Advertising model management method, device, electronic device and computer-readable medium

Abstract

The embodiments of the present application provide a management method, device, electronic device and computer-readable medium for an advertising model, and relate to the field of Internet technology. The method comprises: sending an advertising model issuance request to an advertising server, so that the advertising server sends the advertising model and the category information of the input features of the advertising model to the advertising client based on the advertising model issuance request; receiving the advertising model and the category information of the input features of the advertising model issued by the advertising server based on the advertising model issuance request; obtaining the input feature information of the advertising model based on the category information of the input features of the advertising model; inputting the input feature information of the advertising model into the advertising model to obtain the result predicted by the advertising model based on the input feature information. Through the embodiments of the present application, not only the timeliness of the input features of the advertising model can be effectively improved, but also the success rate of requesting advertising model prediction can be effectively improved.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of Internet technology, and in particular to a management method, device, electronic device, and computer-readable medium for an advertising model. Background Art

[0002] Advertisements are ubiquitous in our lives, and advertisers have a need to push ads to users to promote their products or services. With the widespread use of the internet and computers, online advertising has become a new form of advertising. In online advertising systems, advertisers pay publishers to place their ads through web pages, browsers, ad clients, or other online media. With the continuous development of international business, online advertising systems handle billions of ad requests daily.

[0003] In order to achieve intelligent advertising decision-making and increase advertising revenue, online advertising systems use advertising models with intelligent functions to achieve intelligent advertising decision-making and increase advertising revenue. However, the international network environment in which the terminal devices of the clients in the online advertising system are located is extremely unbalanced and extremely unstable. Most international regions are in a weak network environment (high latency, high packet loss, secondary network, tertiary network) intermittently or for a long time. In view of this international network environment, the prediction results of the server-side advertising model are often unable to be returned to the client requesting the advertising model prediction due to the network environment, or the time taken to return to the client requesting the advertising model prediction is too long. This will lead to the inability to meet the advertising function requirements. It can be seen that how to effectively improve the success rate of requesting advertising model predictions has become a technical problem that needs to be solved urgently. Summary of the Invention

[0004] The purpose of this application is to propose an advertising model management method, device, electronic device and computer-readable medium to solve the technical problem existing in the prior art of how to effectively improve the success rate of requesting advertising model predictions.

[0005] According to the first aspect of the embodiment of the present application, a method for managing an advertising model is provided. The method includes: sending an advertising model issuance request to an advertising server, so that the advertising server sends the advertising model and category information of the input features of the advertising model to the advertising client based on the advertising model issuance request; receiving the advertising model and category information of the input features of the advertising model issued by the advertising server based on the advertising model issuance request; obtaining input feature information of the advertising model based on the category information of the input features of the advertising model; inputting the input feature information of the advertising model into the advertising model to obtain a result predicted by the advertising model based on the input feature information.

[0006] According to a second aspect of an embodiment of the present application, a method for managing an advertising model is provided. The method comprises: receiving an advertising model delivery request sent by an advertising client; and based on the advertising model delivery request, delivering the advertising model and category information of the input features of the advertising model to the advertising client, so that the advertising client obtains the input feature information of the advertising model based on the category information of the input features of the advertising model, and inputs the input feature information of the advertising model into the advertising model to obtain a result predicted by the advertising model based on the input feature information.

[0007] According to the third aspect of the embodiment of the present application, a method for managing an advertising model is provided. The method includes: sending an advertising model issuance request to an advertising Internet of Things device, so that the advertising Internet of Things device issues the advertising model and category information of the input features of the advertising model to the advertising client based on the advertising model issuance request; receiving the advertising model and category information of the input features of the advertising model issued by the advertising Internet of Things device based on the advertising model issuance request; obtaining the input feature information of the advertising model based on the category information of the input features of the advertising model; inputting the input feature information of the advertising model into the advertising model to obtain the result predicted by the advertising model based on the input feature information.

[0008] According to the fourth aspect of the embodiment of the present application, a device for managing an advertising model is provided. The device includes: a first sending module for sending an advertising model sending request to an advertising server, so that the advertising server sends the advertising model and the category information of the input features of the advertising model to the advertising client based on the advertising model sending request; a first receiving module for receiving the advertising model and the category information of the input features of the advertising model sent by the advertising server based on the advertising model sending request; a first acquiring module for acquiring the input feature information of the advertising model based on the category information of the input features of the advertising model; and a first input module for inputting the input feature information of the advertising model into the advertising model to obtain the result predicted by the advertising model based on the input feature information.

[0009] According to a fifth aspect of an embodiment of the present application, a device for managing an advertising model is provided. The device includes: a second receiving module for receiving an advertising model sending request sent by an advertising client; a first sending module for sending an advertising model and category information of input features of the advertising model to the advertising client based on the advertising model sending request, so that the advertising client obtains the input feature information of the advertising model based on the category information of the input features of the advertising model, and inputs the input feature information of the advertising model into the advertising model to obtain a result predicted by the advertising model based on the input feature information.

[0010] According to the sixth aspect of the embodiment of the present application, a device for managing an advertising model is provided. The device includes: a second sending module for sending an advertising model sending request to an advertising Internet of Things device, so that the advertising Internet of Things device sends the advertising model and the category information of the input features of the advertising model to the advertising client based on the advertising model sending request; a third receiving module for receiving the advertising model and the category information of the input features of the advertising model sent by the advertising Internet of Things device based on the advertising model sending request; a second acquisition module for acquiring the input feature information of the advertising model based on the category information of the input features of the advertising model; and a second input module for inputting the input feature information of the advertising model into the advertising model to obtain the result predicted by the advertising model based on the input feature information.

[0011] According to the seventh aspect of the embodiments of the present application, an electronic device is provided, comprising: one or more processors; a computer-readable medium configured to store one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the advertising model management method as described in the first aspect, the second aspect or the third aspect of the above-mentioned embodiments.

[0012] According to an eighth aspect of an embodiment of the present application, a computer-readable medium is provided, on which a computer program is stored. When the program is executed by a processor, the advertising model management method as described in the first aspect, the second aspect or the third aspect of the above embodiment is implemented.

[0013] According to the management scheme of the advertising model provided in the embodiment of the present application, the advertising client sends an advertising model sending request to the advertising server, and the advertising server sends the advertising model and the category information of the input features of the advertising model to the advertising client based on the advertising model sending request. The advertising client receives the advertising model and the category information of the input features of the advertising model sent by the advertising server based on the advertising model sending request, and obtains the input feature information of the advertising model based on the category information of the input features of the advertising model, and then inputs the input feature information of the advertising model into the advertising model to obtain the result predicted by the advertising model based on the input feature information. Compared with other existing methods, by sending the advertising model of the advertising server and the category information of the input features of the advertising model to the advertising client, the advertising client can input the input feature information of the advertising model into the advertising model based on the category information of the input features of the advertising model to obtain the result predicted by the advertising model, thereby reducing the interference of the network environment on the advertising model prediction result, thereby effectively improving the success rate of requesting advertising model prediction. In addition, the advertising client obtains the input feature information of the advertising model based on the category information of the input feature of the advertising model, and inputs the input feature information of the advertising model into the advertising model, which can effectively improve the timeliness of the input feature of the advertising model. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Other features, objects and advantages of the present application will become more apparent upon reading the detailed description of non-limiting embodiments made with reference to the following drawings:

[0015] Figure 1 This is a flowchart of the steps of the advertising model management method in Example 1 of the present application;

[0016] Figure 2 This is a flowchart of the steps of the advertising model management method in Example 2 of this application;

[0017] Figure 3A This is a flowchart of the steps of the advertising model management method in Example 3 of this application;

[0018] Figure 3B A schematic diagram of a usage scenario of the advertising model management method provided in accordance with the third embodiment of the present application;

[0019] Figure 4 This is a flowchart of the steps of the advertising model management method in the fourth embodiment of the present application;

[0020] Figure 5 This is a schematic diagram of the structure of the management device of the advertising model in Example 5 of the present application;

[0021] Figure 6 This is a schematic diagram of the structure of the management device of the advertising model in Example 6 of the present application;

[0022] Figure 7 This is a schematic diagram of the structure of the management device of the advertising model in Example 7 of the present application;

[0023] Figure 8 This is a schematic structural diagram of an electronic device in Example 8 of the present application;

[0024] Figure 9 This is the hardware structure of the electronic device in Example 9 of the present application. DETAILED DESCRIPTION

[0025] The present application will be further described in detail below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are merely configured to explain the relevant invention and are not intended to limit the invention. It should also be noted that, for ease of description, only the portions relevant to the relevant invention are shown in the accompanying drawings.

[0026] It should be noted that, in the absence of conflict, the embodiments and features of the embodiments in this application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0027] Reference Figure 1 , showing a step flow chart of the advertising model management method in Example 1 of the present application.

[0028] This embodiment describes the management method of the advertising model provided by this embodiment from the perspective of the advertising client. Specifically, the management method of the advertising model provided by this embodiment includes the following steps:

[0029] In step S101, an advertisement model sending request is sent to an advertisement server, so that the advertisement server sends the advertisement model and category information of input features of the advertisement model to the advertisement client based on the advertisement model sending request.

[0030] In the embodiment of the present application, the advertising service end can be understood as a service end that provides services for advertising. The advertising model sending request can be understood as a request for sending the advertising model and the category information of the input features of the advertising model. The advertising model can be understood as a model for advertising. Specifically, the model for advertising can be a neural network model for advertising or other machine learning models for advertising, such as a decision tree. In a specific usage scenario, the advertising model can be an advertising click-through rate prediction model, an advertising conversion cost prediction model, an advertising click type prediction model, etc. The category information of the input features of the advertising model can be understood as the type of input features of the advertising model, such as a user identifier, an advertising client identifier, an identifier of the terminal device to which the advertising client belongs, the geographical location of the terminal device to which the advertising client belongs, etc. The advertising client can be understood as a client for providing advertising display services. It can be understood that the above description is only exemplary and the embodiment of the present application does not impose any limitation on this.

[0031] In some optional embodiments, when sending an advertising model delivery request to an advertising server, an initialization operation is received for the advertising client; based on the initialization operation, the advertising model delivery request is sent to the advertising server. Thus, the initialization operation for the advertising client can trigger the advertising client to send the advertising model delivery request to the advertising server. It should be understood that the above description is merely exemplary and is not intended to be limiting in any way in this embodiment of the present application.

[0032] In a specific example, the initialization operation on the advertising client may be an operation to start the advertising client, for example, a user clicking an icon of the advertising client to start the advertising client. The initialization operation on the advertising client may also be an operation on an initialization control of the advertising client, for example, a user clicking, dragging, or long pressing an initialization control of the advertising client. It will be understood that the above description is merely exemplary and is not intended to be limiting in any way in the present embodiments.

[0033] In step S102, the advertisement model and category information of the input features of the advertisement model, which are sent by the advertisement server based on the advertisement model sending request, are received.

[0034] In the embodiment of the present application, the advertising client receives the advertising model and the category information of the input features of the advertising model issued by the advertising server based on the advertising model issuance request. It is understood that the above description is only exemplary and the embodiment of the present application does not impose any limitation on this.

[0035] In some optional embodiments, the advertising model includes an advertising dynamic model. When receiving the advertising model and the category information of the input features of the advertising model issued by the advertising server based on the advertising model issuance request, the advertising dynamic model, the category information of the dynamic input features of the advertising dynamic model, and the dynamic coding feature query table of the advertising dynamic model issued by the advertising server based on the advertising model issuance request are received, wherein the dynamic coding feature query table is used to query the dynamic input feature representation information encoding the dynamic input feature information of the advertising dynamic model. Thus, by receiving the advertising dynamic model, the category information of the dynamic input features of the advertising dynamic model, and the dynamic coding feature query table of the advertising dynamic model issued by the advertising server based on the advertising model issuance request, it is possible to request the advertising dynamic model to predict the corresponding result. It will be understood that the above description is only exemplary and the embodiments of the present application do not impose any limitation on this.

[0036] In a specific example, the advertising dynamic model can be understood as a dynamic model obtained by splitting the advertising model. The dynamic model can be understood as a model in which the input feature information of the model changes over time. By sending the dynamic model obtained by splitting the advertising model to the advertising client, the advertising client can have a unique dynamic model. The category information of the dynamic input features of the advertising dynamic model can be understood as the type of dynamic input features of the advertising dynamic model. The dynamic input feature information can be understood as input feature information that changes over time. The dynamic input feature representation information can be understood as a representation vector of the dynamic input feature information. It can be understood that the above description is only exemplary and the embodiments of the present application do not impose any limitation on this.

[0037] In step S103, based on the category information of the input features of the advertisement model, the input feature information of the advertisement model is acquired.

[0038] In some optional embodiments, when obtaining the input feature information of the advertising model based on the category information of the input features of the advertising model, the input feature information of the advertising model is called from the full feature library of the advertising model based on the category information of the input features of the advertising model. Thus, by calling the input feature information of the advertising model from the full feature library of the advertising model, the input feature information of the advertising model can be effectively obtained. It will be understood that the above description is merely exemplary and is not limited in any way by the embodiments of this application.

[0039] In a specific example, before calling the input feature information of the advertising model from the full feature library of the advertising model based on the category information of the input features of the advertising model, the method further includes: obtaining the full feature information of the full feature library of the advertising model based on the full feature configuration information of the full feature library of the advertising model; and constructing the full feature library of the advertising model based on the full feature information of the full feature library of the advertising model. Among them, the full feature configuration information can be understood as the configuration information of the types of full features in the full feature library. Thereby, by obtaining the full feature information of the full feature library of the advertising model through the full feature configuration information of the full feature library of the advertising model, the full feature library of the advertising model can be effectively constructed. It can be understood that the above description is only exemplary and the embodiments of the present application do not impose any limitation on this.

[0040] In step S104, the input feature information of the advertising model is input into the advertising model to obtain a result predicted by the advertising model based on the input feature information.

[0041] In some optional embodiments, when the input feature information of the advertising model is input into the advertising model to obtain the result predicted by the advertising model based on the input feature information, the dynamic coding feature query table is queried based on the dynamic input feature information of the advertising dynamic model to obtain the dynamic input feature representation information corresponding to the dynamic input feature information of the advertising dynamic model; the dynamic input feature representation information of the advertising dynamic model is input into the advertising dynamic model to obtain the result predicted by the advertising dynamic model based on the dynamic input feature representation information. Thus, by converting the dynamic input feature information of the advertising dynamic model into the corresponding dynamic input feature representation information and inputting it into the advertising dynamic model, the advertising dynamic model itself does not need to convert the dynamic input feature information of the advertising dynamic model into the corresponding dynamic input feature representation information, thereby facilitating the advertising dynamic model to effectively obtain the result predicted by the advertising dynamic model based on the dynamic input feature representation information. It will be understood that the above description is only exemplary and the embodiments of the present application do not impose any limitation on this.

[0042] In some optional embodiments, the advertising model also includes an advertising fixed model paired with the advertising dynamic model, and the method further includes: fusing the result predicted by the advertising dynamic model based on the dynamic input feature representation information with the result predicted by the advertising fixed model based on the fixed input feature information, so as to obtain the final prediction result of the advertising dynamic model and the advertising fixed model. Wherein, the advertising model includes an advertising dynamic model and an advertising fixed model paired with the advertising dynamic model, and the fixed input feature information can be understood as input feature information that does not change with time. Thus, by fusing the result predicted by the advertising dynamic model based on the dynamic input feature representation information with the result predicted by the advertising fixed model based on the fixed input feature information, the prediction result of the advertising model can be effectively obtained. It can be understood that the above description is only exemplary, and the embodiments of the present application do not impose any limitation on this.

[0043] In a specific example, the advertising fixed model can be understood as a fixed model obtained by splitting the advertising model. The fixed model can be understood as a model whose input feature information does not change with time. By splitting the advertising model, not only the advertising dynamic model can be obtained, but also the advertising fixed model paired with the advertising dynamic model can be obtained. By splitting the advertising model, the size of the advertising model is reduced, and the advertising model is made lightweight, thereby effectively improving the success rate of the advertising model issuance. It can be understood that the above description is only exemplary, and the embodiments of the present application do not impose any limitations on this.

[0044] Through the advertising model management method provided in the embodiment of the present application, the advertising client sends an advertising model sending request to the advertising server, and the advertising server sends the advertising model and the category information of the input features of the advertising model to the advertising client based on the advertising model sending request. The advertising client receives the advertising model and the category information of the input features of the advertising model sent by the advertising server based on the advertising model sending request, and obtains the input feature information of the advertising model based on the category information of the input features of the advertising model, and then inputs the input feature information of the advertising model into the advertising model to obtain the result predicted by the advertising model based on the input feature information. Compared with other existing methods, by sending the advertising model of the advertising server and the category information of the input features of the advertising model to the advertising client, the advertising client can input the input feature information of the advertising model into the advertising model based on the category information of the input features of the advertising model to obtain the result predicted by the advertising model, thereby reducing the interference of the network environment on the advertising model prediction result, thereby effectively improving the success rate of requesting advertising model prediction. In addition, the advertising client obtains the input feature information of the advertising model based on the category information of the input feature of the advertising model, and inputs the input feature information of the advertising model into the advertising model, which can effectively improve the timeliness of the input feature of the advertising model.

[0045] The advertising model management method provided in this embodiment can be executed by any appropriate device with data processing capabilities, including but not limited to: cameras, terminals, mobile terminals, PCs, servers, vehicle-mounted equipment, entertainment equipment, advertising equipment, personal digital assistants (PDAs), tablet computers, laptop computers, handheld game consoles, smart glasses, smart watches, wearable devices, virtual display devices or display enhancement devices, etc.

[0046] Reference Figure 2 , showing a step flow chart of the advertising model management method in Example 2 of the present application.

[0047] This embodiment describes the management method of the advertising model provided by this embodiment from the perspective of the advertising server. Specifically, the management method of the advertising model provided by this embodiment includes the following steps:

[0048] In step S201, an advertisement model delivery request sent by an advertisement client is received.

[0049] In some optional embodiments, before receiving the advertising model sending request sent by the advertising client, the method further includes: generating the advertising model based on the configured advertising model generation information, and configuring the category information of the input features of the generated advertising model. Wherein, the advertising model generation information includes at least one of the following: the training samples of the advertising model, the training cycle of the advertising model, the update cycle of the advertising model, the output location of the prediction results of the advertising model, the identifier of the advertising model, and the code of the generation logic of the advertising model. Thereby, the advertising model can be effectively generated through the configured advertising model generation information. In addition, the category information of the input features of the generated advertising model can also be configured. It can be understood that the above description is only exemplary and the embodiments of the present application do not impose any limitation on this.

[0050] In a specific example, the advertising model includes an advertising dynamic model and an advertising fixed model paired with the advertising dynamic model. When generating the advertising model based on the configured advertising model generation information and configuring the category information of the input features of the generated advertising model, the advertising dynamic model is generated based on the configured advertising dynamic model generation information and the category information of the input features of the generated advertising dynamic model is configured; and when generating the advertising fixed model based on the configured advertising fixed model generation information and configuring the category information of the input features of the generated advertising fixed model. Among them, the advertising fixed model can be understood as a fixed model obtained by splitting the advertising model, and the advertising dynamic model can be understood as a dynamic model obtained by splitting the advertising model. That is to say, by splitting the advertising model, not only the advertising dynamic model can be obtained, but also the advertising fixed model paired with the advertising dynamic model can be obtained. The advertising dynamic model generation information includes at least one of the following: the training samples of the advertising dynamic model, the training cycle of the advertising dynamic model, the update cycle of the advertising dynamic model, the output location of the prediction results of the advertising dynamic model, the identifier of the advertising dynamic model, and the code of the generation logic of the advertising dynamic model. The advertising fixed model generation information includes at least one of the following: the training samples of the advertising fixed model, the training cycle of the advertising fixed model, the update cycle of the advertising fixed model, the output location of the prediction results of the advertising fixed model, the identifier of the advertising fixed model, and the code of the generation logic of the advertising fixed model. It can be understood that the above description is only exemplary and the embodiments of the present application do not impose any limitation on this.

[0051] In step S202, based on the request issued by the advertising model, the advertising model and the category information of the input features of the advertising model are sent to the advertising client, so that the advertising client obtains the input feature information of the advertising model based on the category information of the input features of the advertising model, and inputs the input feature information of the advertising model into the advertising model to obtain the result predicted by the advertising model based on the input feature information.

[0052] In some optional embodiments, when sending the advertising model and the category information of the input features of the advertising model to the advertising client based on the advertising model sending request, the advertising model sending condition of the advertising client is obtained based on the advertising model sending request; the advertising model whose sending condition matches the advertising model sending condition of the advertising client is determined; and the advertising model whose sending condition matches the advertising model sending condition of the advertising client and the category information of the input features of the advertising model whose sending condition matches the advertising model sending condition of the advertising client are sent to the advertising client. Thus, by obtaining the advertising model sending condition of the advertising client, the advertising model whose sending condition matches the advertising model sending condition of the advertising client and the category information of the input features of the advertising model whose sending condition matches the advertising model sending condition of the advertising client can be sent to the advertising client. It can be understood that the above description is only exemplary and the embodiments of the present application do not impose any limitation on this.

[0053] In a specific example, when obtaining the advertising model sending conditions of the advertising client based on the advertising model sending request, the advertising model sending conditions of the advertising client carried in the advertising model sending request are obtained from the advertising model sending request; or, the identifier of the terminal device to which the advertising client belongs carried in the advertising model sending request is obtained from the advertising model sending request, and based on the identifier of the terminal device to which the advertising client belongs carried in the advertising model sending request, a request for obtaining the advertising model sending conditions is sent to the terminal device to obtain the advertising model sending conditions of the advertising client. It will be understood that the above description is merely exemplary and the embodiments of the present application do not impose any limitation on this.

[0054] In one specific example, before determining an advertising model whose delivery conditions match the advertising model delivery conditions of the advertising client, the method further includes: the advertising server configuring corresponding delivery conditions for all advertising models in the advertising server. It will be understood that the above description is merely exemplary and is not intended to be limiting in any way in the present embodiments.

[0055] In a specific example, when determining an advertising model whose delivery conditions match the delivery conditions of the advertising model of the advertising client, the advertising model whose delivery conditions are the same as the delivery conditions of the advertising model of the advertising client is determined to be the advertising model whose delivery conditions match the delivery conditions of the advertising model of the advertising client. It will be understood that the above description is merely exemplary and is not limited in any way by the present embodiment.

[0056] In a specific example, the conditions for issuing the advertising model of the advertising client include at least one of the following: the version number of the advertising client, the sub-version number of the advertising client, the advertising position of the advertising client, the network environment of the terminal device to which the advertising client belongs, the memory of the terminal device to which the advertising client belongs, the operating system version of the terminal device to which the advertising client belongs, and the geographical location of the terminal device to which the advertising client belongs. Among them, the network environment of the terminal device to which the advertising client belongs may be a 2G network, a 3G network, a 4G network, a 5G network, a local area network, etc. It can be understood that the above description is only exemplary and the embodiments of the present application do not impose any limitations on this.

[0057] In some optional embodiments, the method further includes: if an update of the advertising model is detected, the updated advertising model and the category information of the updated input features of the advertising model are sent to the advertising client, so that the advertising client obtains the updated input feature information of the advertising model based on the category information of the updated input features of the advertising model, and inputs the updated input feature information of the advertising model into the updated advertising model to obtain the result predicted by the updated advertising model based on the updated input feature information. Thus, when an update of the advertising model is detected, the updated advertising model and the category information of the updated input features of the advertising model are sent to the advertising client, and there is no need to update the advertising model through the update of the advertising client, which greatly improves the update efficiency of the advertising model. It can be understood that the above description is only exemplary and the embodiments of the present application do not impose any limitations on this.

[0058] Through the advertising model management method provided in the embodiment of the present application, the advertising server receives the advertising model sending request sent by the advertising client, and based on the advertising model sending request, sends the advertising model and the category information of the input features of the advertising model to the advertising client. The advertising client receives the advertising model and the category information of the input features of the advertising model sent by the advertising server based on the advertising model sending request, and obtains the input feature information of the advertising model based on the category information of the input features of the advertising model, and then inputs the input feature information of the advertising model into the advertising model to obtain the result predicted by the advertising model based on the input feature information. Compared with other existing methods, the advertising model and the category information of the input features of the advertising model are sent to the advertising client through the advertising server, and the advertising client can input the input feature information of the advertising model into the advertising model based on the category information of the input features of the advertising model to obtain the result predicted by the advertising model, thereby reducing the interference of the network environment on the advertising model prediction result, thereby effectively improving the success rate of requesting advertising model prediction. In addition, the advertising client obtains the input feature information of the advertising model based on the category information of the input feature of the advertising model, and inputs the input feature information of the advertising model into the advertising model, which can effectively improve the timeliness of the input feature of the advertising model.

[0059] The advertising model management method provided in this embodiment can be executed by any appropriate device with data processing capabilities, including but not limited to: cameras, terminals, mobile terminals, PCs, servers, vehicle-mounted equipment, entertainment equipment, advertising equipment, personal digital assistants (PDAs), tablet computers, laptop computers, handheld game consoles, smart glasses, smart watches, wearable devices, virtual display devices or display enhancement devices, etc.

[0060] Reference Figure 3A , showing a step flow chart of the advertising model management method in Example 3 of the present application.

[0061] This embodiment describes the advertising model management method provided by this embodiment from the perspective of the interaction between the advertising client and the advertising server. Specifically, the advertising model management method provided by this embodiment includes the following steps:

[0062] In step S301, the advertising client sends an advertising model delivery request to the advertising server.

[0063] Since the specific implementation of step S301 is similar to the specific implementation of step S101 in the above embodiment 1, it will not be repeated here.

[0064] In step S302, the advertising server receives the advertising model sending request sent by the advertising client.

[0065] Since the specific implementation of step S302 is similar to the specific implementation of step S201 in the above-mentioned embodiment 2, it will not be repeated here.

[0066] In step S303, the advertising server sends the advertising model and category information of the input features of the advertising model to the advertising client based on the advertising model sending request.

[0067] Since the specific implementation of step S303 is similar to the specific implementation of step S202 in the above-mentioned embodiment 2, it will not be repeated here.

[0068] In step S304, the advertising client receives the advertising model and category information of the input features of the advertising model, which are sent by the advertising server based on the advertising model sending request.

[0069] Since the specific implementation of step S304 is similar to the specific implementation of step S102 in the above embodiment 1, it will not be repeated here.

[0070] In step S305 , the advertising client obtains input feature information of the advertising model based on the category information of the input feature of the advertising model.

[0071] Since the specific implementation of step S305 is similar to the specific implementation of step S103 in the above embodiment 1, it will not be repeated here.

[0072] In step S306 , the advertising client inputs the input feature information of the advertising model into the advertising model to obtain a result predicted by the advertising model based on the input feature information.

[0073] Since the specific implementation of step S306 is similar to the specific implementation of step S104 in the above-mentioned embodiment 1, it will not be repeated here.

[0074] In a specific example, Figure 3B As shown, a structural diagram of an online advertising system for implementing the advertising model management method provided in an embodiment of the present application is provided. The system may include an advertising server and an advertising client in a terminal device A. It should be understood that Figure 3B The presented advertising server and the advertising client in the terminal device A are merely exemplary and do not limit the implementation forms of the two.

[0075] In actual applications, the advertising server and the terminal device A can be connected by a wired or wireless network. Specifically, the communication connection can be achieved through mobile networks such as GSM, GPRS, LTE, or through Bluetooth, WIFI, infrared, etc. The embodiment of the present application does not limit the specific communication connection method between the advertising server and the terminal device A.

[0076] The advertising server may be a server provided in a service device that provides services to users, and specifically may be an independent application service device. The embodiment of the present application does not limit the structure and implementation form of the advertising server.

[0077] Terminal device A can be a user-facing terminal capable of interacting with the user, such as a mobile phone, laptop, computer, iPad, smart speaker, etc. It can also be various self-service terminals, such as self-service kiosks in hospitals, banks, stations, etc. In addition, terminal device A can also be an intelligent machine that supports interaction, such as a chat robot, a sweeping robot, or a food ordering service robot. The embodiments of this application do not limit the product type and physical form of the terminal device. The embodiments of this application require that it have interactive functions, which can be achieved by installing interactive applications such as advertising clients.

[0078] When issuing an advertising model, the advertising client in terminal device A can send an advertising model issuance request to the advertising server via the network. The advertising server receives the advertising model issuance request sent by the advertising client in terminal device A and, based on the advertising model issuance request, sends the advertising model and category information of the input features of the advertising model to the advertising client. Therefore, the advertising model management method provided in this embodiment of the present application can be executed by the advertising server, and the specific implementation process can refer to the description of the above-mentioned method embodiment 2.

[0079] Through the advertising model management method provided by the embodiment of the present application, the advertising client sends an advertising model sending request to the advertising server, the advertising server receives the advertising model sending request sent by the advertising client, and based on the advertising model sending request, sends the advertising model and the category information of the input features of the advertising model to the advertising client, the advertising client receives the advertising model and the category information of the input features of the advertising model sent by the advertising server based on the advertising model sending request, and based on the category information of the input features of the advertising model, obtains the input feature information of the advertising model, and then inputs the input feature information of the advertising model into the advertising model to obtain the result predicted by the advertising model based on the input feature information. Compared with other existing methods, the advertising server sends the advertising model and the category information of the input features of the advertising model to the advertising client, and the advertising client can input the input feature information of the advertising model into the advertising model based on the category information of the input features of the advertising model to obtain the result predicted by the advertising model, thereby reducing the interference of the network environment on the advertising model prediction result, thereby effectively improving the success rate of requesting advertising model prediction. In addition, the advertising client obtains the input feature information of the advertising model based on the category information of the input feature of the advertising model, and inputs the input feature information of the advertising model into the advertising model, which can effectively improve the timeliness of the input feature of the advertising model.

[0080] The advertising model management method provided in this embodiment can be executed by any appropriate device with data processing capabilities, including but not limited to: cameras, terminals, mobile terminals, PCs, servers, vehicle-mounted equipment, entertainment equipment, advertising equipment, personal digital assistants (PDAs), tablet computers, laptop computers, handheld game consoles, smart glasses, smart watches, wearable devices, virtual display devices or display enhancement devices, etc.

[0081] Reference Figure 4 , showing a step flow chart of the advertising model management method in Example 4 of the present application.

[0082] This embodiment describes the management method of the advertising model provided by this embodiment from the perspective of the advertising client. Specifically, the management method of the advertising model provided by this embodiment includes the following steps:

[0083] In step S401, an advertising model sending request is sent to an advertising Internet of Things device, so that the advertising Internet of Things device sends the advertising model and category information of the input features of the advertising model to the advertising client based on the advertising model sending request.

[0084] In this embodiment, the advertising IoT device can be understood as an IoT device with edge computing capabilities and providing advertising services. The advertising model sending request can be understood as a request for sending the advertising model and the category information of the input features of the advertising model. The advertising model can be understood as a model used for advertising. Specifically, the model used for advertising can be a neural network model for advertising or other machine learning models for advertising, such as a decision tree. In a specific usage scenario, the advertising model can be an advertising click-through rate prediction model, an advertising conversion cost prediction model, an advertising click type prediction model, etc. The category information of the input features of the advertising model can be understood as the type of input features of the advertising model, such as a user identifier, an advertising client identifier, an identifier of the terminal device to which the advertising client belongs, the geographical location of the terminal device to which the advertising client belongs, etc. The advertising client can be understood as a client used to provide advertising display services. It can be understood that the above description is only exemplary and the embodiments of this application do not impose any limitations on this.

[0085] In step S402, the advertising model and category information of the input features of the advertising model sent by the advertising Internet of Things device based on the advertising model sending request are received.

[0086] In this embodiment, the advertising client receives the advertising model and the category information of the input features of the advertising model sent by the advertising IoT device based on the advertising model sending request. It is understood that the above description is only exemplary and the present embodiment does not impose any limitation on this.

[0087] In step S403, the input feature information of the advertisement model is acquired based on the category information of the input feature of the advertisement model.

[0088] Since the specific implementation of step S403 is similar to the specific implementation of step S103 in the above embodiment 1, it will not be repeated here.

[0089] In step S404, the input feature information of the advertising model is input into the advertising model to obtain a result predicted by the advertising model based on the input feature information.

[0090] Since the specific implementation of step S404 is similar to the specific implementation of step S104 in the above-mentioned embodiment 1, it will not be repeated here.

[0091] According to the advertising model management method provided in this embodiment, by sending the advertising model of the advertising Internet of Things device and the category information of the input features of the advertising model to the advertising client, the advertising client can input the input feature information of the advertising model into the advertising model based on the category information of the input features of the advertising model to obtain the results predicted by the advertising model, thereby reducing the interference of the network environment on the prediction results of the advertising model, thereby effectively improving the success rate of requesting advertising model predictions. In addition, the advertising client obtains the input feature information of the advertising model based on the category information of the input features of the advertising model, and inputs the input feature information of the advertising model into the advertising model, which can effectively improve the timeliness of the input features of the advertising model.

[0092] The advertising model management method provided in this embodiment can be executed by any appropriate device with data processing capabilities, including but not limited to: cameras, terminals, mobile terminals, PCs, servers, vehicle-mounted equipment, entertainment equipment, advertising equipment, personal digital assistants (PDAs), tablet computers, laptop computers, handheld game consoles, smart glasses, smart watches, wearable devices, virtual display devices or display enhancement devices, etc.

[0093] Reference Figure 5 , shows a structural diagram of the management device of the advertising model in Example 5 of the present application.

[0094] The management device of the advertising model provided in this embodiment includes: a first sending module 501, used to send an advertising model issuance request to the advertising server, so that the advertising server sends the advertising model and the category information of the input features of the advertising model to the advertising client based on the advertising model issuance request; a first receiving module 502, used to receive the advertising model and the category information of the input features of the advertising model issued by the advertising server based on the advertising model issuance request; a first acquisition module 503, used to obtain the input feature information of the advertising model based on the category information of the input features of the advertising model; a first input module 504, used to input the input feature information of the advertising model into the advertising model to obtain the result predicted by the advertising model based on the input feature information.

[0095] Optionally, the first sending module 501 is specifically configured to: receive an initialization operation for the advertising client; and send the advertising model issuance request to the advertising server based on the initialization operation.

[0096] Optionally, the first acquisition module 503 includes: a calling submodule 5033, which is used to call the input feature information of the advertising model from the full feature library of the advertising model based on the category information of the input feature of the advertising model.

[0097] Optionally, before calling the sub-module 5033, the first acquisition module 503 also includes: an acquisition sub-module 5031, which is used to obtain the full feature information of the full feature library of the advertising model based on the full feature configuration information of the full feature library of the advertising model; and a construction sub-module 5032, which is used to construct the full feature library of the advertising model based on the full feature information of the full feature library of the advertising model.

[0098] Optionally, the advertising model includes an advertising dynamic model, and the first receiving module 502 is specifically used to: receive the advertising dynamic model, the category information of the dynamic input features of the advertising dynamic model, and the dynamic coding feature query table of the advertising dynamic model, which are sent by the advertising server based on the request of the advertising model, wherein the dynamic coding feature query table is used to query the dynamic input feature representation information that encodes the dynamic input feature information of the advertising dynamic model.

[0099] Optionally, the first input module 504 is specifically used to: query the dynamic coding feature query table based on the dynamic input feature information of the advertising dynamic model to obtain the dynamic input feature representation information corresponding to the dynamic input feature information of the advertising dynamic model; input the dynamic input feature representation information of the advertising dynamic model into the advertising dynamic model to obtain the result predicted by the advertising dynamic model based on the dynamic input feature representation information.

[0100] Optionally, the advertising model also includes an advertising fixed model paired with the advertising dynamic model, and the device also includes: a fusion module 505, which is used to fuse the results predicted by the advertising dynamic model based on the dynamic input feature representation information with the results predicted by the advertising fixed model based on the fixed input feature information to obtain the final prediction results of the advertising dynamic model and the advertising fixed model.

[0101] The advertising model management device provided in this embodiment is used to implement the corresponding advertising model management methods in the aforementioned multiple method embodiments, and has the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0102] Reference Figure 6 , showing a structural diagram of the management device of the advertising model in Example 6 of the present application.

[0103] The advertising model management device provided in this embodiment includes: a second receiving module 602, used to receive an advertising model sending request sent by an advertising client; a first sending module 603, used to send the advertising model and the category information of the input features of the advertising model to the advertising client based on the advertising model sending request, so that the advertising client obtains the input feature information of the advertising model based on the category information of the input features of the advertising model, and inputs the input feature information of the advertising model into the advertising model to obtain the result predicted by the advertising model based on the input feature information.

[0104] Optionally, the first sending module 603 is specifically used to: obtain the advertising model sending conditions of the advertising client based on the advertising model sending request; determine the advertising model whose sending conditions match the advertising model sending conditions of the advertising client; and send the category information of the input features of the advertising model whose sending conditions match the advertising model sending conditions of the advertising client and the advertising model whose sending conditions match the advertising model sending conditions of the advertising client to the advertising client.

[0105] Optionally, the conditions for issuing the advertising model of the advertising client include at least one of the following: the version number of the advertising client, the sub-version number of the advertising client, the advertising position of the advertising client, the network environment of the terminal device to which the advertising client belongs, the memory of the terminal device to which the advertising client belongs, the operating system version of the terminal device to which the advertising client belongs, and the geographical location of the terminal device to which the advertising client belongs.

[0106] Optionally, before the second receiving module 602, the apparatus further includes: a generating module 601, configured to generate the advertising model based on the configured advertising model generation information, and configure category information of input features of the generated advertising model.

[0107] Optionally, the advertising model includes an advertising dynamic model and an advertising fixed model paired with the advertising dynamic model, and the generation module 601 is specifically used to: generate the advertising dynamic model based on the configured advertising dynamic model generation information, and configure the category information of the input features of the generated advertising dynamic model; generate the advertising fixed model based on the configured advertising fixed model generation information, and configure the category information of the input features of the generated advertising fixed model.

[0108] Optionally, the advertising model generation information includes at least one of the following: training samples of the advertising model, training cycle of the advertising model, update cycle of the advertising model, output location of the prediction results of the advertising model, identification of the advertising model, and code of the generation logic of the advertising model.

[0109] Optionally, the device also includes: a second sending module 604, which is used to send the updated advertising model and the category information of the updated input features of the advertising model to the advertising client if it is detected that the advertising model is updated, so that the advertising client obtains the updated input feature information of the advertising model based on the category information of the updated input features of the advertising model, and inputs the updated input feature information of the advertising model into the updated advertising model to obtain the result predicted by the updated advertising model based on the updated input feature information.

[0110] The advertising model management device provided in this embodiment is used to implement the corresponding advertising model management methods in the aforementioned multiple method embodiments, and has the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0111] Reference Figure 7 , showing a structural diagram of the management device of the advertising model in Example 7 of the present application.

[0112] The management device of the advertising model provided in this embodiment includes: a second sending module 701, which is used to send an advertising model issuance request to the advertising Internet of Things device, so that the advertising Internet of Things device sends the advertising model and the category information of the input features of the advertising model to the advertising client based on the advertising model issuance request; a third receiving module 702, which is used to receive the advertising model and the category information of the input features of the advertising model issued by the advertising Internet of Things device based on the advertising model issuance request; a second acquisition module 703, which is used to obtain the input feature information of the advertising model based on the category information of the input features of the advertising model; a second input module 704, which is used to input the input feature information of the advertising model into the advertising model to obtain the result predicted by the advertising model based on the input feature information.

[0113] The advertising model management device provided in this embodiment is used to implement the corresponding advertising model management methods in the aforementioned multiple method embodiments, and has the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0114] Figure 8 This is a schematic diagram of the structure of an electronic device in Example 8 of the present application; the electronic device may include:

[0115] One or more processors 801;

[0116] The computer readable medium 802 may be configured to store one or more programs,

[0117] When the one or more programs are executed by the one or more processors, the one or more processors implement the advertising model management method as described in the above-mentioned embodiment 1, embodiment 2, embodiment 3 or embodiment 4.

[0118] Figure 9 The hardware structure of the electronic device in the ninth embodiment of the present application is as follows; Figure 9 As shown, the hardware structure of the electronic device may include: a processor 901, a communication interface 902, a computer-readable medium 903 and a communication bus 904;

[0119] The processor 901, the communication interface 902, and the computer-readable medium 903 communicate with each other via a communication bus 904;

[0120] Optionally, the communication interface 902 may be an interface of a communication module, such as an interface of a GSM module;

[0121] In particular, the processor 901 can be specifically configured to: send an advertising model sending request to the advertising server, so that the advertising server sends the advertising model and the category information of the input features of the advertising model to the advertising client based on the advertising model sending request; receive the advertising model and the category information of the input features of the advertising model sent by the advertising server based on the advertising model sending request; obtain the input feature information of the advertising model based on the category information of the input features of the advertising model; input the input feature information of the advertising model into the advertising model to obtain the result predicted by the advertising model based on the input feature information. In addition, the processor 901 can also be configured to: receive an advertising model sending request sent by the advertising client; send the advertising model and the category information of the input features of the advertising model to the advertising client based on the advertising model sending request, so that the advertising client obtains the input feature information of the advertising model based on the category information of the input features of the advertising model, and inputs the input feature information of the advertising model into the advertising model to obtain the result predicted by the advertising model based on the input feature information. In addition, the processor 901 can also be configured to: send an advertising model sending request to the advertising Internet of Things device, so that the advertising Internet of Things device sends the advertising model and the category information of the input features of the advertising model to the advertising client based on the advertising model sending request; receive the advertising model and the category information of the input features of the advertising model sent by the advertising Internet of Things device based on the advertising model sending request; obtain the input feature information of the advertising model based on the category information of the input features of the advertising model; input the input feature information of the advertising model into the advertising model to obtain the result predicted by the advertising model based on the input feature information.

[0122] The processor 901 can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic device, a discrete gate or transistor logic device, or a discrete hardware component. It can implement or execute the various methods, steps, and logic block diagrams disclosed in the embodiments of the present application. The general-purpose processor can be a microprocessor or any conventional processor.

[0123] The computer-readable medium 903 may be, but is not limited to, a random access memory (RAM), a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), etc.

[0124] In particular, according to an embodiment of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code configured to execute the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part, and / or installed from a removable medium. When the computer program is executed by a central processing unit (CPU), the above-mentioned functions defined in the method of the present application are executed. It should be noted that the computer-readable medium described in the present application can be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable medium can be, for example, but is 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 having one or more conductors, 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 medium, a magnetic storage medium, or any suitable combination thereof. In this application, a computer-readable storage medium may be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In this application, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, which carries computer-readable program code. Such a propagated data signal may take a variety of forms, including, but not limited to, an electromagnetic signal, an optical signal, or any suitable combination thereof. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium that can transmit, propagate, or transfer a program configured for use by or in conjunction with an instruction execution system, apparatus, or device. Program code embodied on a computer-readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0125] Computer program code configured to perform the operations of the present application may be written in one or more programming languages, or a combination thereof, including 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 stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on the remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer through 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., through the Internet using an Internet service provider).

[0126] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of the systems, methods, and computer program products according to various embodiments of the present application. In this regard, each box in the flowchart or block diagram can represent a module, program segment, or part of the code, which contains one or more executable instructions configured to implement the specified logical functions. There are specific sequential relationships in the above-mentioned specific embodiments, but these sequential relationships are only exemplary. When implementing the specific embodiments, these steps may be fewer, more, or the execution order may be adjusted. That is, in some alternative implementations, the functions marked in the boxes may also occur in an order different from that marked in the accompanying drawings. For example, two boxes shown in succession can actually be executed substantially in parallel, or they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, as well as the combination of boxes in the block diagram and / or flowchart, can be implemented using a dedicated hardware-based system that performs the specified functions or operations, or can be implemented using a combination of dedicated hardware and computer instructions.

[0127] The modules involved in the embodiments of the present application may be implemented in software or in hardware. The modules described may also be provided in a processor. For example, they may be described as follows: a processor comprising a first sending module, a first receiving module, a first acquiring module, and a first input module. The names of these modules do not, in some cases, constitute a limitation on the modules themselves. For example, the first sending module may also be described as a module that "sends an advertising model request to the advertising server, so that the advertising server sends the advertising model and the category information of the input features of the advertising model to the advertising client based on the advertising model request."

[0128] As another aspect, the present application further provides a computer-readable medium having a computer program stored thereon, which, when executed by a processor, implements the advertising model management method as described in the above-mentioned embodiment 1, embodiment 2, embodiment 3 or embodiment 4.

[0129] As another aspect, the present application also provides a computer-readable medium, which may be included in the device described in the above embodiment; or it may exist independently and not be assembled into the device. The above computer-readable medium carries one or more programs. When the above one or more programs are executed by the device, the device: sends an advertising model sending request to the advertising server, so that the advertising server sends the advertising model and the category information of the input features of the advertising model to the advertising client based on the advertising model sending request; receives the advertising model and the category information of the input features of the advertising model sent by the advertising server based on the advertising model sending request; obtains the input feature information of the advertising model based on the category information of the input features of the advertising model; inputs the input feature information of the advertising model into the advertising model to obtain the result predicted by the advertising model based on the input feature information. In addition, the device is also configured to: receive an advertising model sending request sent by an advertising client; based on the advertising model sending request, send the advertising model and the category information of the input features of the advertising model to the advertising client, so that the advertising client obtains the input feature information of the advertising model based on the category information of the input features of the advertising model, and inputs the input feature information of the advertising model into the advertising model to obtain the result predicted by the advertising model based on the input feature information. In addition, the device is also configured to: send an advertising model sending request to an advertising Internet of Things device, so that the advertising Internet of Things device sends the advertising model and the category information of the input features of the advertising model to the advertising client based on the advertising model sending request; receive the advertising model and the category information of the input features of the advertising model sent by the advertising Internet of Things device based on the advertising model sending request; based on the category information of the input features of the advertising model, obtain the input feature information of the advertising model; input the input feature information of the advertising model into the advertising model to obtain the result predicted by the advertising model based on the input feature information.

[0130] The terms "first," "second," "the first," or "the second" used in various embodiments of the present disclosure may modify various components regardless of order and / or importance, but these terms do not limit the corresponding components. The above terms are configured solely for the purpose of distinguishing an element from other elements. For example, a first user device and a second user device represent different user devices, even though both are user devices. For example, a first element may be referred to as a second element, and similarly, a second element may be referred to as a first element without departing from the scope of the present disclosure.

[0131] When one element (for example, a first element) is referred to as being “(operably or communicably) coupled” or “(operably or communicably) coupled to” or “connected to” another element (for example, a second element), it should be understood that the one element is directly connected to the other element or that the one element is indirectly connected to the other element via yet another element (for example, a third element). Conversely, it should be understood that when an element (for example, a first element) is referred to as being “directly connected” or “directly coupled” to another element (the second element), there is no element (for example, a third element) interposed therebetween.

[0132] The above description is merely a preferred embodiment of the present application and an illustration of the technical principles employed. Those skilled in the art should understand that the scope of the invention involved in this application is not limited to the technical solutions formed by the specific combination of the above-mentioned technical features, but also encompasses other technical solutions formed by any combination of the above-mentioned technical features or their equivalents without departing from the above-mentioned inventive concept. For example, a technical solution formed by replacing the above-mentioned features with (but not limited to) technical features with similar functions disclosed in this application.

Claims

1. A method for managing an advertising model, the method being applied to an advertising client, the method comprising: Sending an advertisement model sending request to an advertisement server, so that the advertisement server sends the advertisement model and category information of input features of the advertisement model to the advertisement client based on the advertisement model sending request; receiving the advertisement model and category information of the input features of the advertisement model issued by the advertisement server based on the advertisement model issuance request; Based on the category information of the input features of the advertising model, acquiring input feature information of the advertising model; Input feature information of the advertising model is input into the advertising model to obtain a result predicted by the advertising model based on the input feature information.

2. The method according to claim 1, wherein The sending of the advertisement model issuance request to the advertisement server includes: receiving an initialization operation for the advertising client; Based on the initialization operation, the advertisement model issuance request is sent to the advertisement server.

3. The method according to claim 1, wherein The acquiring the input feature information of the advertising model based on the category information of the input feature of the advertising model includes: Based on the category information of the input features of the advertising model, the input feature information of the advertising model is called from the full feature library of the advertising model.

4. The method according to claim 3, wherein: Before calling the input feature information of the advertising model from the full feature library of the advertising model based on the category information of the input feature of the advertising model, the method further includes: Based on the full feature configuration information of the full feature library of the advertising model, obtaining the full feature information of the full feature library of the advertising model; Based on the full feature information of the full feature library of the advertising model, a full feature library of the advertising model is constructed.

5. The method according to claim 1, wherein The advertising model includes an advertising dynamic model, The receiving the advertisement model and the category information of the input features of the advertisement model sent by the advertisement server based on the advertisement model sending request includes: receiving the advertisement dynamic model, category information of the dynamic input features of the advertisement dynamic model, and a dynamic coding feature query table of the advertisement dynamic model, which are issued by the advertisement server based on the advertisement model issuance request; The dynamic coding feature query table is used to query the dynamic input feature representation information encoded for the dynamic input feature information of the advertisement dynamic model.

6. The method according to claim 5, wherein: Inputting the input feature information of the advertising model into the advertising model to obtain a result predicted by the advertising model based on the input feature information includes: Based on the dynamic input feature information of the advertisement dynamic model, querying the dynamic coding feature query table to obtain dynamic input feature representation information corresponding to the dynamic input feature information of the advertisement dynamic model; The dynamic input feature representation information of the advertising dynamic model is input into the advertising dynamic model to obtain a result predicted by the advertising dynamic model based on the dynamic input feature representation information.

7. The method according to claim 6, wherein: The advertising model further includes an advertising fixed model paired with the advertising dynamic model, and the method further includes: The prediction result of the dynamic advertising model based on the dynamic input feature representation information is fused with the prediction result of the fixed advertising model based on the fixed input feature information to obtain the final prediction result of the dynamic advertising model and the fixed advertising model.

8. A method for managing an advertising model, the method being applied to an advertising server, the method comprising: Receive the advertising model delivery request sent by the advertising client; Based on the request issued by the advertising model, the advertising model and the category information of the input features of the advertising model are sent to the advertising client, so that the advertising client obtains the input feature information of the advertising model based on the category information of the input features of the advertising model, and inputs the input feature information of the advertising model into the advertising model to obtain the result predicted by the advertising model based on the input feature information.

9. The method according to claim 8, wherein The sending request based on the advertisement model, sending the advertisement model and category information of the input features of the advertisement model to the advertisement client, includes: Based on the advertisement model issuance request, obtaining the advertisement model issuance condition of the advertisement client; Determining an advertising model whose delivery conditions match the advertising model delivery conditions of the advertising client; The advertisement model whose delivery condition matches the advertisement model delivery condition of the advertisement client and the category information of the input features of the advertisement model whose delivery condition matches the advertisement model delivery condition of the advertisement client are delivered to the advertisement client.

10. The method according to claim 9, wherein: The advertisement model delivery condition of the advertisement client includes at least one of the following: The version number of the advertising client, the sub-version number of the advertising client, the advertising position of the advertising client, the network environment of the terminal device to which the advertising client belongs, the memory of the terminal device to which the advertising client belongs, the operating system version of the terminal device to which the advertising client belongs, and the geographical location of the terminal device to which the advertising client belongs.

11. The method according to claim 8, wherein Before receiving the advertisement model sending request sent by the advertisement client, the method further includes: Based on the configured advertisement model generation information, the advertisement model is generated, and category information of input features of the generated advertisement model is configured.

12. The method according to claim 11, wherein The advertising model includes an advertising dynamic model and an advertising fixed model paired with the advertising dynamic model. The generating information of the configured advertising model, generating the advertising model, and configuring category information of input features of the generated advertising model include: Based on the configured advertising dynamic model generation information, generate the advertising dynamic model, and configure category information of input features of the generated advertising dynamic model; Based on the configured advertisement fixed model generation information, the advertisement fixed model is generated, and category information of input features of the generated advertisement fixed model is configured.

13. The method according to claim 11, wherein The advertisement model generation information includes at least one of the following: The training samples of the advertising model, the training cycle of the advertising model, the update cycle of the advertising model, the output location of the prediction result of the advertising model, the identification of the advertising model, and the code of the generation logic of the advertising model.

14. The method according to claim 8, wherein The method further comprises: If it is detected that the advertising model is updated, the updated advertising model and the category information of the updated input features of the advertising model will be sent to the advertising client, so that the advertising client obtains the updated input feature information of the advertising model based on the category information of the updated input features of the advertising model, and inputs the updated input feature information of the advertising model into the updated advertising model to obtain the prediction result of the updated advertising model based on the updated input feature information.

15. A method for managing an advertising model, the method comprising: Sending an advertising model delivery request to the advertising IoT device, so that the advertising IoT device delivers the advertising model and category information of the input features of the advertising model to the advertising client based on the advertising model delivery request; receiving the advertising model and category information of the input features of the advertising model issued by the advertising Internet of Things device based on the advertising model issuance request; Based on the category information of the input features of the advertising model, acquiring input feature information of the advertising model; Input feature information of the advertising model is input into the advertising model to obtain a result predicted by the advertising model based on the input feature information.

16. A device for managing an advertising model, the device comprising: a first sending module, configured to send an advertisement model sending request to an advertisement server, so that the advertisement server sends the advertisement model and category information of input features of the advertisement model to the advertisement client based on the advertisement model sending request; A first receiving module is configured to receive the advertisement model and category information of the input features of the advertisement model, which are sent by the advertisement server based on the advertisement model sending request; a first acquisition module, configured to acquire input feature information of the advertising model based on category information of the input feature of the advertising model; The first input module is used to input the input feature information of the advertising model into the advertising model to obtain the result predicted by the advertising model based on the input feature information.

17. A device for managing an advertising model, the device comprising: The second receiving module is used to receive the advertisement model sending request sent by the advertisement client; The first sending module is used to send the advertising model and the category information of the input features of the advertising model to the advertising client based on the sending request of the advertising model, so that the advertising client obtains the input feature information of the advertising model based on the category information of the input features of the advertising model, and inputs the input feature information of the advertising model into the advertising model to obtain the result predicted by the advertising model based on the input feature information.

18. A device for managing an advertising model, the device comprising: a second sending module, configured to send an advertising model sending request to the advertising Internet of Things device, so that the advertising Internet of Things device sends the advertising model and category information of the input features of the advertising model to the advertising client based on the advertising model sending request; A third receiving module is configured to receive the advertising model and category information of the input features of the advertising model, which are sent by the advertising IoT device based on the advertising model sending request; a second acquisition module, configured to acquire input feature information of the advertising model based on category information of the input feature of the advertising model; The second input module is used to input the input feature information of the advertising model into the advertising model to obtain the result predicted by the advertising model based on the input feature information.

19. An electronic device, comprising: one or more processors; A computer readable medium configured to store one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors implement the advertising model management method as described in any one of claims 1 to 7, or implement the advertising model management method as described in any one of claims 8 to 14, or implement the advertising model management method as described in claim 15.

20. A computer-readable medium having a computer program stored thereon, which, when executed by a processor, implements the advertising model management method as described in any one of claims 1 to 7, or the advertising model management method as described in any one of claims 8 to 14, or the advertising model management method as described in claim 15.

Images