Method, device and equipment for determining a delivery population for targeted content
By acquiring and analyzing exposure information and user profiles of exposed users, the target audience for targeted content can be determined, solving the problem of inaccurate ad delivery for merchants and achieving higher accuracy and effectiveness in ad delivery.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- TENCENT TECHNOLOGY (SHENZHEN) CO LTD
- Filing Date
- 2021-03-12
- Publication Date
- 2026-06-02
AI Technical Summary
In existing technologies, the user identification for merchant advertising pushes is not accurate enough, resulting in insufficient accuracy in advertising pushes.
By acquiring the exposure users of the related targeted content of the content to be targeted, a candidate set is constructed. Based on the exposure information and user profile, the targeting score is determined, and target exposure users who meet the conditions are selected to be added to the targeted audience.
It improves the accuracy of the predicted results for users to be exposed to targeted content, ensures the accuracy of the targeted audience, and improves the effectiveness of advertising.
Smart Images
Figure CN115082099B_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of artificial intelligence technology, and in particular to a method, apparatus, and device for determining the target audience for targeted content. Background Technology
[0002] Currently, merchants use media platforms to push advertisements to users as a supplement when selling products.
[0003] In related technologies, in order to ensure the accuracy of merchant advertising push, before pushing the target advertisement, the corresponding push users are determined from big data based on the click data of the merchant's other advertisements, and then the target advertisement is pushed to the push users.
[0004] However, in the aforementioned technologies, determining the target users for a campaign ad solely based on click data from other merchant ads is not accurate enough. Summary of the Invention
[0005] This application provides a method, apparatus, and device for determining the target audience for targeted content, which can ensure the accuracy of the estimated results of exposure users for the targeted content to be delivered, thereby ensuring the accuracy of the targeted audience. The technical solution is as follows:
[0006] According to one aspect of the embodiments of this application, a method for determining the target audience for targeted content is provided, the method comprising:
[0007] Obtain multiple exposure users of the related targeted content of the content to be targeted, and construct a candidate set;
[0008] Based on the exposure information and user profile of the exposed users in the candidate set, the targeting score of the content to be delivered is determined relative to the exposed users; wherein, the exposure information is used to indicate the historical browsing behavior of the exposed users towards the related targeting content, and the targeting score is used to indicate the estimated browsing behavior of the exposed users towards the targeting content to be delivered;
[0009] Select target exposure users whose delivery scores meet the criteria from the candidate set and add them to the targeted audience for the content to be delivered.
[0010] According to one aspect of the embodiments of this application, an apparatus for determining the target audience for targeted content is provided, the apparatus comprising:
[0011] The candidate acquisition module is used to acquire multiple exposure users of the related targeted content of the content to be delivered, and construct a candidate set;
[0012] The score determination module is used to determine the delivery score of the targeted content to be delivered relative to the exposed users based on the exposure information and user profile of the exposed users in the candidate set; wherein, the exposure information is used to indicate the historical browsing behavior of the exposed users towards the related targeted content, and the delivery score is used to indicate the estimated browsing behavior of the exposed users towards the targeted content to be delivered;
[0013] The user identification module is used to select target exposure users whose delivery scores meet the conditions from the candidate set and add them to the targeted audience of the content to be delivered.
[0014] According to one aspect of the embodiments of this application, a computer device is provided, the computer device including a processor and a memory, the memory storing at least one instruction, at least one program, code set or instruction set, the at least one instruction, the at least one program, the code set or instruction set being loaded and executed by the processor to implement the above-described method for determining the target audience for targeted content.
[0015] According to one aspect of the embodiments of this application, a computer-readable storage medium is provided, wherein at least one instruction, at least one program, code set, or instruction set is stored in the computer-readable storage medium, wherein the at least one instruction, the at least one program, the code set, or the instruction set is loaded and executed by a processor to implement the above-described method for determining the target audience for targeted content.
[0016] According to one aspect of the embodiments of this application, a computer program product or computer program is provided, which includes computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the aforementioned method for determining the target audience for targeted content.
[0017] The technical solutions provided in this application embodiment may have the following beneficial effects:
[0018] By leveraging exposure information and user profiles of exposed users, the system can predict their browsing activity for targeted content. Exposure information indicates the historical browsing history of exposed users for related targeted content. Combining this exposure information with user profiles ensures the accuracy of the predicted browsing activity, thereby guaranteeing the accuracy of the targeted audience. Furthermore, since exposed users are those who have viewed related targeted content, selecting the target audience from among exposed users effectively ensures the delivery effectiveness of the targeted content. Attached Figure Description
[0019] 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.
[0020] Figure 1 An exemplary diagram illustrates a process for determining the target audience for targeted content.
[0021] Figure 2 This is a flowchart illustrating a method for determining the target audience for targeted content, provided in one embodiment of this application.
[0022] Figure 3 An exemplary schematic diagram of a user interface is shown;
[0023] Figure 4 An exemplary diagram illustrates an update control for targeted audience delivery;
[0024] Figure 5 An exemplary diagram illustrates another method for determining the target audience for targeted content.
[0025] Figure 6 This is a block diagram of a device for determining the target audience for targeted content, provided in one embodiment of this application;
[0026] Figure 7 This is a block diagram of a device for determining the target audience for targeted content, provided in another embodiment of this application;
[0027] Figure 8 This is a schematic diagram of a computer device structure provided in one embodiment of this application. Detailed Implementation
[0028] 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.
[0029] Artificial intelligence (AI) is the theory, methods, technology, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to achieve optimal results. In other words, AI is a comprehensive technology within computer science that attempts to understand the essence of intelligence and produce a new kind of intelligent machine that can react in a way similar to human intelligence. AI studies the design principles and implementation methods of various intelligent machines, enabling them to possess the functions of perception, reasoning, and decision-making.
[0030] Artificial intelligence (AI) is a comprehensive discipline encompassing a wide range of fields, including both hardware and software technologies. Fundamental AI technologies generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing, operating / interactive systems, and mechatronics. AI software technologies primarily include computer vision, speech processing, natural language processing, and machine learning / deep learning.
[0031] Machine learning (ML) is a multidisciplinary field involving probability theory, statistics, approximation theory, convex analysis, and algorithm complexity theory. 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 learn-by-doing.
[0032] With the research and advancement of artificial intelligence (AI) technology, AI is being studied and applied in various fields, such as smart homes, smart wearable devices, virtual assistants, smart speakers, smart marketing, autonomous driving, drones, robots, smart healthcare, and smart customer service. It is believed that with the development of technology, AI will be applied in more fields and play an increasingly important role.
[0033] The solution provided in this application involves technologies such as machine learning in artificial intelligence. A conversion score prediction model is trained using machine learning techniques. This model is then used to predict the conversion score of the targeted content relative to the exposed users based on their exposure information and user profiles. The exposure information indicates the historical browsing history of the exposed users regarding related targeted content, which refers to targeted content related to the content to be targeted. The conversion score indicates the estimated probability that the exposed users will browse the targeted content and perform a related operation, which refers to an operation related to the targeted content. Optionally, when acquiring the exposed users, feature information of a third candidate related content is obtained using computer vision and natural language processing techniques. Based on this feature information, targeted content similar to the content to be targeted is selected from the third candidate related content as related targeted content. The exposed users are then determined based on the exposed users of this related targeted content.
[0034] For example, in conjunction with the reference Figure 1This paper briefly introduces the method for determining the target audience for targeted content in this application. First, multiple users who have been exposed to the targeted content are obtained. Optionally, these users are obtained through related targeted content. Further, it is determined whether the number of these multiple users exceeds a threshold. If the number of users exceeds the threshold, a conversion score prediction model is used to predict the exposure information and user profiles of the users, obtaining their conversion scores. Based on these conversion scores, users with conversion scores greater than the target value are selected as the target audience for the targeted content. If the number of users is less than or equal to the threshold, the related targeted content is expanded, and the expanded related targeted content is used to re-obtain users who have been exposed to the targeted content.
[0035] Optionally, the various pre-stored data involved in this application (such as information about each user, information about each targeted content, etc.) can be stored on the blockchain.
[0036] The technical solution of this application will be described below with reference to several embodiments.
[0037] For ease of explanation, the following method embodiments only use a computer device as the execution subject of each step as an example. This computer device can be any electronic device with computing and storage capabilities. Exemplarily, the computer device can be a server, which can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN (Content Delivery Network), and big data and artificial intelligence platforms. It should be noted that in this embodiment, the execution subject of each step can be the same computer device, or it can be executed by multiple different computer devices interacting and cooperating; this is not limited here.
[0038] Please refer to Figure 2 The diagram illustrates a flowchart of a method for determining the target audience for targeted content, provided in one embodiment of this application. The method may include the following steps (201-203):
[0039] Step 201: Obtain multiple exposure users of the related targeted content of the content to be targeted, and construct a candidate set.
[0040] Targeted content refers to content that is specifically targeted. Optionally, this targeting may be relative to a specific group of people; or, it may be relative to a particular field; and so on. Targeted content to be delivered refers to targeted content that has not yet been delivered. This targeted content to be delivered can be any content with browsing functionality, such as advertisements, videos, articles, news, etc.
[0041] In this embodiment of the application, before the targeted content to be delivered is delivered, the computer device obtains multiple exposure users of the related targeted content of the targeted content to be delivered, and constructs a candidate set. Here, related targeted content refers to targeted content that is associated with the targeted content to be delivered.
[0042] In one possible implementation, the aforementioned associated targeted content and the content to be targeted belong to the same advertiser. Optionally, when acquiring the aforementioned associated targeted content, the computer device determines the advertiser of the content to be targeted and acquires the advertiser's previously delivered targeted content, and then acquires the aforementioned associated targeted content based on the previously delivered targeted content. Optionally, the computer device directly determines the previously delivered targeted content as the aforementioned associated targeted content; or, the computer device selects content from the previously delivered targeted content that meets the associated content filtering criteria as the aforementioned associated targeted content. For example, the aforementioned associated content filtering criteria include content delivered within the last 3 days, content similar to the content to be targeted, and content with more than a certain number of views, etc.
[0043] In another possible implementation, the aforementioned related targeted content and the target content do not belong to the same advertiser. Optionally, when acquiring the aforementioned related targeted content, the computer device determines the feature information of the target content and selects similar already-delivered targeted content based on the feature information, and then acquires the aforementioned related targeted content based on the already-delivered targeted content. Optionally, the computer device directly determines the already-delivered targeted content as the aforementioned related targeted content; or, the computer device selects content that meets the related content filtering criteria from the already-delivered targeted content as the aforementioned related targeted content. For example, the aforementioned related content filtering criteria include content delivered in the last 3 days, content similar to the target content, and content with more than a certain number of views.
[0044] Of course, in practical applications, the filtering conditions for related content can be flexibly set and adjusted according to the actual situation, and this application embodiment does not limit this.
[0045] Optionally, the candidate set may include all or some of the aforementioned exposed users.
[0046] In one possible implementation, the candidate set includes all users of the aforementioned plurality of exposed users. Optionally, after obtaining the plurality of exposed users of the aforementioned associated targeted content, the computer device directly constructs the aforementioned candidate set from the plurality of exposed users.
[0047] In another possible implementation, the candidate set includes a subset of the aforementioned multiple exposed users. Optionally, after acquiring the multiple exposed users of the aforementioned targeted content, the computer device selects exposed users who meet the candidate user filtering criteria from the multiple exposed users to construct the aforementioned candidate set. For example, the candidate user filtering criteria may be, for instance, the number of times the targeted content has been viewed exceeds a certain value, the targeted content has been viewed within the last 3 days, or the viewing time of the targeted content exceeds a certain target value, etc.
[0048] It should be noted that, in this embodiment, the computer device can construct the aforementioned candidate set for the targeted content after receiving a delivery instruction; or, the computer device can construct the aforementioned candidate set after receiving a targeted audience update instruction. The targeted audience update instruction can be directed to the targeted content itself or to the advertiser of the targeted content. In other words, the advertiser can update their targeted audience based on actual circumstances, and then use the latest updated targeted audience as the target audience for the next delivery of targeted content.
[0049] For example, such as Figure 3 As shown, the user interface 30 includes a targeted audience update method 31. After detecting the advertiser's selection of update method 31, the aforementioned targeted audience update instruction is generated. Alternatively, as... Figure 4 As shown, when a selection operation is detected for the expansion selection button 41, the above-mentioned targeted audience update instruction is generated.
[0050] Step 202: Based on the exposure information and user profiles of the candidate users, determine the delivery score of the targeted content to be delivered relative to the exposed users.
[0051] In this embodiment of the application, after obtaining the above-mentioned candidate set, the computer device determines the delivery score of the targeted content to be delivered relative to the exposed users based on the exposure information and user profile of the exposed users in the candidate set.
[0052] The aforementioned exposure information is used to indicate the historical browsing history of exposed users for the relevant targeted content. Optionally, this exposure information includes the number of times the exposed user has historically viewed the relevant targeted content, and the browsing time corresponding to each view. The browsing time can also be referred to as the exposure time. In one possible implementation, when the computer device obtains multiple exposed users for the aforementioned relevant targeted content, it obtains the exposure information of each exposed user to ensure the timeliness of subsequent score acquisition. In another possible implementation, to avoid unnecessary resource overhead, after obtaining the candidate set, the computer device determines the exposed users in the candidate set and then obtains the exposure information of the exposed users in the candidate set.
[0053] The aforementioned user profile is used to reflect the representative characteristics of users who have been exposed to the content. This user profile is obtained by a computer device through the behavioral data of the exposed users, which records the actions taken by the exposed users towards the task content at any given time. Optionally, the computer device obtains the user profile when acquiring multiple exposed users of the aforementioned associated targeted content; or, after acquiring a candidate set, the computer device acquires the user profiles of the exposed users within the candidate set.
[0054] The aforementioned performance score indicates the estimated viewing activity of exposed users for the targeted content. Viewing activity can be either clicks or conversions. Conversions refer to exposed users clicking on the targeted content and performing related actions. Optionally, advertisers can configure viewing metrics based on actual circumstances. For example, such as... Figure 3 As shown, if the advertiser needs to optimize the click-through rate of the targeted content, select "Click-through rate" in update method 31; if the advertiser needs to optimize the conversion rate of the targeted content, select "Conversion rate" in update method 31.
[0055] It should be noted that different targeted content corresponds to different associated operations. For example, if the targeted content is an advertisement, the associated operation is to purchase the product corresponding to the advertisement; if the targeted content is a story-based video series, the associated operation is to click on the next episode of the story; if the targeted content is news information, the associated operation is to post a comment on the news information; and so on. Of course, in practical applications, the associated operations of targeted content can be flexibly set and adjusted according to the actual situation. One targeted content can correspond to one or more associated operations, and this application embodiment does not limit this.
[0056] Optionally, in this embodiment of the application, the exposure information and user profile of the exposed user are information that is pre-acquired by a computer device and stored in a specific database.
[0057] Step 203: Select target exposure users whose delivery scores meet the criteria from the candidate set and add them to the targeted audience for the targeted content to be delivered.
[0058] In this embodiment, after obtaining the aforementioned delivery score, the computer device selects target exposure users whose delivery scores meet certain conditions from the candidate set based on the delivery score, and adds these target exposure users to the targeted audience for the content to be delivered. Here, the aforementioned conditions refer to the criteria for determining exposure users.
[0059] In one possible implementation, the condition is that the delivery score is greater than the delivery threshold. Optionally, after obtaining the delivery score, the computer device compares the delivery score with the delivery threshold. If the delivery score is greater than the delivery threshold, the user corresponding to the delivery score is determined to be a target user; if the delivery score is less than or equal to the delivery threshold, the user corresponding to the delivery score is determined not to be a target user. It should be noted that the delivery threshold can be any value, and this embodiment does not limit it.
[0060] In another possible implementation, the condition described above is that the ranking of the ad placement score is greater than a ranking threshold. Optionally, after obtaining the ad placement scores, the computer device sorts the scores from largest to smallest and selects the users whose ranking is higher than the ranking threshold as target users. It should be noted that the ranking threshold is determined by the number of targeted users specified by the advertiser, and this ranking threshold ensures that the number of targeted users ultimately obtained by the computer device is greater than or equal to the number of targeted users specified by the advertiser.
[0061] In summary, the technical solution provided in this application estimates the browsing activity of exposed users for targeted content by using their exposure information and user profiles. The exposure information indicates the historical browsing activity of exposed users for related targeted content. By combining the exposure information with the user profile to estimate browsing activity, the accuracy of the estimation results for exposed users regarding the targeted content is ensured, thereby guaranteeing the accuracy of the targeted audience. Furthermore, since exposed users are those who have browsed the related targeted content of the users to be targeted, selecting the targeted audience from exposed users effectively guarantees the delivery effect of the targeted content.
[0062] The following section describes how the candidate set was obtained.
[0063] In an exemplary embodiment, step 201 above includes the following steps:
[0064] 1. Obtain the first related users of the targeted content to be delivered.
[0065] In this embodiment, when acquiring the exposure users of associated targeted content, the computer device acquires the exposure users of the first associated targeted content of the targeted content to be delivered. The first associated targeted content refers to targeted content belonging to the same advertiser as the targeted content to be delivered. Optionally, before delivering the targeted content to be delivered, the computer device acquires the identifier information of the advertiser of the targeted content to be delivered, and based on the identifier information, acquires the targeted content already delivered by the advertiser, acquires the aforementioned first associated targeted content, and then acquires the exposure users of the first associated targeted content. It should be noted that one advertiser can correspond to one or more identifiers. Optionally, an advertiser can use different identifiers to deliver different types of targeted content.
[0066] Optionally, when acquiring candidate users of the aforementioned first associated targeted content, the computer device acquires the exposure users of the first candidate associated targeted content of the targeted content to be delivered. Here, the first candidate associated targeted content refers to the targeted content delivered by the advertiser of the targeted content to be delivered within a first historical time period. Further, the number of exposure users of the first candidate associated content is detected. If the number of exposure users of the first candidate associated targeted content meets a second requirement, then the exposure users of the first associated targeted content are determined based on the exposure users of the first candidate associated targeted content; if the number of exposure users of the first candidate associated targeted content does not meet the second requirement, then the exposure users of the second candidate associated targeted content of the targeted content to be delivered are acquired, and the exposure users of the first associated targeted content are determined based on the exposure users of the first candidate associated targeted content and the second candidate associated targeted content. Here, the second candidate associated targeted content refers to the targeted content delivered by the advertiser of the targeted content to be delivered within a second historical time period.
[0067] Optionally, the second historical time period precedes the first historical time period. That is, when acquiring relevant targeted content, the computer device first acquires the first candidate relevant targeted content within the first historical time period closest to the current time. Then, if the number of users exposed to the first candidate relevant targeted content does not meet the second requirement, it acquires the second candidate relevant targeted content within the second historical time period farther from the current time. The time ranges of the first and second historical time periods do not overlap. Optionally, when determining the second historical time period, the computer device can base its determination on the difference between the number of users exposed to the first candidate relevant targeted content and the second requirement, combined with the content delivery frequency of the advertiser of the targeted content to be delivered, to avoid unnecessary overhead caused by an excessively long second historical time period.
[0068] The second requirement mentioned above refers to the requirement for determining the number of users exposed. Optionally, the second requirement is that the number of users exposed is greater than a quantity determination threshold. After obtaining the number of users exposed to the first candidate related targeted content, the computer device compares the number of users exposed with the quantity determination threshold. If the number of users exposed is greater than the quantity determination threshold, then the users exposed to the first related targeted content are determined based on the users exposed to the first candidate related targeted content; if the number of users exposed is less than or equal to the quantity determination threshold, then the second candidate related targeted content is obtained, and the users exposed to the first related targeted content are determined based on the users exposed to the first candidate related targeted content and the users exposed to the second candidate related targeted content.
[0069] Optionally, when determining the users exposed to the targeted content based on the exposure users of the first candidate related targeted content (or the exposure users of the first candidate related targeted content and the second candidate related targeted content), the computer device may directly determine the exposure users of the first candidate related targeted content (or the exposure users of the first candidate related targeted content and the second candidate related targeted content) as the exposure users of the targeted content; or, it may filter the exposure users of the first candidate related targeted content (or the exposure users of the first candidate related targeted content and the second candidate related targeted content) according to a first filtering condition to determine the exposure users of the targeted content. For example, the first filtering condition may be, for instance, the number of times the targeted content has been viewed exceeds a certain value, the targeted content has been viewed within the last 3 days, or the viewing time for the targeted content exceeds a certain target value.
[0070] 2. If the number of users exposed to the first associated targeted content meets the first requirement, then a candidate set is constructed based on the users exposed to the first associated targeted content.
[0071] In this embodiment of the application, after obtaining the users who have been exposed to the first associated targeted content, the computer device detects the number of users who have been exposed to the first associated targeted content. If the number of users who have been exposed to the first associated targeted content meets a first requirement, a candidate set is constructed based on the users who have been exposed to the first associated targeted content.
[0072] The aforementioned first requirement refers to the requirement for determining the number of users exposed. Optionally, the first requirement is that the number of users exposed is greater than a quantity determination threshold. After obtaining the number of users exposed to the aforementioned first associated targeted content, the computer device compares the number of users exposed with the quantity determination threshold. If the number of users exposed is greater than the quantity determination threshold, a candidate set is constructed based on the users exposed to the first candidate associated targeted content.
[0073] It should be noted that the second requirement described above may be the same as or different from the first requirement described above, and this application embodiment does not limit this. Optionally, if the second requirement is different from the first requirement, then the difficulty of achieving the second requirement is lower than the difficulty of achieving the first requirement.
[0074] 3. If the number of users exposed to the first associated targeted content does not meet the first requirement, then obtain the users exposed to the second associated targeted content of the targeted content to be delivered; construct a candidate set based on the users exposed to the first associated targeted content and the users exposed to the second associated targeted content.
[0075] Optionally, in this embodiment, if the number of users exposed to the first associated targeted content does not meet the first requirement, then the users exposed to the second associated targeted content of the targeted content to be delivered are obtained, and a candidate set is constructed based on the users exposed to the first associated targeted content and the second associated targeted content. The second associated targeted content refers to targeted content that is similar to the targeted content to be delivered.
[0076] Optionally, when acquiring the exposure users of the aforementioned second related targeted content, the computer device acquires the tag information corresponding to the advertiser of the targeted content to be delivered. This tag information indicates the industry sector of the advertiser of the targeted content to be delivered. Further, other advertisers with the same tag information as the advertiser of the targeted content to be delivered are identified, and a third candidate related targeted content is determined based on the already delivered targeted content of these other advertisers. Then, based on the similarity between the third candidate related targeted content and the targeted content to be delivered, the second related targeted content is selected from the third candidate related targeted content, thereby acquiring the exposure users of the second related targeted content.
[0077] Optionally, when acquiring the second related targeted content, the computer device acquires first feature information and second feature information. The first feature information refers to the feature information of the candidate related targeted content, and the second feature information refers to the feature information of the targeted content already delivered by the advertiser of the targeted content to be delivered. Further, the similarity between the first feature information and the second feature information is acquired, and candidate related targeted content whose similarity meets the target condition is determined as the second related targeted content. Optionally, in this embodiment, the feature information of the targeted content is the image and text feature information of the targeted content. The aforementioned target condition refers to the similarity evaluation condition between different targeted content. For example, the target condition could be a similarity less than a certain threshold, a similarity ranking within the top n, etc.
[0078] The following section explains how the scores for the above-mentioned placements are obtained.
[0079] In an exemplary embodiment, step 202 above includes the following steps:
[0080] 1. Determine the exposure intensity of users based on their exposure information.
[0081] Exposure information is used to indicate the historical browsing history of exposed users for related targeted content. Optionally, this exposure information includes the number of times the exposed user has historically browsed the related targeted content, as well as the browsing time corresponding to each browsing. The browsing time can also be referred to as the exposure time.
[0082] In this embodiment of the application, after obtaining the aforementioned candidate set, the computer device determines the exposure intensity of the exposed users based on the exposure information of the exposed users in the candidate set. Here, exposure intensity refers to the browsing intensity of the exposed user towards the associated targeted content.
[0083] Optionally, when acquiring the exposure intensity of an exposed user, the computer device determines the difference between the exposure time and the current time based on the exposure time in the exposure information. Here, the exposure time refers to the moment the exposed user browses the associated targeted content. Further, the computer device obtains the single exposure intensity value corresponding to the exposed user based on this difference, and sums up the individual exposure intensity values to obtain the exposure intensity of the exposed user. Here, there is a positive correlation between the single exposure intensity value and the difference.
[0084] For example, suppose the exposure time corresponding to the i-th view of the relevant targeted content by the exposed user is t. i Then the exposure intensity s of the user is:
[0085] ;
[0086] Where e is the natural logarithm, t cur The current time is given, and 'a' is the time decay factor, which can optionally be 0.5.
[0087] It should be noted that, in this embodiment, after obtaining the exposure intensity, the computer device processes the exposure intensity according to the user corresponding to that exposure intensity. Optionally, if the user belongs to a first-class user, the exposure intensity is determined as the final exposure intensity; if the user belongs to a second-class user, the exposure intensity is attenuated to obtain the final exposure intensity. Here, the first-class user refers to the user identified based on the already delivered targeted content of the advertiser of the content to be targeted, and the second-class user refers to the user identified based on targeted content that is similar to the content to be targeted. The final exposure intensity is used to combine with the user profile to generate feature information of the user.
[0088] Optionally, in this embodiment of the application, the computer device can multiply the exposure intensity by a factor to obtain the final exposure intensity when performing exposure intensity attenuation processing. For example, the factor is 0.6.
[0089] 2. Based on the exposure intensity and user profile of the exposed users, generate the characteristic information of the exposed users.
[0090] In this embodiment, after acquiring the exposure intensity of the user being exposed, the computer device generates feature information of the user based on the exposure intensity and user profile. Optionally, the computer device can stitch together the exposure intensity and user profile of the user to obtain the feature information of the user being exposed.
[0091] 3. Based on the characteristic information of the exposed users, determine the delivery score of the targeted content relative to the exposed users.
[0092] In this embodiment, after acquiring the feature information of the exposed user, the computer device determines the targeting score of the content to be delivered relative to the exposed user based on the feature information. Optionally, the targeting score includes a conversion score, which is used to indicate the estimated probability that the exposed user will browse the targeting content and perform an associated operation.
[0093] In one possible implementation, when the computer device obtains the conversion score, it performs a prediction process based on the characteristic information of the exposed users through a conversion score prediction model to obtain the conversion score of the targeted content to be delivered relative to the exposed users.
[0094] In another possible implementation, when obtaining the conversion score, the computer device performs prediction processing based on the feature information of the exposed user using a click-through rate prediction model to obtain the predicted click-through rate of the targeted content relative to the exposed user; and performs prediction processing based on the feature information of the exposed user using a conversion score prediction model to obtain the predicted conversion score of the targeted content relative to the exposed user. The predicted click-through rate indicates the estimated probability that the exposed user will browse the targeted content, and the predicted conversion score indicates the estimated probability that the exposed user will execute the associated operation. Further, the computer device determines the conversion score of the targeted content relative to the exposed user based on the predicted click-through rate and the predicted conversion score. Optionally, the computer device multiplies the predicted click-through rate and the predicted conversion score to obtain the conversion score.
[0095] Optionally, in this embodiment of the application, the training process of the above-mentioned click score prediction model is as follows:
[0096] 1. Obtain the exposure information and user profiles of users who have been exposed to the targeted content specified by the advertiser of the targeted content to be delivered, as the second positive sample; wherein, the targeted content refers to the targeted content already delivered by other advertisers specified by the advertiser of the targeted content to be delivered;
[0097] 2. Obtain exposure information and user profiles of unexposed users of the specified targeted content from the candidate set as a second negative sample;
[0098] 3. The click score prediction model is trained using the second positive sample and the second negative sample.
[0099] Optionally, in this embodiment of the application, the training process of the above-mentioned conversion score prediction model is as follows:
[0100] 1. Obtain the exposure information and user profile of the converted users corresponding to the targeted content already delivered by the advertiser of the targeted content to be delivered, as well as the exposure information and user profile of the seed users specified by the advertiser of the targeted content to be delivered, as the first positive sample; wherein, the converted user refers to the user who has viewed the targeted content and performed the associated operation;
[0101] 2. Obtain the exposure information and user profiles of unconverted users from the candidate set as the first negative sample;
[0102] 3. The conversion score prediction model is trained using the first positive sample and the first negative sample.
[0103] Below, in conjunction with references Figure 5 This application provides a complete description of the methods used to determine the target audience for targeted content.
[0104] Step 501: Obtain the first associated targeting content of the content to be targeted.
[0105] Step 502: Determine whether the number of users exposed to the first associated targeted content meets the first requirement. If the number of users exposed to the first associated targeted content meets the first requirement, proceed to step 505; if the number of users exposed to the first associated targeted content does not meet the first requirement, proceed to step 503.
[0106] Step 503: Expand the first associated targeted content to obtain the second associated targeted content.
[0107] Step 504: Based on the users exposed to the first associated targeted content and the users exposed to the second associated targeted content, determine the users exposed to the associated targeted content.
[0108] Step 505: Based on the users exposed to the first related targeted content, determine the users exposed to the related targeted content.
[0109] Step 506: Construct a candidate set based on users exposed to the relevant targeted content.
[0110] Step 507: Based on the exposure information and user profiles of the candidate users, determine the delivery score of the targeted content to be delivered relative to the exposed users.
[0111] Step 508: Select target exposure users whose delivery scores meet the criteria from the candidate set and add them to the targeted audience for the targeted content to be delivered.
[0112] The following are embodiments of the apparatus described in this application, which can be used to execute the embodiments of the method described in this application. For details not disclosed in the apparatus embodiments of this application, please refer to the embodiments of the method described in this application.
[0113] Please refer to Figure 6 This diagram illustrates a block diagram of an audience determination device for targeted content according to an embodiment of this application. The device has the function of implementing the aforementioned audience determination method for targeted content; this function can be implemented in hardware or by hardware executing corresponding software. The device can be a computer device or can be installed within a computer device. The device 600 may include: a candidate acquisition module 610, a score determination module 620, and a user determination module 630.
[0114] The candidate acquisition module 610 is used to acquire multiple exposure users of the related targeted content of the targeted content to be delivered, and construct a candidate set.
[0115] The score determination module 620 is used to determine the delivery score of the targeted content to be delivered relative to the exposed users based on the exposure information and user profile of the exposed users in the candidate set; wherein, the exposure information is used to indicate the historical browsing behavior of the exposed users towards the related targeted content, and the delivery score is used to indicate the estimated browsing behavior of the exposed users towards the targeted content to be delivered.
[0116] The user identification module 630 is used to select target exposure users whose delivery scores meet the conditions from the candidate set and add them to the targeted audience of the content to be delivered.
[0117] In an exemplary embodiment, such as Figure 7 As shown, the candidate acquisition module 610 includes: a first acquisition unit 611, a candidate acquisition unit 612, and a second acquisition unit 613.
[0118] The first acquisition unit 611 is used to acquire the exposure users of the first associated targeted content of the targeted content to be delivered, wherein the first associated targeted content refers to the targeted content that belongs to the same advertiser as the targeted content to be delivered.
[0119] The candidate acquisition unit 612 is used to construct the candidate set based on the exposure users of the first associated targeted content if the number of exposure users of the first associated targeted content meets the first requirement.
[0120] The second acquisition unit 613 is configured to acquire, if the number of users exposed to the first associated targeted content does not meet the first requirement, the users exposed to the second associated targeted content of the targeted content to be delivered, wherein the second associated targeted content refers to targeted content that is similar to the targeted content to be delivered. The candidate acquisition unit 612 is further configured to construct the candidate set based on the users exposed to the first associated targeted content and the users exposed to the second associated targeted content.
[0121] In an exemplary embodiment, the first acquisition unit is configured to acquire the exposure users of a first candidate associated targeted content of the targeted content to be delivered; wherein, the first candidate associated targeted content refers to the targeted content delivered by the advertiser of the targeted content to be delivered within a first historical time period; if the number of exposure users of the first candidate associated targeted content meets a second requirement, then the exposure users of the first associated targeted content are determined based on the exposure users of the first candidate associated targeted content; if the number of exposure users of the first candidate associated targeted content does not meet the second requirement, then the exposure users of a second candidate associated targeted content of the targeted content to be delivered are acquired, wherein the second candidate associated targeted content refers to the targeted content delivered by the advertiser of the targeted content to be delivered within a second historical time period; and the exposure users of the first associated targeted content are determined based on the exposure users of the first candidate associated targeted content and the exposure users of the second candidate associated targeted content.
[0122] In an exemplary embodiment, the second acquisition unit is configured to acquire tag information corresponding to the advertiser of the targeted content to be delivered; wherein the tag information is used to indicate the industry sector of the advertiser of the targeted content to be delivered; determine other advertisers with the same tag information as the advertiser of the targeted content to be delivered; determine a third candidate related targeted content based on the targeted content already delivered by the other advertisers; select the second related targeted content from the third candidate related targeted content based on the similarity between the third candidate related targeted content and the targeted content to be delivered; and acquire the exposure users of the second related targeted content.
[0123] In an exemplary embodiment, the second acquisition unit is further configured to acquire first feature information and second feature information, wherein the first feature information refers to the feature information of the third candidate associated targeted content and the second feature information refers to the feature information of the targeted content already delivered by the advertiser of the targeted content to be delivered; acquire the similarity between the first feature information and the second feature information; and determine the third candidate associated targeted content whose similarity satisfies the target condition as the second associated targeted content.
[0124] In an exemplary embodiment, such as Figure 7 As shown, the score determination module 620 includes: an intensity determination unit 621, a feature generation unit 622, and a score determination unit 623.
[0125] The intensity determination unit 621 is used to determine the exposure intensity of the exposed user based on the exposure information of the exposed user, wherein the exposure intensity refers to the browsing intensity of the exposed user for the associated targeted content.
[0126] The feature generation unit 622 is used to generate feature information of the exposed user based on the exposure intensity and user profile of the exposed user.
[0127] The score determination unit 623 is used to determine the delivery score of the targeted content to be delivered relative to the exposed user based on the feature information of the exposed user.
[0128] In an exemplary embodiment, the delivery score includes a conversion score, which is used to indicate the estimated probability that the exposed user will browse the targeted content to be delivered and perform an associated operation; the score determination unit 623 is used to perform prediction processing based on the feature information of the exposed user through a conversion score prediction model to obtain the conversion score of the targeted content to be delivered relative to the exposed user.
[0129] In an exemplary embodiment, the score determination unit 623 is configured to perform a prediction process based on the feature information of the exposed user using a click score prediction model to obtain a click-based predicted score of the targeted content to be delivered relative to the exposed user; wherein the click-based predicted score is used to indicate the predicted probability that the exposed user will browse the targeted content to be delivered; perform a prediction process based on the feature information of the exposed user using a conversion score prediction model to obtain a conversion predicted score of the targeted content to be delivered relative to the exposed user; wherein the conversion predicted score is used to indicate the predicted probability that the exposed user will execute the associated operation; and determine the conversion score of the targeted content to be delivered relative to the exposed user based on the click-based predicted score and the conversion predicted score.
[0130] In an exemplary embodiment, the training process of the conversion score prediction model is as follows:
[0131] The exposure information and user profile of the converted users corresponding to the already delivered targeted content of the advertiser of the targeted content to be delivered, as well as the exposure information and user profile of the seed users specified by the advertiser of the targeted content to be delivered, are obtained as the first positive sample; wherein, the converted user refers to the user who browses the already delivered targeted content and performs the associated operation;
[0132] Exposure information and user profiles of unconverted users are obtained from the candidate set as the first negative sample;
[0133] The conversion score prediction model is trained using the first positive sample and the first negative sample.
[0134] In an exemplary embodiment, the training process of the click score prediction model is as follows:
[0135] The exposure information and user profiles of users who have been exposed to the targeted content specified by the advertiser of the targeted content to be delivered are obtained as a second positive sample; wherein, the targeted content refers to the targeted content already delivered by other advertisers specified by the advertiser of the targeted content to be delivered;
[0136] Exposure information and user profiles of unexposed users of the specified targeted content are obtained from the candidate set as a second negative sample;
[0137] The click score prediction model is trained using the second positive sample and the second negative sample.
[0138] In an exemplary embodiment, the intensity determination unit 621 is configured to determine the difference between the exposure time and the current time based on the exposure time in the exposure information; wherein the exposure time refers to the time when the exposure user browses the associated targeted content; obtain a single exposure intensity value corresponding to the exposure user according to the difference; wherein the single exposure intensity value and the difference are positively correlated; and sum the single exposure intensity values to obtain the exposure intensity of the exposure user.
[0139] In an exemplary embodiment, the intensity determination unit 621 is further configured to: if the exposed user belongs to a first-class exposed user, determine the exposure intensity as the final exposure intensity, wherein the first-class exposed user refers to an exposed user determined based on the already delivered targeted content of the advertiser of the targeted content to be delivered; if the exposed user belongs to a second-class exposed user, perform attenuation processing on the exposure intensity to obtain the final exposure intensity, wherein the second-class exposed user refers to an exposed user determined based on targeted content that is similar to the targeted content to be delivered; wherein the final exposure intensity is used to combine with the user profile to generate the feature information.
[0140] In summary, the technical solution provided in this application estimates the browsing activity of exposed users for targeted content by using their exposure information and user profiles. The exposure information indicates the historical browsing activity of exposed users for related targeted content. By combining the exposure information with the user profile to estimate browsing activity, the accuracy of the estimation results for exposed users regarding the targeted content is ensured, thereby guaranteeing the accuracy of the targeted audience. Furthermore, since exposed users are those who have browsed the related targeted content of the users to be targeted, selecting the targeted audience from exposed users effectively guarantees the delivery effect of the targeted content.
[0141] Please refer to Figure 8 This diagram illustrates the structural block diagram of a computer device provided in one embodiment of this application. This computer device can be used to implement the functions of the above-described method for determining the target audience for targeted content. Specifically:
[0142] Computer device 800 includes a central processing unit (CPU) 801, a system memory 804 including random access memory (RAM) 802 and read-only memory (ROM) 803, and a system bus 805 connecting the system memory 804 and the CPU 801. Computer device 800 also includes a basic input / output (I / O) system 806 that facilitates information transfer between various devices within the computer, and a mass storage device 807 for storing the operating system 813, application programs 814, and other program modules 815.
[0143] The basic input / output system 806 includes a display 808 for displaying information and an input device 809 for user input, such as a mouse or keyboard. Both the display 808 and the input device 809 are connected to the central processing unit 801 via an input / output controller 810 connected to the system bus 805. The basic input / output system 806 may also include the input / output controller 810 for receiving and processing input from multiple other devices such as a keyboard, mouse, or electronic stylus. Similarly, the input / output controller 810 also provides output to a display screen, printer, or other types of output devices.
[0144] Mass storage device 807 is connected to central processing unit 801 via a mass storage controller (not shown) connected to system bus 805. Mass storage device 807 and its associated computer-readable media provide non-volatile storage for computer device 800. That is, mass storage device 807 may include computer-readable media (not shown) such as hard disk or CD-ROM (CompactDisc Read-Only Memory) drive.
[0145] Without loss of generality, computer-readable media can include computer storage media and communication media. Computer storage media includes volatile and non-volatile, removable and non-removable media implemented using any method or technology for storing information such as computer-readable instructions, data structures, program modules, or other data. Computer storage media includes RAM, ROM, EPROM (Erasable Programmable Read Only Memory), EEPROM (Electrically Erasable Programmable Read Only Memory), flash memory or other solid-state storage technologies, CD-ROM, DVD (Digital Video Disc) or other optical storage, magnetic tape cassettes, magnetic tape, disk storage, or other magnetic storage devices. Of course, those skilled in the art will recognize that computer storage media are not limited to the above-mentioned types. The system memory 804 and mass storage device 807 described above can be collectively referred to as memory.
[0146] According to various embodiments of this application, the computer device 800 can also be connected to a remote computer on a network, such as the Internet, for operation. That is, the computer device 800 can be connected to a network 812 via a network interface unit 811 connected to the system bus 805, or the network interface unit 811 can be used to connect to other types of networks or remote computer systems (not shown).
[0147] The memory also includes a computer program stored in the memory and configured to be executed by one or more processors to implement the above-described method for determining the target audience for targeted content.
[0148] In an exemplary embodiment, a computer-readable storage medium is also provided, the storage medium storing at least one instruction, at least one program, code set, or instruction set, wherein the at least one instruction, the at least one program, the code set, or the instruction set, when executed by a processor, implements the above-described method for determining the target audience for targeted content.
[0149] Optionally, the computer-readable storage medium may include: ROM (Read Only Memory), RAM (Random Access Memory), SSD (Solid State Drives), or optical disc, etc. The random access memory may include ReRAM (Resistance Random Access Memory) and DRAM (Dynamic Random Access Memory).
[0150] In an exemplary embodiment, a computer program product or computer program is provided, which includes computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the aforementioned method for determining the target audience for targeted content.
[0151] It should be noted that this application may display prompt interfaces, pop-ups, or output voice prompts before collecting user-related data and during the collection of user-related data (such as user profiles, browsing history, etc. mentioned in this application). These prompt interfaces, pop-ups, or voice prompts are used to inform the user that their relevant data is being collected. This ensures that this application only begins executing the steps related to collecting user-related data after receiving confirmation from the user regarding the prompt interface or pop-up; otherwise (i.e., without receiving confirmation from the user), the steps related to collecting user-related data end, meaning no user-related data is collected. In other words, all user data collected in this application is collected with the user's consent and authorization, and the collection, use, and processing of relevant user data must comply with the relevant laws, regulations, and standards of the relevant countries and regions.
[0152] It should be understood that "multiple" as used herein refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. Furthermore, the step numbers described herein are merely illustrative of one possible execution order. In some other embodiments, the steps may not be executed in numerical order, such as two steps with different numbers being executed simultaneously, or two steps with different numbers being executed in the reverse order of the illustration. This application does not impose any limitations on this.
[0153] The above description is merely an exemplary 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 determining the target audience for targeted content, characterized in that, The method includes: Obtain multiple exposure users of the related targeted content of the content to be targeted, and construct a candidate set; Based on the exposure time in the exposure information, the difference between the exposure time and the current time is determined; wherein, the exposure information is used to indicate the historical browsing of the exposed user for the associated targeted content, and the exposure time refers to the time when the exposed user browsed the associated targeted content; Based on the difference, the single exposure intensity value corresponding to the user is obtained; wherein, the single exposure intensity value and the difference are positively correlated. The exposure intensity of the user is obtained by summing the individual exposure intensity values. Based on the exposure intensity and user profile of the exposed user, the feature information of the exposed user is generated; Based on the characteristic information of the exposed users, the targeting score of the content to be delivered is determined relative to the exposed users. The targeting score is used to indicate the estimated browsing behavior of the exposed users for the targeting content to be delivered. Select target exposure users whose delivery scores meet the criteria from the candidate set and add them to the targeted audience for the content to be delivered.
2. The method according to claim 1, characterized in that, The process involves obtaining multiple users who have been exposed to the associated targeted content of the content to be delivered, and constructing a candidate set, including: Obtain the exposure users of the first associated targeted content of the targeted content to be delivered, where the first associated targeted content refers to the targeted content belonging to the same advertiser as the targeted content to be delivered; If the number of users exposed to the first associated targeted content meets the first requirement, then the candidate set is constructed based on the users exposed to the first associated targeted content; If the number of users exposed to the first associated targeted content does not meet the first requirement, then the users exposed to the second associated targeted content of the targeted content to be delivered are obtained. The second associated targeted content refers to targeted content that is similar to the targeted content to be delivered. Based on the users exposed to the first associated targeted content and the users exposed to the second associated targeted content, the candidate set is constructed.
3. The method according to claim 2, characterized in that, The users who have been exposed to the first associated targeted content of the targeted content to be delivered include: Obtain the exposure users of the first candidate related targeted content of the targeted content to be delivered; wherein, the first candidate related targeted content refers to the targeted content delivered by the advertiser of the targeted content to be delivered within a first historical time period; If the number of users exposed to the first candidate related targeted content meets the second requirement, then the users exposed to the first related targeted content are determined based on the number of users exposed to the first candidate related targeted content. If the number of users exposed to the first candidate related targeted content does not meet the second requirement, then the users exposed to the second candidate related targeted content of the targeted content to be delivered are obtained. The second candidate related targeted content refers to the targeted content delivered by the advertiser of the targeted content to be delivered within the second historical time period. The users exposed to the first related targeted content are determined based on the users exposed to the first candidate related targeted content and the users exposed to the second candidate related targeted content.
4. The method according to claim 2, characterized in that, The users who receive exposure to the second associated targeted content of the targeted content to be delivered include: Obtain the tag information corresponding to the advertiser of the targeted content to be delivered; wherein, the tag information is used to indicate the industry sector of the advertiser of the targeted content to be delivered; Identify other advertisers who have the same tag information as the advertiser of the targeted content to be delivered; The third candidate related targeting content is determined based on the already delivered targeting content of the other advertisers; Based on the similarity between the third candidate related targeting content and the target targeting content, the second related targeting content is selected from the third candidate related targeting content; Obtain the users who have been exposed to the second associated targeted content.
5. The method according to claim 4, characterized in that, The step of selecting the second related targeting content from the third candidate related targeting content based on the similarity between the third candidate related targeting content and the content to be targeted includes: Obtain first feature information and second feature information, wherein the first feature information refers to the feature information of the third candidate associated targeted content, and the second feature information refers to the feature information of the targeted content already delivered by the advertiser of the targeted content to be delivered; Obtain the similarity between the first feature information and the second feature information; The third candidate related content whose similarity meets the target condition is determined as the second related content.
6. The method according to claim 1, characterized in that, The delivery score includes a conversion score, which is used to indicate the estimated probability that the exposed user will browse the targeted content and perform related operations. The step of determining the delivery score of the targeted content relative to the exposed users based on their feature information includes: The conversion score is estimated relative to the exposed users by using a conversion score prediction model based on the characteristic information of the exposed users.
7. The method according to claim 1, characterized in that, The delivery score includes a conversion score, which is used to indicate the estimated probability that the exposed user will browse the targeted content and perform related operations. The step of determining the delivery score of the targeted content relative to the exposed users based on their feature information includes: The click score prediction model is used to predict the click score of the targeted content relative to the exposed user by performing prediction processing based on the feature information of the exposed user; wherein, the click score prediction is used to indicate the estimated probability that the exposed user will browse the targeted content. The conversion score prediction model is used to perform prediction processing based on the feature information of the exposed users to obtain the conversion prediction score of the targeted content to be delivered relative to the exposed users; wherein, the conversion prediction score is used to indicate the predicted probability of the exposed users executing the associated operation; Based on the estimated click score and the estimated conversion score, the conversion score of the targeted content to be delivered relative to the exposed users is determined.
8. The method according to claim 7, characterized in that, The training process of the conversion score prediction model is as follows: The exposure information and user profile of the converted users corresponding to the already delivered targeted content of the advertiser of the targeted content to be delivered, as well as the exposure information and user profile of the seed users specified by the advertiser of the targeted content to be delivered, are obtained as the first positive sample; wherein, the converted user refers to the user who browses the already delivered targeted content and performs the associated operation; Exposure information and user profiles of unconverted users are obtained from the candidate set as the first negative sample; The conversion score prediction model is trained using the first positive sample and the first negative sample.
9. The method according to claim 7, characterized in that, The training process of the click score prediction model is as follows: The exposure information and user profiles of users who have been exposed to the targeted content specified by the advertiser of the targeted content to be delivered are obtained as a second positive sample; wherein, the targeted content refers to the targeted content already delivered by other advertisers specified by the advertiser of the targeted content to be delivered; Exposure information and user profiles of unexposed users of the specified targeted content are obtained from the candidate set as a second negative sample; The click score prediction model is trained using the second positive sample and the second negative sample.
10. The method according to claim 1, characterized in that, After determining the exposure intensity of the exposed user based on the exposure information of the exposed user, the method further includes: If the exposed users belong to a certain type of exposed users, then the exposure intensity is determined as the final exposure intensity. The certain type of exposed users refers to the exposed users determined based on the targeted content already delivered by the advertiser of the targeted content to be delivered. If the exposed user belongs to the second type of exposed user, the exposure intensity is attenuated to obtain the final exposure intensity. The second type of exposed user refers to the exposed user determined based on the targeted content that is similar to the targeted content to be delivered. The attenuation process is used to reduce the exposure intensity. The final exposure intensity is used to combine with the user profile to generate the feature information.
11. A device for determining the target audience for targeted content, characterized in that, The device includes: The candidate acquisition module is used to acquire multiple exposure users of the related targeted content of the content to be delivered, and construct a candidate set; The score determination module is used to determine the difference between the exposure time and the current time based on the exposure time in the exposure information; wherein, the exposure information is used to indicate the historical browsing of the exposure user for the associated targeted content, and the exposure time refers to the time when the exposure user browsed the associated targeted content; The score determination module is further configured to obtain the single exposure intensity value corresponding to the exposure user based on the difference; wherein, the single exposure intensity value and the difference are positively correlated; The score determination module is also used to sum the individual exposure intensity values to obtain the exposure intensity of the user. The score determination module is also used to generate feature information of the exposed user based on the exposure intensity and user profile of the exposed user; The score determination module is further configured to determine the delivery score of the targeted content to be delivered relative to the exposed user based on the characteristic information of the exposed user, and the delivery score is used to indicate the estimated browsing behavior of the exposed user for the targeted content to be delivered; The user identification module is used to select target exposure users whose delivery scores meet the conditions from the candidate set and add them to the targeted audience of the content to be delivered.
12. A computer device, characterized in that, The computer device includes a processor and a memory, the memory storing at least one instruction, at least one program, a code set, or an instruction set, the at least one instruction, the at least one program, the code set, or the instruction set being loaded and executed by the processor to implement the method for determining the target audience for targeted content as described in any one of claims 1 to 10.
13. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores at least one instruction, at least one program, code set, or instruction set, wherein the at least one instruction, the at least one program, the code set, or instruction set is loaded and executed by a processor to implement the method for determining the target audience for targeted content as described in any one of claims 1 to 10.
14. A computer program product comprising computer instructions executed by a processor to implement the method for determining the target audience for targeted content as described in any one of claims 1 to 10.