Advertisement recommendation method, device, equipment and medium

Through the multi-task advertising recommendation model, the network structure layer of shared and independent layers is utilized, combined with multi-task learning, the problems of large workload and poor recommendation results during different types of advertising recommendations are solved, and the accuracy and user experience of advertising recommendations are improved.

CN114820060BActive Publication Date: 2024-12-03BEIJING YOUZHUJU NETWORK TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202210445927.0
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-04-26
Publication Date
2024-12-03
Estimated Expiration
2042-04-26

AI Technical Summary

Technical Problem

When the prior art recommends different types of advertisements to users, it is necessary to implement them separately for each type of advertisement, which leads to a large workload and affects the recommendation effect, and ignores the correlation between different types of advertisements, affecting the accuracy of recommendations.

Method used

Using a multi-task advertising recommendation model, multiple prediction models combine sharing layer and independent layer, and shared network structure layer to predict click probability of different types of advertisements. Multi-task learning is used to correlate data of different types of advertisements to generate multi-task advertising recommendations. model, and then recommend.

Benefits of technology

Improve the accuracy and user experience of advertising recommendations. By sharing information and linking multiple prediction tasks, the individual training needs for different types of advertisements are reduced, and the accuracy of recommendations is improved.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114820060B_ABST
    Figure CN114820060B_ABST
Patent Text Reader

Abstract

The present application discloses an advertisement recommendation method, which obtains multiple advertisement data when different types of advertisements need to be recommended to users. The multiple advertisement data include advertisement data of different types of advertisements. The multiple advertisement data are input into a multi-task advertisement recommendation model to predict the user's click probability for different types of advertisements through the model, and then recommend to the user based on the predicted click probability. It can be seen that the present application can share information between different types of advertisements through the multi-task advertisement recommendation model, associate multiple prediction tasks, thereby improving the accuracy of the prediction, and then recommend content of interest to users, thereby improving the user experience.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to an advertisement recommendation method, device, equipment and medium. Background Art

[0002] With the continuous development of Internet technology, there are more and more ways to place advertisements, and accordingly many different types of advertisements are generated. For example, banner ads, splash ads, interstitial ads, information flow ads, incentive video ads, etc. Usually, different types of advertisements have different delivery performances. When different types of advertisements need to be recommended to users, each type of advertisement needs to be implemented separately. This implementation not only brings a large workload, but also affects the recommendation effect. Summary of the invention

[0003] In view of this, embodiments of the present application provide an advertisement recommendation method, apparatus, device, and medium to recommend advertisements to users through a multi-task advertisement recommendation model and improve the recommendation effect.

[0004] To achieve the above objectives, the technical solutions provided by this application are as follows:

[0005] In a first aspect of the present application, an advertisement recommendation method is provided, the method comprising:

[0006] Acquire multiple pieces of advertisement data, wherein the multiple pieces of advertisement data include advertisement data of different types of advertisements, and the distribution of the different types of advertisements is different;

[0007] Inputting the plurality of advertisement data into a multi-task advertisement recommendation model to obtain a click probability of each advertisement corresponding to each type of advertisement, wherein the multi-task advertisement recommendation model is used to obtain the click probability of different types of advertisements;

[0008] For any type of advertisement, recommend to the user according to the click probability of each advertisement in the type of advertisement;

[0009] The multi-task advertising recommendation model includes multiple prediction models, each of the multiple prediction models is used to predict the click probability of a type of advertisement, and each prediction model includes a shared layer and an independent layer. The shared layer refers to a network structure layer used by different prediction models.

[0010] In a second aspect of the present application, an advertisement recommendation device is provided, the device comprising:

[0011] A first acquisition unit is used to acquire a plurality of pieces of advertisement data, wherein the plurality of pieces of advertisement data include advertisement data of different types of advertisements, and the distribution of the different types of advertisements is different;

[0012] A second acquisition unit is used to input the plurality of advertisement data into a multi-task advertisement recommendation model to obtain a click probability of each advertisement corresponding to each type of advertisement, wherein the multi-task advertisement recommendation model is used to obtain click probabilities of different types of advertisements;

[0013] A recommendation unit, configured to recommend any type of advertisement to the user according to the click probability of each advertisement in the type of advertisement;

[0014] The multi-task advertising recommendation model includes multiple prediction models, each of the multiple prediction models is used to predict the click probability of a type of advertisement, and each prediction model includes a shared layer and an independent layer. The shared layer refers to a network structure layer used by different prediction models.

[0015] In the third aspect of the present application, an electronic device is provided, comprising: a processor and a memory; the memory is used to store instructions or computer programs; the processor is used to execute the instructions or computer programs in the memory, so that the electronic device executes the method described in the first aspect.

[0016] In a fourth aspect of the present application, a computer-readable storage medium is provided, wherein instructions are stored in the computer-readable storage medium. When the instructions are executed on a device, the device executes the method described in the first aspect.

[0017] In a fifth aspect of the present application, a computer program product is provided, wherein the computer program product comprises a computer program / instructions, and when the computer program / instructions are executed by a processor, the method described in the first aspect is implemented.

[0018] It can be seen that this application has the following beneficial effects:

[0019] In the present application, a multi-task advertising recommendation model is pre-trained and generated, and the multi-task advertising recommendation model is used to predict the user's click probability for different types of advertisements. When it is necessary to recommend different types of advertisements to users, multiple pieces of advertising data are obtained. Among them, the multiple pieces of advertising data include advertising data of different types of advertisements. The multiple pieces of advertising data are input into the multi-task advertising recommendation model to predict the user's click probability for different types of advertisements through the model, and then recommend to the user based on the predicted click probability. It can be seen that the present application can share information between different types of advertisements through the multi-task advertising recommendation model, associate multiple prediction tasks, thereby improving the accuracy of the prediction, and then recommend content of interest to users, thereby improving the user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0021] Figure 1 A flow chart of a method for generating a multi-task advertising recommendation model provided in an embodiment of the present application;

[0022] Figure 2 A schematic diagram of a flow chart of an advertisement recommendation method provided in an embodiment of the present application;

[0023] Figure 3 A schematic diagram of a network structure of a multi-task advertising recommendation model provided in an embodiment of the present application;

[0024] Figure 4 A schematic diagram of the structure of an advertisement recommendation device provided in an embodiment of the present application;

[0025] Figure 5 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0026] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0027] In the field of advertising, as the types of advertisements increase, different types of advertisements perform differently in actual delivery. Common types of advertisements include banner ads, interstitial ads, splash ads, information flow ads, patch ads, and rewarded video ads.

[0028] Among them, banner ads can be presented at the top or bottom of the application (APP) interface; they can be presented in the form of static images, GIF images, text links, or animated images with multiple images scrolling. Interstitial ads can pop up in half-screen or full-screen form when the application is opened, paused, passed, jumped, or exited; they mainly include static images, GIF images, videos, etc. Splash screen ads refer to full-screen or half-screen ads that appear when the APP is started and loaded; they can be presented in the form of static images, dynamic videos, Flash, etc. Information flow ads refer to ads inserted when users are browsing or reading content. They mainly include text + small pictures, text + large pictures, text + single pictures, text + multiple pictures, text + video / short video, etc. Incentive video ads are ads that trigger users to get corresponding rewards or user rights after watching the video. Patch ads refer to ads that are attached to publicly screened or broadcast movies, TV programs, and online videos.

[0029] Since different types of ads have different presentation forms and different attractiveness to users, the distribution of different types of ads among users varies. For example, interstitial ads are a common form of ad presentation, and have higher click-through rates and conversion rates than banner ads. Compared with information flow ads, interstitials may be less native, but have more prominent attention and exposure effects. When you need to recommend different types of ads to users, you need to create independent recommendation tasks for different types of ads.

[0030] However, if recommendations for different types of advertisements are processed as independent tasks, the correlation between the data used in making recommendations for different types of advertisements will be ignored, which will affect the accuracy of the recommendations.

[0031] Based on this, an embodiment of the present application provides an advertising recommendation method, which combines recommendation tasks for different types of advertisements based on a multi-task learning approach to learn, so as to generate a multi-task advertising recommendation model using data training under different distributions, and to use the multi-task advertising recommendation model to realize recommendations for different types of advertisements. There is no need to train separately for a certain type of advertisement, so that when recommending, it is possible to combine multi-party data to predict advertisements that users are interested in, thereby improving the accuracy of recommendations.

[0032] Among them, multi-task learning: in simple terms, multiple objective functions are learned simultaneously. Through multi-task learning, it is determined which network structure layers in the multi-task advertising recommendation model are shared between certain tasks and which network structure layers are unique to certain tasks. This not only enables the sharing of related information, but also improves the accuracy of a certain task.

[0033] It is understandable that before using the technical solutions disclosed in the embodiments of this application, the type, scope of use, usage scenarios, etc. of the personal information involved in this application should be informed to the user and the user's authorization should be obtained in an appropriate manner in accordance with relevant laws and regulations.

[0034] For example, in response to receiving an active request from a user, a prompt message is sent to the user to clearly prompt the user that the operation requested to be performed will require obtaining and using the user's personal information. Thus, the user can autonomously choose whether to provide personal information to software or hardware such as an electronic device, application, server, or storage medium that performs the operation of the technical solution of the present disclosure according to the prompt message.

[0035] As an optional but non-limiting implementation, in response to receiving an active request from the user, the prompt information may be sent to the user in the form of a pop-up window, in which the prompt information may be presented in text form. In addition, the pop-up window may also carry a selection control for the user to choose "agree" or "disagree" to provide personal information to the electronic device.

[0036] It is understandable that the above notification and the process of obtaining user authorization are merely illustrative and do not constitute a limitation on the implementation of the present disclosure. Other methods that meet the relevant laws and regulations may also be applied to the implementation of the present disclosure.

[0037] To facilitate understanding of the technical solution provided by the embodiments of the present application, the training process of the multi-task advertising recommendation model in the present application will be described below.

[0038] See also Figure 1 , which is a flow chart of a method for generating a multi-task advertising recommendation model provided by an embodiment of the present application, such as Figure 1 As shown, the method may include:

[0039] S101: Acquire a training sample set, where the training sample set includes advertisement data corresponding to advertisements of different types and click data corresponding to a target advertisement.

[0040] In this embodiment, in order to train and generate a multi-task advertising recommendation model, a large amount of advertising data corresponding to different types of advertisements and click data corresponding to target advertisements can be obtained. For any type of advertisement, advertising data of multiple advertisements of that type can be obtained. Among them, the advertising data may include advertising words, advertising image data, etc. Click data refers to the data corresponding to when a user clicks on a target advertisement, and the data may include user information, browsing time, number of clicks, etc. Among them, user information is used to uniquely indicate a certain user, and click data is obtained when the user clicks on an advertisement with the user's authorization.

[0041] For example, for banner ads, the advertisement data of N1 different banner ads and the click data corresponding to each of the N1 advertisement data are obtained. For interstitial ads, the advertisement data of N2 different interstitial ads and the click data corresponding to each of the N2 advertisement data are obtained. The click data corresponding to each advertisement data may include click data of multiple users.

[0042] S102: Train the initial network model according to the training sample set to generate a multi-task advertising recommendation model.

[0043] After obtaining the training sample set, the initial network model is trained according to the training sample set to generate a multi-task advertising recommendation model. The multi-task advertising recommendation model is used to predict the user's click probability on different types of advertisements, that is, the multi-task advertising recommendation model can achieve the task of recommending different types of advertisements to users.

[0044] Optionally, during training, in response to the network structure of the initial network model being in a locked state, the weight parameters of the initial network model are trained using a training sample set to obtain a first network model; in response to the weight parameters of the first network model being in a locked state, the network structure of the first network model is trained using a training sample set to determine the network structures corresponding to different types of advertisements, and obtain a multi-task advertising recommendation model.

[0045] In this example, during training, the network structure of the initial network model can be fixed in the first training stage, and the weight parameters of the model can be trained using the training sample set, so that the weight parameters of each network structure can be fully learned. After the first stage of training is completed, the first network model is obtained and the second stage is entered. In the second training stage, the weight parameters of the first network model are fixed, and the network structure of the model is trained using the training sample set, so that the network structures corresponding to different types of advertisements are determined. That is, when predicting the click probability of a certain type of advertisement, it is determined which network structures will be used and which structures will not be used.

[0046] Among them, the multi-task advertising recommendation model generated by training includes multiple prediction models, each of which is used to predict the click probability of a type of advertisement, and each prediction model includes a shared layer and an independent layer. Among them, the shared layer refers to the network structure layer used by different prediction models, and the independent layer is the network structure layer used independently by a certain prediction model. Therefore, when training and generating a multi-task advertising recommendation model, the training will be carried out with the goal of seeking an adaptive feature sharing mechanism, which determines which network layers should be shared between which tasks and which layers should be task-specific. That is, the adaptive feature sharing mechanism is used to determine the shared layer and independent layer corresponding to the prediction model.

[0047] For example, the initial network model includes 5 network structure layers, namely block1-block5. Through the above two-stage training, it is determined that the prediction model for banner ads includes block1, block2 and blcok5; the prediction model for interstitial ads includes block1, block3 and blcok4; the prediction model for information flow ads includes block1, block2 and blcok3. Then, block1, block2 and block3 are shared layers, block4 is a unique network structure layer for the interstitial ad prediction model, and block5 is a unique network structure layer for the banner ad prediction model.

[0048] Optionally, considering that click data is structured data, the multi-task advertisement recommendation model in this embodiment is composed of multiple multi-layer perceptron (MLP) network structures.

[0049] It can be seen that through the above embodiment, a multi-task advertising recommendation model can be trained and generated to achieve the recommendation of multiple different types of advertisements through the multi-task advertising recommendation model, without the need to perform separate training for the recommendation tasks of different types of advertisements. Not only can the information sharing between different types of advertisements be achieved, but also the accuracy of the recommendation can be improved.

[0050] Based on the multi-task advertisement recommendation model generated by the above training, an embodiment of the present application provides an advertisement recommendation method, which will be described below in conjunction with the attached examples.

[0051] See also Figure 2 , which is a flow chart of an advertisement recommendation method provided in an embodiment of the present application. Figure 2 As shown, the method may include:

[0052] S201: Acquire multiple pieces of advertisement data.

[0053] In this embodiment, in order to ensure that the advertisements recommended to the user meet the user's expectations, user data can be first obtained with the user's authorization, and the user data is used to reflect the user's characteristics, such as the user's preferences or interests. At the same time, advertisement data corresponding to the multiple advertisements to be recommended are obtained. The multiple advertisement data include advertisement data of different types of advertisements, and the distribution of different types of advertisements is different.

[0054] Among them, since it is necessary to recommend different types of advertisements, it is necessary to obtain the advertisement data of multiple advertisements under each type of advertisement. For example, for banner advertisements, the advertisement data of 100 different banner advertisements are obtained so as to select one or more advertisements from the 100 advertisements to recommend to the user; for splash screen advertisements, the advertisement data of 200 different splash screen advertisements are obtained so as to select one or more advertisements from the 200 advertisements to recommend to the user.

[0055] S202: Inputting a plurality of advertisement data into a multi-task advertisement recommendation model to obtain a click probability of each advertisement corresponding to each type of advertisement.

[0056] When user data is obtained, the user data and multiple advertisement data are input into the multi-task advertisement recommendation model, so as to predict the user's click probability for any of the multiple advertisements input by the multi-task advertisement recommendation model. That is, the multi-task advertisement recommendation model can make predictions for different types of advertisements.

[0057] The multi-task advertising recommendation model includes multiple prediction models, each of which is used to predict the click-through rate of a type of advertisement. Different prediction models include different network structure layers. Usually, a prediction model includes a shared layer and an independent layer. The shared layer refers to the network structure layer used by multiple prediction layers, and the independent layer is a network structure layer unique to a certain prediction.

[0058] Optionally, multiple pieces of advertisement data are input into the multi-task advertisement recommendation model to obtain the click probability of each advertisement corresponding to each type of advertisement, including: for the advertisement data of any advertisement, determining the target prediction model according to the type of the advertisement; inputting the advertisement data into the target prediction model to obtain the click probability of the advertisement. Since the network structure layers included in the models performing different types of advertisement prediction tasks are different, in order to achieve accurate prediction, when performing the prediction task, the prediction model matching the advertisement type of the received advertisement data must first be determined, and then the probability of the user clicking on the advertisement of the type must be predicted using the matched prediction model.

[0059] For example, the multi-task advertising recommendation model includes a banner advertising prediction model, an interstitial advertising prediction model, and an information flow advertising prediction model. Among them, the banner advertising prediction model includes block1, block2, and blcok5; the interstitial advertising prediction model includes block1, block3, and blcok4; the information flow advertising prediction model includes block1, block2, and blcok3. There are three tasks, Task 1 predicts the click probability of each banner ad in multiple banner ads, Task 2 predicts the click probability of each interstitial ad in multiple interstitial ads, and Task 2 predicts the click probability of each information flow ad in multiple information flow ads.

[0060] When user data is obtained, the user data and advertisement data are input into the target prediction model to obtain the click probability of the advertisement.

[0061] S203: For any type of advertisement, recommend to the user according to the click probability of each advertisement in the type of advertisement.

[0062] After obtaining the click probabilities for different types of ads through the multi-task ad recommendation model, for any type of ads, ads of that type are recommended to users based on the click probability of each ad in that type of ads. Specifically, ads with the highest click probability are recommended to users, or ads with a click probability greater than a preset probability threshold are recommended to users.

[0063] For example, task 1 outputs a click probability of 0.2 for ad 1, 0.6 for ad 2, and 0.8 for ad 3; task 2 outputs a click probability of 0.3 for ad 4, 0.1 for ad 5, and 0.6 for ad 6; task 3 outputs a click probability of 0.2 for ad 7, 0.5 for ad 8, and 0.7 for ad 9. Then ad 3 in the banner ad is recommended to the user, ad 6 in the interstitial ad is recommended to the user, and ad 9 in the information flow ad is recommended to the user.

[0064] It can be seen that when it is necessary to recommend different types of advertisements to users, user data and multiple pieces of advertisement data can be obtained. Among them, user data is used to reflect the characteristics of users, and multiple pieces of advertisement data include advertisement data of different types of advertisements. The user data and multiple pieces of advertisement data are input into the multi-task advertisement recommendation model to predict the user's click probability for different types of advertisements through the model, and then recommend to the user based on the predicted click probability. It can be seen that the present application can share information between different types of advertisements through the multi-task advertisement recommendation model, associate multiple prediction tasks, thereby improving the accuracy of the prediction, and then recommend content of interest to users, thereby improving the user experience.

[0065] For easier understanding, see Figure 3 The network structure diagram of a multi-task advertising recommendation model shown in FIG. 1 includes an input layer, Block 1-Block 4, and an output layer. Figure 3 There are three tasks, namely Task 1, Task 2 and Task 3. The processing of Task 1 includes passing through the input layer, Block 1, Block 4 and the output layer, the processing of Task 2 includes passing through the input layer, Block 2, Block 3 and the output layer, and the processing of Task 3 includes passing through the input layer, Block 1, Block 2 and the output layer.

[0066] Based on the above method embodiment, the embodiment of the present application provides an advertisement recommendation device, which will be described below in conjunction with the accompanying drawings.

[0067] See also Figure 4 , which is a structural diagram of an advertisement recommendation device provided in an embodiment of the present application, such as Figure 4 As shown, the device 400 may include: a first acquisition unit 401 , a second acquisition unit 402 , and a recommendation unit 403 .

[0068] A first acquisition unit 401 is used to acquire multiple pieces of advertisement data, wherein the multiple pieces of advertisement data include advertisement data of different types of advertisements, and the distribution of the different types of advertisements is different;

[0069] A second acquisition unit 402 is used to input the plurality of advertisement data into a multi-task advertisement recommendation model to obtain a click probability of each advertisement corresponding to each type of advertisement, wherein the multi-task advertisement recommendation model is used to obtain click probabilities of different types of advertisements;

[0070] The recommendation unit 403 is used to recommend any type of advertisement to the user according to the click probability of each advertisement in the type of advertisement.

[0071] Among them, the multi-task advertising recommendation model includes multiple prediction models, each of the multiple prediction models is used to predict the click probability of a type of advertisement, and each prediction model includes a shared layer and an independent layer. The shared layer refers to a network structure layer used by different prediction models.

[0072] In a possible implementation, the second acquisition unit is specifically configured to determine a target prediction model for the advertisement data of any advertisement according to the type of the advertisement; input the advertisement data into the target prediction model to obtain the click probability of the advertisement.

[0073] In a possible implementation, the device further includes: a third acquisition unit and a generation unit;

[0074] A third acquisition unit is used to acquire a training sample set, wherein the training sample set includes advertisement data corresponding to advertisements of different types and click data corresponding to a target advertisement, wherein the target advertisement is any advertisement of the different types;

[0075] A generating unit is used to train an initial network model according to the training sample set to generate the multi-task advertisement recommendation model.

[0076] In one possible implementation, the generating unit is specifically used to, in response to the network structure of the initial network model being in a locked state, train the weight parameters of the initial network model using the training sample set to obtain a first network model; in response to the weight parameters of the first network model being in a locked state, train the network structure of the first network model using the training sample set to determine the network structures corresponding to different types of advertisements, and obtain the multi-task advertising recommendation model.

[0077] In one possible implementation, the generating unit is specifically used to train the initial network model according to the training sample set with the goal of seeking an adaptive feature sharing mechanism to generate the multi-task advertising recommendation model, and the adaptive feature sharing mechanism is used to determine the shared layer and independent layer corresponding to the prediction model.

[0078] In a possible implementation, the multi-task advertisement recommendation model is a multi-layer perceptron MLP network.

[0079] In a possible implementation, the different types of advertisements include banner ads, interstitial ads, splash ads, information flow ads, patch ads, and incentive video ads.

[0080] It should be noted that the implementation of each unit in this embodiment can refer to the relevant description in the above method embodiment, and this embodiment will not be repeated here.

[0081] See also Figure 5 , which shows a schematic diagram of the structure of an electronic device 500 suitable for implementing the embodiment of the present application. The terminal device in the embodiment of the present application may include but is not limited to mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (portable android devices), PMPs (Portable Media Players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs (televisions), desktop computers, etc. Figure 5The electronic device shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0082] like Figure 5 As shown, the electronic device 500 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 501, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 502 or a program loaded from a storage device 508 into a random access memory (RAM) 503. In the RAM 503, various programs and data required for the operation of the electronic device 500 are also stored. The processing device 501, the ROM 502, and the RAM 503 are connected to each other via a bus 504. An input / output (I / O) interface 505 is also connected to the bus 504.

[0083] Typically, the following devices may be connected to the I / O interface 505: an input device 506 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 507 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 508 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 509. The communication device 509 may allow the electronic device 500 to communicate with other devices wirelessly or by wire to exchange data. Although Figure 5 The electronic device 500 is shown with various devices, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead.

[0084] In particular, according to an embodiment of the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present application includes a computer program product, which includes a computer program carried on a non-transitory computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication device 509, or installed from the storage device 508, or installed from the ROM502. When the computer program is executed by the processing device 501, the above-mentioned functions defined in the method of the embodiment of the present application are executed.

[0085] The electronic device provided in the embodiment of the present application and the advertisement recommendation method provided in the above embodiment belong to the same inventive concept. The technical details not fully described in this embodiment can be referred to the above embodiment, and this embodiment has the same beneficial effects as the above embodiment.

[0086] An embodiment of the present application provides a computer-readable medium having a computer program stored thereon, wherein when the program is executed by a processor, the advertisement recommendation method as described in any of the above embodiments is implemented.

[0087] It should be noted that the computer-readable medium mentioned above in the present application may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present application, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The computer readable signal medium may also be any computer readable medium other than a computer readable storage medium, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer readable medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0088] In some embodiments, the client and the server may communicate using any currently known or future developed network protocol such as HTTP (Hyper Text Transfer Protocol), and may be interconnected with any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), an internet (e.g., the Internet), and a peer-to-peer network (e.g., an ad hoc peer-to-peer network), as well as any currently known or future developed network.

[0089] The computer-readable medium may be included in the electronic device, or may exist independently without being incorporated into the electronic device.

[0090] The computer-readable medium carries one or more programs. When the one or more programs are executed by the electronic device, the electronic device executes the advertisement recommendation method.

[0091] Computer program code for performing the operations of the present application may be written in one or more programming languages ​​or a combination thereof, including, but not limited to, object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0092] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present application. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0093] The units involved in the embodiments described in this application may be implemented by software or hardware. The name of a unit / module does not limit the unit itself in some cases. For example, a voice data acquisition module may also be described as a "data acquisition module".

[0094] The functions described above herein may be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), and the like.

[0095] In the context of the present application, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0096] It should be noted that the various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments, and the same or similar parts between the various embodiments can be referred to each other. For the system or device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part description.

[0097] It should be understood that in the present application, "at least one (item)" means one or more, and "plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the objects associated before and after are in an "or" relationship. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0098] It should also be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0099] The steps of the method or algorithm described in conjunction with the embodiments disclosed herein may be implemented directly using hardware, a software module executed by a processor, or a combination of the two. The software module may be placed in a random access memory (RAM), a memory, a read-only memory (ROM), an electrically programmable ROM, an electrically erasable programmable ROM, a register, a hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.

[0100] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. An advertisement recommendation method, characterized in that: The method comprises: Acquire multiple pieces of advertisement data, wherein the multiple pieces of advertisement data include advertisement data of different types of advertisements, the distribution of the different types of advertisements is different, and the advertisement data includes advertisement words and / or advertisement image data; Inputting the plurality of advertisement data into a multi-task advertisement recommendation model to obtain a click probability of each advertisement corresponding to each type of advertisement, wherein the multi-task advertisement recommendation model is used to obtain the click probability of different types of advertisements to achieve the task of recommending different types of advertisements to users; For any type of advertisement, recommend to the user according to the click probability of each advertisement in the type of advertisement; Among them, the multi-task advertising recommendation model includes multiple prediction models, each of the multiple prediction models is used to predict the click probability of a type of advertisement, and each prediction model includes a shared layer and an independent layer. The shared layer refers to a network structure layer used by different prediction models. The shared layer enables information between different types of advertisements to be shared and associates multiple prediction tasks. The independent layer is a network structure layer used independently by each prediction model.

2. The method according to claim 1, characterized in that The step of inputting the plurality of advertisement data into a multi-task advertisement recommendation model to obtain a click probability of each advertisement corresponding to each type of advertisement includes: For any advertisement data, determining a target prediction model according to the type of the advertisement; The advertisement data is input into the target prediction model to obtain the click probability of the advertisement.

3. The method according to claim 1, characterized in that The training process of the multi-task advertising recommendation model includes: Acquire a training sample set, wherein the training sample set includes advertisement data corresponding to advertisements of different types and click data corresponding to a target advertisement, wherein the target advertisement is any advertisement of the different types; The initial network model is trained according to the training sample set to generate the multi-task advertisement recommendation model.

4. The method according to claim 3, characterized in that The step of training the initial network model according to the training sample set to generate the multi-task advertisement recommendation model includes: In response to the network structure of the initial network model being in a locked state, training weight parameters of the initial network model using the training sample set to obtain a first network model; In response to the weight parameters of the first network model being in a locked state, the network structure of the first network model is trained using the training sample set to determine network structures corresponding to different types of advertisements, thereby obtaining the multi-task advertisement recommendation model.

5. The method according to claim 3 or 4, characterized in that: The step of training the initial network model according to the training sample set to generate the multi-task advertisement recommendation model includes: With the goal of seeking an adaptive feature sharing mechanism, the initial network model is trained according to the training sample set to generate the multi-task advertising recommendation model, and the adaptive feature sharing mechanism is used to determine the shared layer and the independent layer corresponding to the prediction model.

6. The method according to claim 1, characterized in that The multi-task advertisement recommendation model is a multi-layer perceptron MLP network.

7. The method according to claim 1, characterized in that The different types of ads include banner ads, interstitial ads, splash ads, information flow ads, patch ads, and incentive video ads.

8. An advertisement recommendation device, characterized in that: The device comprises: A first acquisition unit is used to acquire a plurality of pieces of advertisement data, wherein the plurality of pieces of advertisement data include advertisement data of different types of advertisements, the distribution of the different types of advertisements is different, and the advertisement data includes advertisement words and / or advertisement image data; A second acquisition unit is used to input the plurality of advertisement data into a multi-task advertisement recommendation model to obtain a click probability of each advertisement corresponding to each type of advertisement, wherein the multi-task advertisement recommendation model is used to obtain the click probability of different types of advertisements to achieve the task of recommending different types of advertisements to users; A recommendation unit, used for recommending any type of advertisement to a user according to the click probability of each advertisement in the type of advertisement; Among them, the multi-task advertising recommendation model includes multiple prediction models, each of the multiple prediction models is used to predict the click probability of a type of advertisement, and each prediction model includes a shared layer and an independent layer. The shared layer refers to a network structure layer used by different prediction models. The shared layer enables information between different types of advertisements to be shared and associates multiple prediction tasks. The independent layer is a network structure layer used independently by each prediction model.

9. An electronic device, characterized in that: The device comprises: a processor and a memory; The memory is used to store instructions or computer programs; The processor is used to execute the instructions or computer programs in the memory so that the electronic device executes the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores instructions, and when the instructions are executed on a device, the device executes the method according to any one of claims 1 to 7.

11. A computer program product, characterized in that The computer program product comprises a computer program / instruction, and when the computer program / instruction is executed by a processor, the method according to any one of claims 1 to 7 is implemented.