Advertisement putting method and device, computing equipment, medium and program product
By comprehensively considering consumer and seller characteristics and advertising exposure information, and using neural network models to generate advertising delivery strategies, the problem of poor delivery results caused by relying on user profiles in existing technologies is solved, and more efficient advertising delivery results are achieved.
Patent Information
- Application Number
- CN202511028722.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-24
- Publication Date
- 2025-11-11
AI Technical Summary
Existing advertising delivery technologies rely on user profiles, resulting in poor delivery performance and failing to meet advertisers' requirements.
Taking into account consumer characteristics, seller characteristics, advertising exposure information, and human characteristics, an advertising strategy is generated. Feature extraction and prediction are performed using models such as attention neural networks, fully connected layers, cross-attention, and gated neural networks.
This generates more realistic advertising strategies, improving campaign effectiveness and conversion rates among target consumers.
Smart Images
Figure CN120931336A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of advertising delivery technology, and in particular to an advertising delivery method, apparatus, computing device, medium and program product. Background Technology
[0002] Currently, advertising has become an important means for businesses to promote their brands, products, and services. With the rapid development of internet technology, advertising channels and methods are constantly innovating and evolving.
[0003] Existing advertising delivery technologies typically rely on user profiles to determine the target audience for ads. However, this approach is affected by the accuracy of the user profiles, resulting in poor ad delivery performance and failing to meet advertisers' requirements. Summary of the Invention
[0004] This application aims to at least address the technical problem of poor advertising performance in the background art. Therefore, one objective of this application is to provide an advertising delivery method to obtain a reasonable advertising delivery strategy and achieve better delivery results.
[0005] An embodiment of the first aspect of this application provides an advertising delivery method, comprising: acquiring consumer characteristic information, wherein the consumer characteristic information indicates the purchase history of the target consumers of the advertisement to be delivered; acquiring seller characteristic information, wherein the seller characteristic information indicates the sales strategy of the seller of the advertisement to be delivered; generating interactive information between the consumer characteristic information and the seller characteristic information based on the consumer characteristic information and the seller characteristic information; acquiring advertisement exposure information, wherein the advertisement exposure information indicates the exposure history of the advertisement to be delivered; and determining the delivery strategy of the advertisement to be delivered based on the interactive information, the advertisement exposure information and predetermined artificial characteristics.
[0006] An embodiment of the second aspect of this application provides an advertising delivery device, comprising: a first acquisition module for acquiring consumer characteristic information, wherein the consumer characteristic information indicates the purchase history of the target consumers of the advertisement to be delivered; a second acquisition module for acquiring seller characteristic information, wherein the seller characteristic information indicates the sales strategy of the seller of the advertisement to be delivered; an interaction information determination module for generating interaction information between the consumer characteristic information and the seller characteristic information based on the consumer characteristic information and the seller characteristic information; a third acquisition module for acquiring advertisement exposure information, wherein the advertisement exposure information indicates the exposure history of the advertisement to be delivered; and a delivery strategy determination module for determining the delivery strategy of the advertisement to be delivered based on the interaction information, the advertisement exposure information, and predetermined artificial characteristics.
[0007] An embodiment of the third aspect of this application provides a computing device, including: at least one processor; and at least one memory communicatively connected to the at least one processor, the at least one memory storing instructions that, when executed individually or jointly by the at least one processor, cause the computing device to perform the advertising delivery method described above.
[0008] An embodiment of the fourth aspect of this application provides a computer-readable storage medium storing instructions that, when executed individually or jointly by one or more processors of a computing device, cause the computing device to perform the advertising delivery method described above.
[0009] An embodiment of the fifth aspect of this application provides a computer program product, including instructions that, when executed individually or jointly by one or more processors of a computing device, cause the computing device to perform the advertising delivery method described above.
[0010] The technical solution in this application takes into account factors such as consumer purchase information, seller's sales strategy, and advertising exposure information, which is more in line with the actual situation of advertising placement and can generate a reasonable advertising placement plan, effectively improving the advertising placement effect.
[0011] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, the following are specific embodiments of this application. Attached Figure Description
[0012] In the accompanying drawings, unless otherwise specified, the same reference numerals throughout the various drawings denote the same or similar parts or elements. These drawings are not necessarily drawn to scale. It should be understood that these drawings depict only some embodiments disclosed in this application and should not be construed as limiting the scope of this application.
[0013] Figure 1 This is a flowchart illustrating the advertising delivery method of some embodiments of this application;
[0014] Figure 2 This is a schematic diagram illustrating the process of obtaining consumer characteristic information in some embodiments of this application;
[0015] Figure 3 This is a schematic diagram illustrating the process of obtaining seller characteristic information in some embodiments of this application;
[0016] Figure 4 This is a schematic diagram illustrating the process of generating interactive information in some embodiments of this application;
[0017] Figure 5This is a schematic diagram illustrating the process of obtaining advertising exposure information in some embodiments of this application;
[0018] Figure 6 This is a flowchart illustrating the process of determining the delivery strategy in some embodiments of this application;
[0019] Figure 7 This is a flowchart illustrating the process of determining the conversion probability in some embodiments of this application;
[0020] Figure 8 This is a schematic block diagram of an advertising delivery device according to some embodiments of this application;
[0021] Figure 9 This is a schematic block diagram of a computing device according to some embodiments of this application;
[0022] Figure 10 This is a flowchart illustrating an advertising delivery method according to some embodiments of this application. Detailed Implementation
[0023] The embodiments of the technical solution of this application will now be described in detail with reference to the accompanying drawings. These embodiments are only used to more clearly illustrate the technical solution of this application and are therefore merely examples, and should not be used to limit the scope of protection of this application.
[0024] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains; the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the application; the terms “comprising” and “having”, and any variations thereof, in the specification, claims, and foregoing description of the drawings are intended to cover non-exclusive inclusion.
[0025] In the description of the embodiments of this application, technical terms such as "first" and "second" are used only to distinguish different objects and should not be construed as indicating or implying relative importance or implicitly specifying the number, specific order, or primary and secondary relationship of the indicated technical features. In the description of the embodiments of this application, "multiple" means two or more, unless otherwise explicitly defined.
[0026] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.
[0027] In the description of the embodiments in this application, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Additionally, the character " / " in this document generally indicates that the preceding and following related objects have an "or" relationship.
[0028] In the description of the embodiments of this application, the term "multiple" refers to two or more (including two), and similarly, "multiple groups" refers to two or more (including two).
[0029] The collection, storage, use, processing, transmission, provision, and disclosure of user personal information involved in the technical solution of this application all comply with the provisions of relevant laws and regulations and do not violate public order and good morals.
[0030] Currently, advertising has become an important means for businesses to promote their brands, products, and services. With the rapid development of internet technology, advertising channels and methods are constantly innovating and evolving.
[0031] Existing advertising delivery technologies typically rely on user profiles to determine the target audience for ads. However, this approach is affected by the accuracy of user profiles. In real-world scenarios, verifying the authenticity of user information is difficult, leading to poor ad delivery results and failing to meet advertisers' requirements.
[0032] To achieve better advertising results, a comprehensive approach can be taken, considering consumer characteristics, seller characteristics, ad exposure data, and pre-defined human factors to generate an advertising strategy. Strategies developed in this way are more aligned with real-world needs and can be tailored to specific advertising requirements, effectively improving advertising performance and increasing conversion rates among target consumers.
[0033] This application provides a method for placing advertisements. (See reference) Figure 1 The advertising placement method 100 includes steps 110 to 150.
[0034] Step 110: Obtain consumer characteristic information. Consumer characteristic information indicates the purchase history of the target consumers for whom the advertisement is to be placed.
[0035] Step 120: Obtain seller characteristic information. Seller characteristic information indicates the sales strategy of the seller to be advertised.
[0036] Step 130: Generate interactive information between consumer characteristic information and seller characteristic information based on consumer characteristic information and seller characteristic information.
[0037] Step 140: Obtain ad exposure information. Ad exposure information indicates the exposure history of the ad to be delivered.
[0038] Step 150: Determine the placement strategy for the advertisement to be placed based on interaction information, advertisement exposure information and pre-determined human characteristics.
[0039] In embodiments of this application, consumers can be the targets of advertising, and the advertisements to be advertised will be pushed to the consumers. Consumer characteristic information is related to the consumer's purchase history and can be obtained based on the consumer's purchase history, such as the consumer's purchase history of the products and / or services involved in the advertisements.
[0040] In some embodiments, reference Figure 2 Step 110 includes steps 210 to 230.
[0041] Step 210: Obtain the purchase history of the target consumer.
[0042] Step 220: Based on purchase history, determine the single purchase characteristics of the target consumer.
[0043] Step 230: Based on purchase history, determine the multiple purchase characteristics of the target consumer.
[0044] For a given advertisement to be delivered, after determining the target audience, i.e., the target consumers, the purchase history of these target consumers can be obtained. The purchase history can include each purchase record of the target consumers, including information such as the number of purchases, purchase frequency, and the price of each purchase.
[0045] In step 220, the single-purchase characteristic information of the target consumer will be determined based on the acquired purchase history. The single-purchase characteristic information can indicate information about the target consumer in a particular purchase process, such as the purchase price.
[0046] In some embodiments, step 220 includes: using the purchase history as input to a first extraction model, and using the first extraction model to output single purchase feature information. The first extraction model is used to extract features from the purchase history.
[0047] In some embodiments, the first extraction model includes an attention neural network model.
[0048] A trained attention-based neural network (ADNN) model can be used as the first extraction model to extract features from the purchase history. The purchase history obtained in step 210 is used as input to the ADNN model for feature extraction, yielding single-purchase feature information. In some embodiments, a sample of consumers can be selected, and the ADNN model can be trained based on their purchase history to obtain the first extraction model for feature extraction from the purchase history.
[0049] In step 230, the multiple purchase characteristics of the target consumer will be determined based on the purchase history obtained in step 210. These multiple purchase characteristics indicate information about the target consumer's multiple purchases, such as the total number of purchases and purchase frequency.
[0050] In some embodiments, step 230 includes: using purchase history and / or single purchase feature information as input to a second extraction model, and using the second extraction model to output multiple purchase feature information. The second extraction model is used to extract features from the purchase history.
[0051] In some embodiments, the second extraction model includes an attention neural network model.
[0052] Similar to the first extraction model, the second extraction model can also be an attention neural network model. The input to the second extraction model can include the purchase history obtained in step 210, or the single-purchase feature information obtained in step 220. The purchase history and / or single-purchase feature information are used as input to the trained attention neural network model for feature extraction, resulting in multiple purchase feature information. In some embodiments, a sample of consumers can be selected, and the attention neural network model can be trained based on their purchase history to obtain the second extraction model for feature extraction of the purchase history.
[0053] In some embodiments, single-purchase feature information and multiple-purchase feature information can be in the form of feature vectors.
[0054] By acquiring single-purchase and multiple-purchase characteristic information of target consumers, we can determine their purchasing tendencies for the products and / or services involved in the advertisements to be placed, thereby enabling more precise ad delivery to target consumers.
[0055] In embodiments of this application, the seller can be an advertiser of the advertisement to be placed, or a business provider offering the products and / or services involved in the advertisement. The seller's characteristic information is related to the seller's sales strategy and can be obtained based on the seller's sales strategy. The sales strategy may include, for example, limited-time offers or discounts on the products and / or services involved in the advertisement.
[0056] In some embodiments, reference Figure 3 Step 120 includes steps 310 to 320.
[0057] Step 310: Obtain the seller's sales strategy.
[0058] Step 320: Based on the sales strategy, determine the seller's sales characteristic information.
[0059] In step 310, the sales strategies of the sellers whose advertisements are to be placed will be obtained. In step 320, the seller's characteristic information will be determined based on these obtained sales strategies.
[0060] In some embodiments, seller characteristic information may be in the form of a feature vector.
[0061] In some embodiments, step 320 includes: using the sales strategy as input to the third extraction model, and using the third extraction model to output salesperson feature information. The third extraction model is used to extract features from the sales strategy.
[0062] In some embodiments, the third extraction model includes a fully connected layer model.
[0063] A trained fully connected layer model can be used as a third extraction model to extract features from the sales strategy. The fully connected layer can be one layer within this model. The sales strategy obtained in step 310 is used as input to the fully connected layer model, which performs feature extraction to obtain the seller's feature information. In some embodiments, a sample of sellers can be selected, and the fully connected layer model can be trained based on the sales strategies of these sample sellers to obtain the third extraction model for feature extraction from the sales strategy.
[0064] By identifying seller characteristics, the resulting advertising strategies can be tailored to different sales strategies, effectively improving advertising performance.
[0065] After obtaining consumer and seller characteristic information, interaction information between the two can be generated. This interaction information can reveal the correlation between the consumer and seller characteristics, such as the purchasing behavior of consumers under different sales strategies employed by the seller. Advertising strategies derived from this interaction information can more accurately align with consumer habits, achieving better targeting results.
[0066] In some embodiments, reference Figure 4 Step 130 includes steps 410 to 420.
[0067] Step 410: Input the consumer feature information and seller feature information into the fourth extraction model. The fourth extraction model is used to perform cross-feature extraction on the consumer feature information and seller feature information.
[0068] Step 420: Use the fourth extraction model to generate interactive information.
[0069] In some embodiments, the interaction information between consumer characteristic information and seller characteristic information can also be in the form of a feature vector.
[0070] In some embodiments, the fourth extraction model includes a cross-attention model.
[0071] A trained cross-attention model can be used as the fourth extraction model to generate interaction features between consumer and seller features. The cross-attention model involves the query vector, key vector, and value vector from the attention mechanism. Cross-feature extraction is performed based on these three vectors to output the interaction information. In one example, the seller features determined in step 320 can be used as the query of the cross-attention model, and the single-purchase features determined in step 220 and / or the multiple-purchase features determined in step 230 can be used as the key and / or value of the cross-attention model. The cross-attention model then generates the interaction information between the consumer and seller features.
[0072] Since interactive information can reflect the relationship between consumer and seller characteristics, generating interactive information can further reveal the purchasing tendencies of target consumers under different sales strategies, thereby enabling more precise advertising to be pushed to target consumers.
[0073] In embodiments of this application, the exposure history may include the history of how the advertisement was displayed, such as the number of times it was pushed to the screen. Ad exposure information is related to the exposure history of the advertisement to be delivered, and can be determined based on the exposure data of the advertisement before this delivery.
[0074] In some embodiments, reference Figure 5 Step 140 includes steps 510 to 520.
[0075] Step 510: Obtain the exposure history of the advertisement to be placed.
[0076] Step 520: Use the exposure history as input to the fifth extraction model, and use the fifth extraction model to output the advertisement exposure information. The fifth extraction model is used to extract features from the exposure history.
[0077] In some embodiments, ad exposure information may be in the form of a feature vector.
[0078] In some embodiments, the fifth extraction model includes a gated neural network model or a long short-term memory neural network model.
[0079] A trained gated recurrent unit (GRU) model or a long short-term memory (LSTM) model can be used as the fifth extraction model to extract features from the exposure history. The exposure history obtained in step 510 is used as input to the fifth extraction model to extract features and obtain the ad exposure information. In some embodiments, a number of sample ads can be selected, and the GRU or LSTM model can be trained based on the exposure history of these sample ads to obtain the fifth extraction model for feature extraction from the exposure history.
[0080] By obtaining the ad exposure information of the ads to be placed, the ad placement can be based on the ad's exposure history, making the placement strategy more reasonable.
[0081] Based on information such as consumers' purchase history, sellers' sales strategies, and the exposure history of the advertisements to be placed, some artificial characteristics can be predetermined, such as the number of purchases in the past seven days or the number of purchases in the past month. In step 150, in addition to the interaction information between consumer characteristic information and seller characteristic information, and advertisement exposure information, these artificial characteristics will also be considered to ultimately obtain the placement strategy for the advertisements to be placed.
[0082] In some embodiments, reference Figure 6 Step 150 includes steps 610 to 620.
[0083] Step 610: Determine the conversion probability of the target consumer based on interaction information, advertising exposure information, and human characteristics.
[0084] Step 620: Determine the delivery strategy based on the conversion probability.
[0085] In embodiments of this application, the conversion probability of a target consumer can indicate the probability that a target consumer will engage in consumption behavior (e.g., purchase a product or participate in a related activity) regarding the products and / or services mentioned in the advertisement after receiving the advertisement to be delivered. The conversion probability can reflect the target consumer's acceptance of the advertisement to be delivered, their participation tendency, etc. In some embodiments, the conversion probability may include, for example, the probability that the advertisement to be delivered is exposed to the target consumer, the probability that the target consumer is successfully converted after receiving the advertisement to be delivered, etc.
[0086] In some embodiments, reference Figure 7 Step 610 includes steps 710 to 720.
[0087] Step 710: Input the interaction information, ad exposure information, and human-generated features into the prediction model. The prediction model is used to predict the ad delivery strategy.
[0088] Step 720: Use the prediction model to generate conversion probability categories and / or conversion probability feature vectors for the target consumers. The conversion probability category indicates the conversion probability corresponding to at least one campaign strategy category. The conversion probability feature vector includes a vector corresponding to the conversion probability obtained from the conversion probability.
[0089] In step 710, the interaction information of consumer characteristic information and seller characteristic information, advertising exposure information and artificial characteristics obtained from the steps mentioned above will be used as input to the prediction model, which will then predict the conversion probability of the target consumer.
[0090] In step 720, the prediction model will output the conversion probability category and / or conversion probability feature vector of the target consumer based on these inputs. The conversion probability category for a target consumer reflects the probability of that consumer converting after receiving a certain advertising strategy. For example, for a certain advertising strategy, the conversion probability category can include three categories: highly likely to convert, very likely to convert, and certain to convert. The prediction model will predict the conversion probability of the target consumer based on the input information and output one of these three categories. Alternatively, for a certain advertising strategy, the conversion probability category can include four categories: conversion probability 100%, conversion probability ≥ 50%, 50% > conversion probability ≥ 30%, and conversion probability < 30%. The prediction model will predict the conversion probability of the target consumer based on the input information and output one of these four categories. Or, it can be for multiple advertising strategies, such as multiple ad creatives or multiple ad media platforms. The conversion probability category indicates the conversion probability of consumers for each ad creative or ad media platform. The prediction model will predict based on the input information and output the recommended ad creative or ad media platform based on the conversion probability corresponding to each ad creative or ad media platform, that is, the ad creative or ad media platform with the highest conversion probability. The conversion probability feature vector of the target consumer can be a feature vector corresponding to the conversion probability of the target consumer. This feature vector can be used in other models, such as as input to other models. In one example, after the prediction model predicts the conversion probability of the target consumer, a targeting strategy determination model can be set up to determine the specific targeting strategy based on the conversion probability of the target consumer. In this example, the output of the prediction model in steps 710 and 720 can be used as the input of the targeting strategy determination model, that is, the probability conversion feature vector obtained by the prediction model is input into the targeting strategy determination model to obtain the specific targeting strategy.
[0091] In some embodiments, the prediction model can use a fully connected layer model. In a fully connected layer model, one fully connected layer is included. Different outputs of the prediction model in step 720 can be achieved by changing the subsequent output layer of the fully connected layer.
[0092] For predictive models that require generating conversion probability categories, a category output layer can be added after the fully connected layer. The output of the category output layer is a single scalar (e.g., category 1, 2, 3, etc.), allowing for the output of different conversion probability categories, such as conversion probability category 1, conversion probability category 2, or conversion probability category 3, etc. During the training process of this predictive model, pre-defined training materials such as sample delivery materials and sample media platforms can be used. The model is trained based on whether sample consumers have made purchases related to each sample delivery material and sample media platform.
[0093] For prediction models that require generating conversion probability feature vectors, a vector output layer can be added after the fully connected layer. The output of the vector output layer is a feature vector, thus outputting the feature vector corresponding to the conversion probability. For this type of prediction model, during training, the conversion probabilities can be assigned to random vectors and fixed, and contrastive learning can be used to train the model. When this prediction model predicts the conversion probability feature vector, it will output based on vector similarity.
[0094] By using predictive models to output results in different formats, multimodal outputs can be achieved, enabling customized advertising based on user needs. This allows for precise recommendations of advertising creatives, media platforms, and placement locations, thus improving the flexibility of advertising strategies.
[0095] In step 620, based on the obtained conversion probability, the placement strategy for the advertisement to be placed can be further determined. For example, when the conversion probability category of the target consumer is determined to be very likely to convert or certain to convert, the advertisement can be placed to them. Or when the conversion probability category of the target consumer is determined to be conversion probability < 30%, the advertisement can not be placed to them. Or when the advertisement placement material with the highest conversion probability of the target consumer is determined, the advertisement placement material can be used for placement.
[0096] In some embodiments, the advertising delivery method 100 further includes:
[0097] After determining the transformation probability, perform at least one of the following steps:
[0098] Based on the predetermined number of target consumers, determine the number of convertible consumers among the target consumers;
[0099] Based on the predetermined number of consumers to be converted, determine the number of target consumers, including those to be converted.
[0100] After determining the conversion probability of the target consumers, and assuming a certain number of target consumers to be targeted with the advertisement, the conversion probability can be used to determine the number of consumers among these target consumers who can be successfully converted, that is, the number of consumers who can make a purchase after receiving the advertisement. In addition, if the number of consumers who need to be converted (i.e., consumers to be converted) is determined, the conversion probability can also be used to determine the number of target consumers to whom the aforementioned advertising method 100 should be applied.
[0101] In some embodiments, the determination can be made according to the following formula:
[0102] B = A × P (model exposure) × P (model conversion) = C × P (model conversion) (1)
[0103] In the formula, A represents the number of predetermined target consumers, that is, the number of target consumers to whom the advertising delivery method 100 needs to be executed; B represents the number of convertible consumers, that is, the number of consumers among these target consumers who can be successfully converted; P (model exposure) represents the probability that the advertisement to be delivered will be exposed to the target consumers; C represents the number of target consumers who are exposed, that is, the number of target consumers who receive the advertisement to be delivered; and P (model conversion) represents the probability that the target consumers will be successfully converted after receiving the advertisement to be delivered.
[0104] In equation (1) above, P (model exposure) can be further determined according to the following formula:
[0105] P(Model Exposure) = P(Model Exposure | Convertible) × P(Convertible) + P(Model Exposure | Non-Convertible) × P(Non-Convertible) (2)
[0106] In the formula, P(model exposure) represents the probability that the advertisement to be delivered will be exposed to the target consumer, P(model exposure | convertible) is the True Positive Rate (TPR) of the prediction model, which represents the probability that the advertisement to be delivered will be delivered to the consumer who is actually converted, P(model exposure | non-convertible) is the False Positive Rate (FPR) of the prediction model, which represents the probability that the advertisement to be delivered will be delivered to the consumer who is actually not converted, P(convertible) represents the probability that the consumer is actually converted, and P(non-convertible) represents the probability that the consumer is not actually converted.
[0107] In the above example, when using the prediction model for prediction, formulas (1) and (2) can be used to convert between the number of target consumers and the number of convertible consumers and / or between the number of consumers to be converted and the number of target consumers, based on the performance of the prediction model and the predicted conversion probability. This allows for the rapid identification of the number of convertible consumers and / or the number of target consumers, enabling the rapid determination of advertising placement strategies and improving prediction efficiency.
[0108] Based on the same technical concept, embodiments of this application provide an advertising delivery device. Embodiments of the advertising delivery device can be referenced to embodiments of the advertising delivery method; repeated details will not be repeated. Reference Figure 8 The advertising delivery device 800 includes a first acquisition module 810, a second acquisition module 820, an interactive information determination module 830, a third acquisition module 840, and a delivery strategy determination module 850.
[0109] The first acquisition module 810 is used to acquire consumer characteristic information. The consumer characteristic information indicates the purchase history of the target consumers for whom the advertisement is to be delivered.
[0110] The second acquisition module 820 is used to acquire seller characteristic information. This includes seller characteristic information indicators and the sales strategy of the seller to whom the advertisement is to be placed.
[0111] The interaction information determination module 830 is used to generate interaction information between consumer characteristic information and seller characteristic information based on consumer characteristic information and seller characteristic information.
[0112] The third acquisition module 840 is used to acquire ad exposure information. Ad exposure information indicates the exposure history of the ad to be delivered.
[0113] The placement strategy determination module 850 is used to determine the placement strategy for the advertisement to be placed based on interaction information, advertisement exposure information and pre-determined human characteristics.
[0114] The first acquisition module 810, the second acquisition module 820, the interaction information determination module 830, the third acquisition module 840, and the delivery strategy determination module 850 in the advertising delivery device 800 can correspond to steps 110 to 150 in the advertising delivery method 100, and will not be described in detail here for the sake of brevity. It should be understood that, corresponding to the embodiment of the advertising delivery method 100, the embodiment of the advertising delivery device 800 may also include more modules.
[0115] It should be noted that the functions of the modules discussed herein can be divided into multiple modules, and / or at least some functions of multiple modules can be combined into a single module. The specific actions performed by a particular module discussed herein include the specific module itself performing the action, or alternatively, the specific module calling or otherwise accessing another component or module that performs the action (or performs the action in conjunction with the specific module). Therefore, a specific module performing an action can include the specific module performing the action itself and / or another module that performs the action, called or otherwise accessed by the specific module.
[0116] It should also be understood that this article can describe various technologies in the general context of software and hardware components or program modules. The above regarding... Figure 8 The described modules can be implemented in hardware or in hardware in combination with software and / or firmware. For example, these modules can be implemented as computer program code / instructions configured to execute in one or more processors and stored in a computer-readable storage medium. Alternatively, these modules can be implemented as hardware logic / circuit. Hardware logic / circuit may include integrated circuit chips (which include processors (e.g., central processing unit (CPU), microcontrollers, microprocessors, digital signal processors (DSPs), etc.), memory, one or more communication interfaces, and / or one or more components of other circuitry), and may optionally execute received program code and / or include embedded firmware to perform functions.
[0117] This application provides a computing device 900, such as... Figure 9 As shown. Figure 9 An example configuration of a computing device 900 that can be used to implement the advertising delivery method 100 described herein is shown. For example, the advertising delivery device 800 described above may be implemented wholly or at least partially by the computing device 900 or a similar device or system.
[0118] The computing device 900 may include at least one processor 905 capable of communicating with each other, such as via a bus 904 or other suitable connection, a memory 907, multiple communication interfaces 902, a display device 901, other input / output (I / O) devices 903, and one or more mass storage devices 906. Instructions are stored on the memory 907, which, when executed by the processor 905, cause the processor 905 to perform the advertising delivery method as described in the above embodiments.
[0119] The computing device 900 can be a variety of different types of devices. Examples of the computing device 900 include, but are not limited to: desktop computers, server computers, laptop or netbook computers, mobile devices (e.g., tablet computers, cellular or other wireless phones (e.g., smartphones), notebook computers, mobile stations), wearable devices (e.g., glasses, watches), entertainment devices (e.g., entertainment appliances, set-top boxes communicatively coupled to a display device, game consoles), televisions or other display devices, automotive computers, and so on.
[0120] Processor 905 may be a single processing unit or multiple processing units, and all processing units may include single or multiple computing units or multiple cores. Processor 905 may be implemented as one or more microprocessors, microcomputers, microcontrollers, digital signal processors, central processing units, state machines, logic circuits, and / or any device that manipulates signals based on operating instructions. Among other capabilities, processor 905 may be configured to fetch and execute computer-readable instructions stored in memory 907, mass storage device 906, or other computer-readable media, such as program code of operating system 908, program code of application program 909, program code of other program 910, etc.
[0121] Memory 907 and mass storage device 906 are examples of computer-readable storage media for storing instructions executed by processor 905 to perform the various functions described above. For example, memory 907 can generally include both volatile and non-volatile memory (e.g., RAM, ROM, etc.). Furthermore, mass storage device 906 can generally include hard disk drives, solid-state drives, removable media, including external and removable drives, memory cards, flash memory, floppy disks, optical disks (e.g., CDs, DVDs), storage arrays, network-attached storage, storage area networks, etc. Both memory 907 and mass storage device 906 can be collectively referred to herein as memory or computer-readable storage media, and can be non-transitory media capable of storing computer-readable, processor-executable program instructions as computer program code, which can be executed by processor 905 as a specific machine configured to perform the operations and functions described in the examples herein.
[0122] Multiple programs can be stored on mass storage device 906. These programs include operating system 908, one or more application programs 909, other programs 910, and program data 911, and they can be loaded into memory 907 for execution. Examples of such application programs or program modules may include, for example, computer program logic (e.g., computer program code or instructions) for implementing the following components / functions: advertising delivery device 800 (including first acquisition module 810, second acquisition module 820, interaction information determination module 830, third acquisition module 840, and delivery strategy determination module 850), advertising delivery method 100 (including any suitable steps of advertising delivery method 100), and / or other embodiments described herein.
[0123] Although Figure 9 The data is illustrated as being stored in memory 907 of computing device 900, but operating system 908, application program 909, other programs 910 and program data 911 or portions thereof may be implemented using any form of computer-readable medium accessible by computing device 900.
[0124] One or more communication interfaces 902 are used for exchanging data with other devices, such as via a network, direct connection, etc. Such communication interfaces can be one or more of the following: any type of network interface (e.g., a network interface card (NIC)), wired or wireless (such as IEEE 802.11 Wireless LAN (WLAN)) interface, Wi-MAX interface, Ethernet interface, Universal Serial Bus (USB) interface, cellular network interface, Bluetooth™ interface, Near Field Communication (NFC) interface, etc. Communication interface 902 can facilitate communication across various network and protocol types, including wired networks (e.g., LAN, cable, etc.) and wireless networks (e.g., WLAN, cellular, satellite, etc.), the Internet, etc. Communication interface 902 can also provide communication with external storage devices (not shown), such as storage arrays, network-attached storage, storage area networks, etc.
[0125] In some examples, a display device 901, such as a monitor, may be included for displaying information and images to the user. Other I / O devices 903 may be devices that receive various inputs from the user and provide various outputs to the user, and may include touch input devices, gesture input devices, cameras, keyboards, remote controls, mice, printers, audio input / output devices, and so on.
[0126] The technologies described herein can be supported by these various configurations of computing device 900, and are not limited to specific examples of the technologies described herein. For example, the functionality can also be implemented wholly or partially on a “cloud” using a distributed system. A cloud includes and / or represents a platform for resources. The platform abstracts the underlying functionality of the cloud’s hardware (e.g., servers) and software resources. Resources may include applications and / or data that can be used when performing computational processing on servers remote from computing device 900. Resources may also include services provided via the Internet and / or via subscriber networks such as cellular or Wi-Fi networks. The platform can abstract resources and functionality to connect computing device 900 to other computing devices. Therefore, the implementation of the functionality described herein can be distributed throughout the cloud. For example, the functionality may be implemented partly on computing device 900 and partly through a platform that abstracts the functionality of the cloud.
[0127] This application also provides a computer-readable storage medium storing instructions that, when executed individually or jointly by one or more processors of a computing device, cause the computing device to perform the methods described in any of the above embodiments.
[0128] Computer-readable storage media include volatile and non-volatile, removable and non-removable media implemented by any method or technology for storing information such as computer-readable instructions, data structures, program modules, or other data. Computer-readable storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, DVD, or other optical storage devices, magnetic cassettes, magnetic tapes, disk storage devices or other magnetic storage devices, or any other non-transmission medium that can be used to store information for access by computer equipment.
[0129] This application also provides a computer program product including instructions that, when executed individually or jointly by one or more processors of a computing device, cause the computing device to perform the methods as described in any of the above embodiments.
[0130] A specific embodiment of this application is described below. It should be understood that this specific embodiment is described for illustrative purposes only and should not be construed as limiting the scope of this application.
[0131] like Figure 10 As shown, the purchase history of the target consumers is used as input to the first and second extraction models to obtain consumer feature information. The sales strategy of the seller is used as input to the third extraction model to obtain seller feature information. The exposure history of the advertisement to be placed is used as input to the fifth extraction model to obtain advertisement exposure information.
[0132] The fourth extraction model is used to perform cross-feature extraction on consumer feature information and seller feature information to obtain the interaction information of consumer feature information and seller feature information.
[0133] By inputting the interaction information of consumer and seller characteristics, advertising exposure information, and pre-determined human characteristics into the prediction model, the model will output the conversion probability of the target consumer. Based on the conversion probability, a targeted advertising strategy can be further determined.
[0134] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and not to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. These modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application, and they should all be covered within the scope of the claims and specification of this application. In particular, as long as there is no structural conflict, the various technical features mentioned in the embodiments can be combined in any way. This application is not limited to the specific embodiments disclosed herein, but includes all technical solutions falling within the scope of the claims.
Claims
1. An advertising placement method, comprising: Obtain consumer characteristic information, which indicates the purchase history of the target consumers to be advertised; Obtain seller characteristic information, which indicates the sales strategy of the seller to be advertised; Based on the consumer characteristic information and the seller characteristic information, interactive information between the consumer characteristic information and the seller characteristic information is generated; Obtain advertising exposure information, which indicates the exposure history of the advertisement to be delivered; as well as Based on the interaction information, the ad exposure information, and the pre-determined human characteristics, the ad delivery strategy is determined.
2. The advertising delivery method according to claim 1, wherein, The acquisition of consumer characteristic information includes: Obtain the purchase history of the target consumer; Based on the purchase history, determine the single-purchase characteristic information of the target consumer; and Based on the purchase history, the multiple purchase characteristics of the target consumer are determined.
3. The advertising delivery method according to claim 2, wherein, The determination of the single-purchase characteristic information of the target consumer based on the purchase history includes: The purchase history is used as input to the first extraction model, and the first extraction model is used to output the single purchase feature information. The first extraction model is used to extract features from the purchase history.
4. The advertising delivery method according to claim 3, wherein, The first extraction model includes an attention neural network model.
5. The advertising placement method according to any one of claims 2-4, wherein, The determination of the target consumer's multiple purchase characteristics based on the purchase history includes: The purchase history and / or the single purchase feature information are used as input to the second extraction model, and the multiple purchase feature information is output using the second extraction model. The second extraction model is used to extract features from the purchase history.
6. The advertising delivery method according to claim 5, wherein, The second extraction model includes an attention neural network model.
7. The advertising placement method according to any one of claims 1-6, wherein, The acquisition of seller characteristic information includes: Obtain the sales strategy of the seller; and Based on the sales strategy, determine the seller's sales characteristic information.
8. The advertising delivery method according to claim 7, wherein, The process of determining the seller's sales characteristic information based on the sales strategy includes: The sales strategy is used as input to the third extraction model, and the third extraction model is used to output the salesperson feature information. The third extraction model is used to extract features from the sales strategy.
9. The advertising delivery method according to claim 8, wherein, The third extraction model includes a fully connected layer model.
10. The advertising delivery method according to any one of claims 1-9, wherein, The step of generating interactive information for the consumer characteristic information and the seller characteristic information based on the consumer characteristic information and the seller characteristic information includes: The consumer feature information and the seller feature information are input into a fourth extraction model, which is used to perform cross-feature extraction on the consumer feature information and the seller feature information; and The interaction information is generated using the fourth extraction model.
11. The advertising delivery method according to claim 10, wherein, The fourth extraction model includes the cross-attention model.
12. The advertising delivery method according to any one of claims 1-11, wherein, The acquisition of advertising exposure information includes: Obtain the exposure history of the advertisement to be delivered; and The exposure history is used as input to the fifth extraction model, and the fifth extraction model is used to output the advertisement exposure information. The fifth extraction model is used to extract features from the exposure history.
13. The advertising delivery method according to claim 12, wherein, The fifth extraction model includes a gated neural network model or a long short-term memory neural network model.
14. The advertising delivery method according to any one of claims 1-13, wherein, The step of determining the delivery strategy for the advertisement to be delivered based on the interaction information, the advertisement exposure information, and pre-determined human characteristics includes: Based on the interaction information, the ad exposure information, and the human characteristics, the conversion probability of the target consumer is determined; and The delivery strategy is determined based on the conversion probability.
15. The advertising delivery method according to claim 14, wherein, Determining the conversion probability of the target consumer based on the interaction information, the ad exposure information, and the human characteristics includes: The interaction information, the ad exposure information, and the artificial features are input into a prediction model, which is used to predict the delivery strategy for the ad to be delivered; and The prediction model is used to generate the conversion probability category and / or conversion probability feature vector of the target consumer, wherein the conversion probability category indicates the conversion probability corresponding to at least one delivery strategy category, and the conversion probability feature vector includes a vector corresponding to the conversion probability obtained based on the conversion probability.
16. The advertising delivery method according to claim 14 or 15, further comprising: After determining the transformation probability, perform at least one of the following steps: Based on the predetermined number of target consumers, determine the number of convertible consumers among the target consumers; Based on the predetermined number of consumers to be converted, determine the number of target consumers, including those to be converted.
17. An advertising delivery device, comprising: The first acquisition module is used to acquire consumer characteristic information, which indicates the purchase history of the target consumers to be advertised. The second acquisition module is used to acquire seller characteristic information, which indicates the sales strategy of the seller to be advertised. An interaction information determination module is used to generate interaction information between the consumer characteristic information and the seller characteristic information based on the consumer characteristic information and the seller characteristic information; The third acquisition module is used to acquire advertising exposure information, which indicates the exposure history of the advertisement to be delivered; as well as The delivery strategy determination module is used to determine the delivery strategy of the advertisement to be delivered based on the interaction information, the advertisement exposure information and the pre-determined human characteristics.
18. A computing device, comprising: At least one processor; as well as At least one memory communicatively connected to the at least one processor, the at least one memory storing instructions that, when executed individually or jointly by the at least one processor, cause the computing device to perform the advertising delivery method according to any one of claims 1 to 16.
19. A computer-readable storage medium storing instructions that, when executed individually or jointly by one or more processors of a computing device, cause the computing device to perform the advertising delivery method according to any one of claims 1 to 16.
20. A computer program product comprising instructions that, when executed individually or jointly by one or more processors of a computing device, cause the computing device to perform the advertising delivery method of any one of claims 1 to 16.