target object is a product, the operation information of the account on the target object may include the account's order information for that product; if the target object is a piece of news, the operation information of the account on the target object may include the account's browsing information on the target object.

[0138] 1002. Based on the operation information and conversion rate threshold of the accounts in the first set of accounts to be displayed to the target display object, determine the fourth set of account attribute tags.

[0139] In implementation, based on the operation information of the accounts in the first set of accounts to be displayed to the target display object, it can be determined whether the accounts in the first set of accounts to be displayed have successfully converted to the target display object. Then, based on the conversion success or failure information for each account, the conversion rate corresponding to each account attribute tag in the account attribute tag corresponding to the first set of accounts to be displayed is calculated. Then, the account attribute tags with conversion rates greater than or less than a preset conversion rate threshold are grouped into a set, which is the fourth set of account attribute tags.

[0140] 1003. Determine the second reference account set based on the fourth account attribute tag set.

[0141] In implementation, after determining the fourth set of account attribute tags, the accounts corresponding to each account attribute tag in the fourth set of account attribute tags are grouped into a set. The accounts in the set are deduplicated. Then, a preset fourth preset number of accounts are randomly selected from the deduplicated set to form the second reference account set.

[0142] Optionally, the fourth preset number can be equal to any one of the first preset number, the second preset number, and the third preset number, or it can be different from all three preset numbers. It can be set according to the specific situation, and this application embodiment does not limit it.

[0143] 1004. Based on at least one account attribute label, at least one object attribute label, and the conversion prediction model for each account in the second reference account set, determine the conversion rate corresponding to each account attribute label in the second reference account set.

[0144] In implementation, after determining the second set of reference accounts, each account in the second set of reference accounts is processed as follows: the account's account attribute tag sequence and the target display object's object attribute tag sequence are input into the conversion prediction model, which outputs the conversion prediction information for that account regarding the target display object. Based on the above processing method, the conversion prediction information for each account in the second set of reference accounts regarding the target display object is obtained, thereby determining whether each account has successfully converted. Based on the conversion prediction information for each account in the second set of reference accounts, and at least one account attribute tag corresponding to each account, the conversion rate corresponding to each account attribute tag in the second set of reference accounts is obtained.

[0145] 1005. Based on the conversion rate of each account attribute tag in the second reference account set, determine the third set of accounts to be displayed, and send the display information of the target display object to the accounts in the third set of accounts to be displayed.

[0146] In implementation, the third set of accounts to be displayed can be formed by grouping account attribute tags whose conversion rates are greater than or equal to the conversion rate threshold from the account attribute tags corresponding to the second set of reference accounts. The third set of accounts to be displayed is obtained by further filtering based on the operation information of each account in the first set of accounts to be displayed, resulting in the set of accounts to be displayed to the target audience in the next push.

[0147] Send the display information of the target display object to the accounts in the third set of accounts to be displayed, thereby recommending the target display object to the users to which the accounts in the third set of accounts to be displayed belong.

[0148] After step 1005, after a second preset time has elapsed since the display information of the target display object was sent to the accounts in the third set of accounts to be displayed, the operation information of the accounts in the third set of accounts to be displayed regarding the target display object can be obtained. Then, based on this operation information and the conversion rate threshold, a fifth set of account attribute tags is determined. A third set of reference accounts is determined based on the fifth set of attribute tags. Based on at least one account attribute tag, at least one object attribute tag, and the conversion prediction model for each account in the third set of reference accounts, the conversion rate corresponding to each account attribute tag in the third set of reference accounts is determined. Based on the conversion rate corresponding to each account attribute tag in the third set of reference accounts, a fourth set of accounts to be displayed is determined, and the display information of the target display object is sent to the accounts in the fourth set of accounts to be displayed. This process can be repeated multiple times, from steps 1001 to 1005, to perform the next recommendation based on the previous set of accounts to be displayed, until all the recommendation share of the display information of the target display object has been sent.

[0149] The beneficial effects of the technical solution provided in this application are as follows: The solution mentioned in this application can first determine a first set of account attribute tags based on at least one object attribute tag of the target display object, and then determine a first set of reference accounts based on the correspondence between account attribute tags and accounts. Further refinement is then performed on the first set of reference accounts. First, based on the conversion prediction model, the conversion rate corresponding to each account attribute tag in the first set of reference accounts is predicted. Then, based on the conversion rate corresponding to the account attribute tag, the account to be displayed is determined, and the display information of the target display object is pushed to it. In this way, accounts corresponding to account attribute tags with higher conversion rates can be selected as the push targets, which can better improve the probability of successful conversion for each account, thereby improving the conversion rate of the target display object.

[0150] The conversion prediction model in this application embodiment is a machine learning model. Machine learning is a multidisciplinary field involving probability, statistics, approximation, convex analysis, algorithm complexity theory, and many other disciplines. It specifically studies how computers can simulate or implement human learning behavior to acquire new knowledge or skills and reorganize existing knowledge structures to continuously improve their performance. Machine learning is the core of artificial intelligence and the fundamental way to endow computers with intelligence; its applications span all areas of artificial intelligence. Machine learning and deep learning typically include techniques such as artificial neural networks, belief networks, reinforcement learning, transfer learning, inductive learning, and instructional learning.

[0151] The conversion prediction model needs to be trained before it can be used. The conversion prediction model used above is a pre-trained conversion prediction model. The training process for the conversion prediction model can be as follows:

[0152] The process involves obtaining the sample account attribute label sequence and the sample object attribute label sequence of the sample displayed object as input data for the conversion prediction model to be trained. The operational information of the sample account is then used to determine its conversion prediction information as baseline output data. This data is then fed into the conversion prediction model to obtain the actual output data of the sample account. The conversion prediction model is then tuned based on the baseline output data and the actual output data, thus completing one training cycle. This process is repeated multiple times with new sample account attribute label sequences and sample object attribute label sequences, until the loss value is less than a preset threshold, resulting in a successfully trained conversion prediction model.

[0153] This application provides an apparatus for sending display information. This apparatus can be the computer device described in the above embodiments, such as... Figure 11 As shown, the device includes:

[0154] The first acquisition module 1110 is used to acquire at least one object attribute label of the target display object;

[0155] The first determining module 1120 is used to determine the first account attribute tag set corresponding to the at least one object attribute tag, and to determine the first reference account set based on the first account attribute tag set.

[0156] The second determining module 1130 is used to determine the conversion rate corresponding to each account attribute tag of the first reference account set based on at least one account attribute tag of each account in the first reference account set, the at least one object attribute tag and the conversion prediction model.

[0157] The first sending module 1140 is used to determine the first set of accounts to be displayed based on the conversion rate corresponding to each account attribute tag of the first set of reference accounts, and to send the display information of the target display object to the accounts in the first set of accounts to be displayed.

[0158] In one possible design, the first determining module 1120 is configured to:

[0159] The set of accounts corresponding to each account attribute tag in the first account attribute tag set is determined as the first reference account set.

[0160] In one possible design, the first determining module 1120 is configured to:

[0161] A first preset number of accounts are randomly selected from the accounts corresponding to the first set of account attribute tags to form a second set of accounts to be displayed.

[0162] Send the display information of the target display object to each account in the second set of accounts to be displayed, and obtain the operation information of each account in the second set of accounts to be displayed on the target display object;

[0163] Based on the operation information and conversion rate threshold of each account in the second set of accounts to be displayed to the target display object, a second set of account attribute tags is determined;

[0164] The first reference account set is determined based on the second account attribute tag set.

[0165] In one possible design, the first determining module 1120 is configured to:

[0166] Based on the operation information of each account in the second set of accounts to be displayed to the target display object and at least one account attribute tag, the conversion rate corresponding to each account attribute tag in the second set of accounts to be displayed is determined;

[0167] In each account attribute tag corresponding to the second set of accounts to be displayed, obtain the account attribute tags whose conversion rate is greater than or equal to the conversion rate threshold, and form the second set of account attribute tags.

[0168] In one possible design, the first determining module 1120 is configured to:

[0169] A second preset number of accounts are randomly selected from the accounts corresponding to the second account attribute tag set to form a first reference account set.

[0170] In one possible design, the second determining module 1130 is used for:

[0171] Based on at least one account attribute label, at least one object attribute label, and the conversion prediction model for each account in the first reference account set, conversion prediction information for each account in the first reference account set is obtained.

[0172] Based on the conversion prediction information of each account in the first reference account set and at least one account attribute tag, the conversion rate corresponding to each account attribute tag in the first reference account set is determined.

[0173] In one possible design, the first transmitting module 1140 is used for:

[0174] In each account attribute tag corresponding to the first reference account set, obtain the account attribute tags whose conversion rate is greater than or equal to the conversion rate threshold, and form a third account attribute tag set.

[0175] A third preset number of accounts are randomly selected from the accounts corresponding to the third account attribute tag set to form the first set of accounts to be displayed.

[0176] In one possible design, the device further includes:

[0177] The second acquisition module is used to acquire operation information of the accounts in the first set of accounts to be displayed on the target display object;

[0178] The third determining module is used to determine the fourth set of account attribute tags based on the operation information and conversion rate threshold of the accounts in the first set of accounts to be displayed to the target display object;

[0179] The fourth determining module is used to determine the second reference account set based on the fourth account attribute tag set;

[0180] The fifth determining module is used to determine the conversion rate corresponding to each account attribute tag of the second reference account set based on at least one account attribute tag of each account in the second reference account set, the at least one object attribute tag, and the conversion prediction model.

[0181] The second sending module is used to determine the third set of accounts to be displayed based on the conversion rate corresponding to each account attribute tag in the second set of reference accounts, and to send the display information of the target display object to the accounts in the third set of accounts to be displayed.

[0182] It should be noted that the device for sending display information provided in the above embodiments is only illustrated by the division of the above functional modules. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. In addition, the device for sending display information and the method embodiment for sending display information provided in the above embodiments belong to the same concept, and the specific implementation process can be found in the method embodiment, which will not be repeated here.

[0183] Figure 12 This is a schematic diagram of a server structure provided in an embodiment of this application. The server 1200 can vary significantly due to different configurations or performance. It may include one or more processors 1201 and one or more memories 1202. The memories 1202 store at least one instruction, which is loaded and executed by the processors 1201 to implement the methods provided in the above-described method embodiments. Of course, the server may also have wired or wireless network interfaces, a keyboard, and input / output interfaces for input and output. The server may also include other components for implementing device functions, which will not be elaborated upon here.

[0184] In an exemplary embodiment, a computer-readable storage medium is also provided, such as a memory including instructions that can be executed by a processor in a terminal to complete the method of sending display information in the above embodiments. This computer-readable storage medium may be non-transitory. For example, the computer-readable storage medium may be a ROM, RAM, CD-ROM, magnetic tape, floppy disk, or optical data storage device, etc.

[0185] Those skilled in the art will understand that all or part of the steps of the above embodiments can be implemented by hardware or by a program instructing related hardware. The program can be stored in a computer-readable storage medium, such as a read-only memory, a disk, or an optical disk.

[0186] The above description is merely an optional embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A method for sending display information, characterized in that, The method includes: Retrieve at least one object attribute label of the target display object; Determine the first set of account attribute tags corresponding to the at least one object attribute tag; A first preset number of accounts are randomly selected from the accounts corresponding to the first set of account attribute tags to form a second set of accounts to be displayed. The display information of the target display object is sent to each account in the second set of accounts to be displayed. When the time after sending reaches the second preset time, the operation information of each account in the second set of accounts to be displayed on the target display object is obtained. The operation information is whether each account successfully or failed to convert the target display object within the time period corresponding to the second preset time. Based on the operation information and conversion rate threshold of each account in the second set of accounts to be displayed to the target display object, a second set of account attribute tags is determined; The first reference account set is determined based on the second account attribute tag set; Based on at least one account attribute tag for each account in the first reference account set, the at least one object attribute tag, and the conversion prediction model, the conversion rate corresponding to each account attribute tag in the first reference account set is determined. Based on the conversion rate corresponding to each account attribute tag in the first reference account set, a first set of accounts to be displayed is determined, and the display information of the target display object is sent to the accounts in the first set of accounts to be displayed.

2. The method according to claim 1, characterized in that, The step of determining the second set of account attribute tags based on the operation information and conversion rate threshold of each account in the second set of accounts to be displayed for the target display object includes: Based on the operation information of each account in the second set of accounts to be displayed to the target display object and at least one account attribute tag, the conversion rate corresponding to each account attribute tag in the second set of accounts to be displayed is determined; In each account attribute tag corresponding to the second set of accounts to be displayed, obtain the account attribute tags whose conversion rate is greater than or equal to the conversion rate threshold, and form the second set of account attribute tags.

3. The method according to claim 2, characterized in that, Determining the first reference account set based on the second account attribute tag set includes: Based on the conversion rate corresponding to each account attribute tag in the second set of account attribute tags, determine the weighted score corresponding to each account attribute tag; Based on the weighted score corresponding to each account attribute tag, determine the total weighted score for each account that has at least one account attribute tag from the second account attribute tag set; A second preset number of accounts are selected in descending order of total weighted score to form the first reference account set.

4. The method according to claim 1, characterized in that, The step of determining the conversion rate corresponding to each account attribute tag in the first reference account set based on at least one account attribute tag of each account in the first reference account set, the at least one object attribute tag, and the conversion prediction model includes: Based on at least one account attribute label, at least one object attribute label, and the conversion prediction model for each account in the first reference account set, conversion prediction information for each account in the first reference account set is obtained. Based on the conversion prediction information of each account in the first reference account set and at least one account attribute tag, the conversion rate corresponding to each account attribute tag in the first reference account set is determined.

5. The method according to claim 1, characterized in that, The step of determining the first set of accounts to be displayed based on the conversion rate corresponding to each account attribute tag in the first set of reference accounts includes: In each account attribute tag corresponding to the first reference account set, obtain the account attribute tags whose conversion rate is greater than or equal to the conversion rate threshold, and form a third account attribute tag set. A third preset number of accounts are selected from the accounts corresponding to the third account attribute tag set to form the first set of accounts to be displayed.

6. The method according to claim 1, characterized in that, After sending the display information of the target display object to the accounts in the first set of accounts to be displayed, the method further includes: Obtain the operation information of the accounts in the first set of accounts to be displayed on the target display object; Based on the operation information and conversion rate threshold of the accounts in the first set of accounts to be displayed to the target display object, a fourth set of account attribute tags is determined; The second reference account set is determined based on the fourth account attribute tag set; Based on at least one account attribute tag, at least one object attribute tag, and the conversion prediction model for each account in the second reference account set, the conversion rate corresponding to each account attribute tag in the second reference account set is determined. Based on the conversion rate corresponding to each account attribute tag in the second reference account set, a third set of accounts to be displayed is determined, and the display information of the target display object is sent to the accounts in the third set of accounts to be displayed.

7. A device for transmitting display information, characterized in that, The device includes: The acquisition module is used to acquire at least one object attribute tag of the target display object; A first determining module is configured to: determine a first set of account attribute tags corresponding to the at least one object attribute tag; randomly select a first preset number of accounts from the accounts corresponding to the first set of account attribute tags to form a second set of accounts to be displayed; send display information of the target display object to each account in the second set of accounts to be displayed; when the time elapsed after sending reaches a second preset time elapsed, obtain operation information of each account in the second set of accounts to be displayed on the target display object, wherein the operation information indicates whether each account successfully or failed to convert the target display object within the time period corresponding to the second preset time elapsed; determine a second set of account attribute tags based on the operation information of each account in the second set of accounts to be displayed on the target display object and a conversion rate threshold; and determine a first set of reference accounts based on the second set of account attribute tags. The second determining module is used to determine the conversion rate corresponding to each account attribute tag of the first reference account set based on at least one account attribute tag of each account in the first reference account set, the at least one object attribute tag and the conversion prediction model. The sending module is used to determine a first set of accounts to be displayed based on the conversion rate corresponding to each account attribute tag in the first set of reference accounts, and to send the display information of the target display object to the accounts in the first set of accounts to be displayed.

8. A computer device, characterized in that, The computer device includes a processor and a memory, the memory storing at least one instruction, which is loaded and executed by the processor to perform the operation performed by the method of sending display information as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The storage medium stores at least one instruction, which is loaded and executed by a processor to perform the operation of the method for sending display information as described in any one of claims 1 to 6.

Images