Data model processing method and device, equipment and storage medium

By processing and grouping the ad serving sample data and generating a parallel training delivery model, the problem that existing advertising delivery methods cannot adapt to user interests is solved, and the accuracy and effectiveness of advertising delivery are improved.

CN118886964AActive Publication Date: 2024-11-01SHENZHEN TAIHAO NETWORK TECHNOLOGY CO LTD
View PDF 8 Cites 0 Cited by

Patent Information

Application Number
CN202410890936.X
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-04
Publication Date
2024-11-01
Estimated Expiration
2044-07-04

AI Technical Summary

Technical Problem

The existing advertising delivery methods cannot adaptively adjust the delivery strategy according to users' interests, resulting in poor advertising delivery results and the inability to accurately attract the attention of target users.

Method used

By obtaining advertising delivery sample data, data processing and grouping, determining the initial delivery model, and conducting parallel training based on product relevance, generating target delivery parameters and models, and optimizing advertising delivery plan.

Benefits of technology

It realizes rapid learning and updating model parameters based on user interests, improves the accuracy and effectiveness of advertising delivery, and solves the problem of poor advertising delivery results.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118886964A_ABST
    Figure CN118886964A_ABST
Patent Text Reader

Abstract

The invention discloses a data model processing method and device, equipment and a storage medium, and relates to the technical field of data processing, and the method comprises the steps: obtaining advertisement putting sample data, carrying out the data processing of the advertisement putting sample data, obtaining training data, determining the number of current GPUs, and determining the grouping number of the training data according to the number of GPUs. Grouping the training data according to the grouping number to obtain training grouped data, determining an initial putting model according to the advertisement putting sample data, determining a product association degree according to the advertisement putting sample data based on the initial putting model, and performing parallel training on the initial putting model according to the product association degree to obtain a target putting parameter, and obtaining a target putting model according to the target putting parameters, carrying out advertisement putting optimization based on the target putting model to obtain an advertisement putting scheme, and carrying out data model processing on the target putting model obtained by carrying out parallel training on the user sample data, rapidly learning the interest of the user and updating the model parameters.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a data model processing method, device, equipment and storage medium. Background Art

[0002] Advertising is a marketing method that can promote products to users and increase product awareness and influence. However, conventional advertising is not targeted at the interests and hobbies of the target audience, resulting in the ads failing to attract the attention of target users and even causing negative emotions among users, thereby increasing marketing costs. In order to solve this problem, the current method is to push ads only based on the usage habits of target users, and the delivery strategy cannot be adjusted adaptively. As a result, the advertising information of the same product will be pushed to target users for a long time, which in turn limits the product content in the ads pushed to target users, reducing the effect of product advertising push. Therefore, the current advertising delivery method has the technical problem of being unable to adapt to the interests of target users and having poor delivery effects.

[0003] The above contents are only used to assist in understanding the technical solution of the present application and do not constitute an admission that the above contents are prior art. Summary of the invention

[0004] The main purpose of this application is to provide a data model processing method, device, equipment and storage medium, aiming to solve the technical problem that the existing technology cannot adaptively change the advertising delivery strategy according to user interests, resulting in reduced advertising delivery effect.

[0005] To achieve the above purpose, the present application proposes a data model processing method, which includes:

[0006] Acquire advertisement delivery sample data, and perform data processing on the advertisement delivery sample data to obtain training data;

[0007] Determine the number of current GPUs, determine the number of groups of the training data according to the number of GPUs, and group the training data according to the number of groups to obtain training group data;

[0008] Determine an initial delivery model according to the advertising delivery sample data, determine product relevance based on the initial delivery model according to the advertising delivery sample data, and perform parallel training on the initial delivery model according to the product relevance to obtain target delivery parameters;

[0009] A target delivery model is obtained according to the target delivery parameters, and advertisement delivery optimization is performed based on the target delivery model to obtain an advertisement delivery plan.

[0010] In one embodiment, the steps of determining an initial delivery model according to the advertising delivery sample data, determining product relevance according to the advertising delivery sample data based on the initial delivery model, and training the initial delivery model in parallel according to the product relevance to obtain target delivery parameters include:

[0011] Determine the product delivery type, target user label, average interest and maximum interest according to the advertisement delivery sample data, and match the initial delivery model according to the product delivery type;

[0012] Determine the target user tag according to the training group data, and determine the product association degree according to the target user tag and the product delivery type;

[0013] The initial delivery model is trained in parallel according to the product association degree, the average interest degree and the maximum interest degree to obtain target delivery parameters.

[0014] In one embodiment, the step of determining the target user tag according to the training group data, and determining the product association degree according to the target user tag and the product delivery type includes:

[0015] Determine the interest circle of the associated user according to the product delivery type and the target user tag;

[0016] Obtaining an associated user tag of the associated user according to the associated user's interest circle;

[0017] The product association degree is obtained based on the associated user tag, the target user tag and the product delivery type.

[0018] In one embodiment, the step of training the initial delivery model in parallel according to the product association, the average interest and the maximum interest to obtain target delivery parameters includes:

[0019] Obtaining the number of the initial delivery models according to the number of groups, and deploying the initial delivery models on GPUs respectively, wherein each GPU corresponds to one initial delivery model;

[0020] Inputting the correlation degree and the training group data into the corresponding initial delivery model respectively to obtain initial model parameters;

[0021] Determine the degree of fit of each of the corresponding initial delivery models according to the initial parameters;

[0022] Sorting the fitness of each of the initial delivery models to obtain the maximum fitness;

[0023] Obtaining a maximum fitting degree threshold value according to the average interest degree and the maximum interest degree;

[0024] When the maximum degree of fit is equal to the maximum degree of fit threshold, the initial parameter corresponding to the maximum degree of fit is determined as the target delivery parameter.

[0025] In one embodiment, the step of sorting the fitness of each initial delivery model to obtain the maximum fitness includes:

[0026] Summarizing the fitness of each initial delivery model and calculating the gradient value of the fitness;

[0027] Determining an error value of the initial delivery model according to the gradient value, and when the error value is greater than a preset error value, updating the initial delivery model corresponding to the gradient value to obtain an updated degree of fit;

[0028] The updated fitness degree and the fitness degrees of each initial delivery model are sorted to obtain the maximum fitness degree.

[0029] In one embodiment, the step of obtaining sample advertising data and performing data processing on the sample advertising data to obtain training data includes:

[0030] Acquire advertisement delivery sample data, and extract characteristic fields from the advertisement delivery sample data;

[0031] The duplicate data in the advertisement delivery sample data is determined according to the characteristic field, the duplicate data is deduplicated, and the deduplicated data is normalized to obtain training data.

[0032] In one embodiment, the step of obtaining a target delivery model according to the target delivery parameters, optimizing advertisement delivery based on the target delivery model, and obtaining an advertisement delivery solution includes:

[0033] Detecting the output data of the target delivery model, and when abnormal data is detected, detecting the abnormal data;

[0034] When the abnormal data is preset abnormal data, data processing is performed on the abnormal data and the advertisement delivery sample data to obtain training data;

[0035] Return to the step of determining the number of current GPUs, determining the number of groups of the training data according to the number of GPUs, and grouping the training data according to the number of groups to obtain training group data.

[0036] In addition, to achieve the above purpose, the present application also proposes a data model processing device, the data model processing device comprising:

[0037] A data preparation module is used to obtain sample data of advertisement delivery and process the sample data of advertisement delivery to obtain training data;

[0038] A data loading module, used to determine the number of current GPUs, determine the number of groups of the training data according to the number of GPUs, and group the training data according to the number of groups to obtain training group data;

[0039] A parameter fitting module is used to determine an initial delivery model according to the advertising delivery sample data, determine product relevance according to the advertising delivery sample data based on the initial delivery model, and perform parallel training on the initial delivery model according to the product relevance to obtain target delivery parameters;

[0040] The model generation module is used to obtain a target delivery model according to the target delivery parameters, and optimize the advertisement delivery based on the target delivery model to obtain an advertisement delivery plan.

[0041] In addition, to achieve the above-mentioned purpose, the present application also proposes a data model processing device, which includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, and the computer program is configured to implement the steps of the data model processing method described above.

[0042] In addition, to achieve the above-mentioned purpose, the present application also proposes a storage medium, which is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by the processor, the steps of the data model processing method described above are implemented.

[0043] In addition, to achieve the above-mentioned purpose, the present application also provides a computer program product, which includes a computer program, and when the computer program is executed by a processor, it implements the steps of the data model processing method described above.

[0044] One or more technical solutions proposed in the present application have at least the following technical effects: obtaining advertising delivery sample data, performing data processing on the advertising delivery sample data to obtain training data, determining the current number of GPUs, determining the number of groups of the training data according to the number of GPUs, grouping the training data according to the number of groups to obtain training group data, determining an initial delivery model according to the advertising delivery sample data, determining product association according to the initial delivery model according to the advertising delivery sample data, performing parallel training on the initial delivery model according to the product association to obtain target delivery parameters, obtaining a target delivery model according to the target delivery parameters, optimizing advertising delivery based on the target delivery model to obtain an advertising delivery plan, and realizing rapid learning of user interests through parallel training of user sample data, updating model parameters, and obtaining a target delivery model for data model processing. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0046] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0047] Figure 1 A flowchart of the first embodiment of the data model processing method of the present application is provided;

[0048] Figure 2 A schematic diagram of initial delivery model training provided in Embodiment 1 of the data model processing method of this application;

[0049] Figure 3 This is a schematic diagram of the module structure of the data model processing device according to an embodiment of the present application;

[0050] Figure 4 A schematic diagram of the device structure of the hardware operating environment involved in the data model processing method in the embodiment of the present application.

[0051] The purpose, features and advantages of this application will be further described in conjunction with the embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0052] It should be understood that the specific embodiments described herein are only used to explain the technical solutions of the present application and are not used to limit the present application.

[0053] In order to better understand the technical solution of the present application, a detailed description will be given below in conjunction with the accompanying drawings and specific implementation methods.

[0054] The main solution of the embodiment of the present application is: obtaining advertising delivery sample data, and performing data processing on the advertising delivery sample data to obtain training data, determining the current number of GPUs, determining the number of groups of the training data according to the number of GPUs, and grouping the training data according to the number of groups to obtain training group data, determining an initial delivery model according to the advertising delivery sample data, and determining a product association based on the initial delivery model according to the advertising delivery sample data, performing parallel training on the initial delivery model according to the product association to obtain target delivery parameters, obtaining a target delivery model according to the target delivery parameters, and optimizing advertising delivery based on the target delivery model to obtain an advertising delivery plan.

[0055] In this embodiment, for ease of description, the following description is made by taking the identification data model processing device as the execution subject.

[0056] Since the existing technology only pushes advertisements based on the usage habits of target users and cannot adapt the delivery strategy, the advertisement information of the same product will be pushed to the target users for a long time, which will limit the product content in the advertisements pushed to the target users and reduce the effect of product advertisement push. Therefore, the current advertising delivery method has the technical problem of being unable to adapt to the interests of target users and having poor delivery effect.

[0057] The present application provides a solution that enables advertisement delivery to quickly learn user interests through parallel training of user sample data, update model parameters, obtain the target delivery model for data model processing, and achieve the purpose of accurate advertisement delivery.

[0058] It can be seen from the above embodiments that the present application obtains advertising delivery sample data, processes the advertising delivery sample data, obtains training data, determines the current number of GPUs, determines the number of groups of training data according to the number of GPUs, groups the training data according to the number of groups, obtains training group data, determines an initial delivery model according to the advertising delivery sample data, determines product association according to the initial delivery model according to the advertising delivery sample data, performs parallel training on the initial delivery model according to the product association, obtains target delivery parameters, obtains a target delivery model according to the target delivery parameters, optimizes advertising delivery based on the target delivery model, obtains an advertising delivery plan, and realizes rapid learning of user interests through parallel training of user sample data, updates model parameters, and obtains a target delivery model for data model processing.

[0059] It should be noted that the execution subject of this embodiment may be a computing service device with data processing, network communication and program running functions, such as a tablet computer, a personal computer, a mobile phone, etc., or an electronic device capable of realizing the above functions, a data model processing device, etc. The following takes the data model processing device as an example to illustrate this embodiment and the following embodiments.

[0060] Based on this, the present application embodiment provides a data model processing method, referring to Figure 1 , Figure 1 This is a flow chart of the first embodiment of the data model processing method of the present application.

[0061] In this embodiment, the data model processing method includes steps S10 to S40:

[0062] Step S10, acquiring advertisement delivery sample data, and performing data processing on the advertisement delivery sample data to obtain training data.

[0063] It should be noted that the advertising sample data refers to the user labels of the target population of the avatar objects, product types, and user interest statistics for the advertising, obtained by sorting out the feedback data of the advertising during the previous advertising process.

[0064] In a specific implementation, when processing the sample data for advertising placement, it is possible to perform data screening, filtering and other operations on the sample data for advertising placement. When processing the sample data for advertising placement, it is possible to obtain the sample data for advertising placement and extract feature fields of the sample data for advertising placement; determine duplicate data in the sample data for advertising placement based on the feature fields, deduplicate the duplicate data, and normalize the deduplicated data to obtain training data. Specifically, by obtaining the sample data of advertisement delivery, the sample data of advertisement delivery is obtained through statistics, saved and transmitted in a database file, and specifically includes the type of delivered product, the role label of the target user, the interest level of the target object in the current product, the average interest level of the user group in the current product, the maximum interest level of the user group in the current product, and other information, by extracting the characteristic fields in the sample data of advertisement delivery, for example, in this embodiment, the user's ID information can be extracted as the characteristic field for data processing, at this time, the advertisement delivery data can be screened for duplicate values ​​according to the user's ID information, and the sample data of advertisement delivery can be sorted and / or traversed with the user's ID information as the primary key value, and the duplicate values ​​in the user ID information are determined, and the sample data of advertisement delivery corresponding to the duplicate ID information is deleted to remove the duplicate values ​​in the data. After the data is deduplicated, in order to facilitate the model to process the data and reduce the size of the data, the deduplicated data can be normalized, and the deduplicated dimensions can be unified to obtain the training data.

[0065] Step S20, determining the number of current GPUs, determining the number of groups of the training data according to the number of GPUs, and grouping the training data according to the number of groups to obtain training group data.

[0066] It should be noted that GPU, which stands for Graphics Processing Unit, is a highly parallel processor that can perform a large number of calculations at the same time. Its data processing speed is more efficient than that of the central processing unit CPU.

[0067] In a specific implementation, the hardware information of the current data model processing device is read to obtain the number of GPUs in the current device, and the number of groups of training data is determined according to the number of GPUs. For example, when the number of GPUs in this embodiment is 4, the current training data can be evenly distributed according to the data volume or data volume entries to obtain 4 groups of training group data with equal or similar data volume entries or data volume sizes. At the same time, the current 4 GPUs can be marked as GPU1, GPU2, GPU3 and GPU4, respectively, and the corresponding 4 grouped training data are marked as Data1, Data2, Data3 and Data4. And Data1 is loaded into GPU1, Data2 is loaded into GPU2, and the grouped training data are loaded into the corresponding GPUs according to a similar loading method. For grouped training data, the original sample data set Data={X1,X2,...,X n The BSP data model of} is defined as a set of M independently generated Bootstrap sample sets, denoted as BSP(Data)={Data1,Data2,...,Datan}. The resulting data set can be considered as a large data set, denoted as an HDFS file directory BSP-HDFS / .

[0068] Step S30, determining an initial delivery model according to the advertising delivery sample data, and determining product relevance based on the initial delivery model according to the advertising delivery sample data, and performing parallel training on the initial delivery model according to the product relevance to obtain target delivery parameters.

[0069] It should be noted that the initial delivery model is a neural network model, which includes a model input layer, a hidden layer, a model output layer, a loss function, and an optimizer. It is used to train the initial delivery model based on the advertising delivery sample data and optimize the parameters in the initial delivery model.

[0070] In the specific implementation, when training the initial delivery model, the initial delivery model can be copied and loaded into each GPU separately, refer to Figure 2 , Figure 2 This is a schematic diagram of the initial delivery model training. Figure 2 In the process, the obtained data set Data is divided according to the number of GPUs, and the group training data is loaded, and the loaded group training data is sent to the input layer of the model. The data is calculated through the hidden layer, and the product association is output from the model output layer. The product association is trained through the loss function and the optimizer to obtain the target delivery parameter. A target delivery parameter will be generated in each GPU.

[0071] Exemplarily, the step of determining an initial delivery model according to the advertising delivery sample data, determining product relevance according to the advertising delivery sample data based on the initial delivery model, and performing parallel training on the initial delivery model according to the product relevance to obtain target delivery parameters includes:

[0072] Determine the product delivery type, target user label, average interest and maximum interest according to the advertisement delivery sample data, and match the initial delivery model according to the product delivery type;

[0073] Determine the target user tag according to the training group data, and determine the product association degree according to the target user tag and the product delivery type;

[0074] The initial delivery model is trained in parallel according to the product association degree, the average interest degree and the maximum interest degree to obtain target delivery parameters.

[0075] In a specific implementation, the sample data of the advertisement delivery is interpreted to determine the product delivery type, target user label, average interest and maximum interest. At this time, it is possible to determine the match between the delivered product type and the target user label, the correlation between the delivered product type and the user label, and the matching degree between the currently delivered product type and the user. The higher the matching degree, the higher the current product correlation degree. Before training, it is necessary to match the initial delivery model according to the delivery type of the product. There are differences between the model parameters of the initial delivery model corresponding to different delivery types. And according to the product correlation, average interest and maximum interest, the average interest is the average of the interest in all sample data, and the maximum interest is the maximum interest in the sample data, representing the maximum tolerance. The obtained product correlation, the average interest and the maximum interest are used to train the initial delivery model in parallel to obtain the target delivery parameters.

[0076] Exemplarily, the step of determining the target user tag according to the training group data, and determining the product association degree according to the target user tag and the product delivery type includes:

[0077] Determine the interest circle of the associated user according to the product delivery type and the target user tag;

[0078] Obtaining an associated user tag of the associated user according to the associated user's interest circle;

[0079] The product association degree is obtained based on the associated user tag, the target user tag and the product delivery type.

[0080] In a specific implementation, after determining the product launch type and the target user tag, the interest circle can be determined according to the product launch type and the target user tag, where the interest circle refers to the tag collection corresponding to the user tag. It can be understood that a target user can have multiple user tags, and correspondingly, a person can have multiple interest circles, where all people in the interest circle have the same identity tag. Assuming that there are currently 3 target users, user A's tag is (photography, travel, food...), user B's tag is (fitness, music, photography...), and user C's tag is (music, photography, e-sports...), then it can be obtained that user A corresponds to the photography interest circle, travel interest circle, food interest circle, and so on. From another perspective, for the photography interest circle, including user A, user B and user C, at this time, determining the interest circle of the associated user according to the product delivery type and the target user tag can be understood as determining the interest circle of the user according to the current user's tag, and determining the identity tags of other users in each interest circle, and determining the interest circle corresponding to the identity tags of other users as the interest circle of the associated user. Still using the above example, for user A, his interest circle is photography, travel, and food, and user B who is in the photography interest circle with user A has fitness and music in addition to photography. At this time, user A can be associated, and it can be determined that user A may also be interested in fitness and music, so it can be determined that user A's associated user tags are fitness and music. At this time, advertising can be delivered based on fitness and music, and the delivery ratio can be determined. The specific implementation method is implemented through the knowledge graph, and then the product association degree is obtained according to the associated user tag, the target user tag and the product delivery type.

[0081] Exemplarily, the step of training the initial delivery model in parallel according to the product association, the average interest and the maximum interest to obtain target delivery parameters includes:

[0082] Obtaining the number of the initial delivery models according to the number of groups, and deploying the initial delivery models on GPUs respectively, wherein each GPU corresponds to one initial delivery model;

[0083] Inputting the correlation degree and the training group data into the corresponding initial delivery model respectively to obtain initial model parameters;

[0084] Determine the degree of fit of each of the corresponding initial delivery models according to the initial parameters;

[0085] Sorting the fitness of each of the initial delivery models to obtain the maximum fitness;

[0086] Obtaining a maximum fitting degree threshold value according to the average interest degree and the maximum interest degree;

[0087] When the maximum degree of fit is equal to the maximum degree of fit threshold, the initial parameter corresponding to the maximum degree of fit is determined as the target delivery parameter.

[0088] In the specific implementation, during training, since the group training data has been loaded into the corresponding GPU, it is also necessary to copy and load the initial delivery model into each GPU according to the number of groups, where each GPU corresponds to an initial delivery model. Input the correlation degree and the training group data into the corresponding initial delivery model to obtain the initial model parameters. Use the ReLU activation function f(x) = max(0,x). Calculate the degree of fit according to the model training formula, the formula is:

[0089]

[0090] Among them, f(t) is the fit, α is the correlation, x i is the sample data, i is the i-th sample data, and f(x) is the activation function value.

[0091] When calculating the goodness of fit, it can be done according to:

[0092]

[0093] Among them, M is the number of categories, y o,c is the one-hot encoding of the true label, p o,c is the predicted value of the model.

[0094] When the degree of fit is greater than 0, it means that the current degree of fit may be too large and the current investment intensity needs to be reduced. At this time, the degree of fit value can be reduced. Correspondingly, when the degree of fit is less than 0, the degree of fit value needs to be increased.

[0095] After obtaining the fitness of all initial delivery models, the fitness values ​​of each model can be sorted to obtain the maximum fitness value. At the same time, the maximum fitness threshold is obtained according to the average interest and the maximum interest. When the maximum fitness is equal to the maximum fitness threshold, the initial parameters corresponding to the maximum fitness are determined as the target delivery parameters.

[0096] Exemplarily, the step of sorting the fitness of each initial delivery model to obtain the maximum fitness includes:

[0097] Summarizing the fitness of each initial delivery model and calculating the gradient value of the fitness;

[0098] Determining an error value of the initial delivery model according to the gradient value, and when the error value is greater than a preset error value, updating the initial delivery model corresponding to the gradient value to obtain an updated degree of fit;

[0099] The updated fitness degree and the fitness degrees of each initial delivery model are sorted to obtain the maximum fitness degree.

[0100] In the specific implementation, the fit of each initial delivery model is summarized and the gradient value is obtained according to the formula:

[0101]

[0102] Among them, L is the loss function, θ is the model parameter, and k is the number of parameters.

[0103] And determine the error value of the initial delivery model based on the gradient value, where the error value is When the error value is greater than the preset error value, the initial delivery model corresponding to the gradient value is updated to obtain the updated fit. When the parameters are updated, Where ω is the learning rate. The updated fitness and the fitness of each initial launch model are sorted to obtain the maximum fitness.

[0104] Step S40, obtaining a target delivery model according to the target delivery parameters, and optimizing the advertisement delivery based on the target delivery model to obtain an advertisement delivery plan.

[0105] In a specific implementation, after obtaining the target delivery model according to the target delivery parameters, the target delivery model can be optimized for advertising delivery, and the corresponding advertising delivery plan can be output. After obtaining the advertising delivery plan, the output data of the target delivery model can be detected. When abnormal data is detected, the abnormal data is detected. When the abnormal data is preset abnormal data, the abnormal data and the advertising delivery sample data are processed to obtain training data, and the number of current GPUs is returned to determine, the number of groups of training data is determined according to the number of GPUs, and the training data is grouped according to the number of groups to obtain the steps of training group data, and the model is retrained.

[0106] The present embodiment provides a data model processing method, which obtains advertising delivery sample data, processes the advertising delivery sample data, obtains training data, determines the current number of GPUs, determines the number of groups of training data according to the number of GPUs, groups the training data according to the number of groups, obtains training group data, determines an initial delivery model according to the advertising delivery sample data, determines product association according to the initial delivery model according to the advertising delivery sample data, performs parallel training on the initial delivery model according to the product association, obtains target delivery parameters, obtains a target delivery model according to the target delivery parameters, optimizes advertising delivery based on the target delivery model, obtains an advertising delivery plan, and realizes rapid learning of user interests through parallel training of user sample data, updates model parameters, and obtains a target delivery model for data model processing.

[0107] This application also provides a data model processing device, please refer to Figure 3 , the data model processing device comprises:

[0108] The data preparation module 10 is used to obtain sample data of advertisement delivery and process the sample data of advertisement delivery to obtain training data;

[0109] A data loading module 20 is used to determine the number of current GPUs, determine the number of groups of the training data according to the number of GPUs, and group the training data according to the number of groups to obtain training group data;

[0110] A parameter fitting module 30 is used to determine an initial delivery model according to the advertising delivery sample data, determine product relevance according to the advertising delivery sample data based on the initial delivery model, and perform parallel training on the initial delivery model according to the product relevance to obtain target delivery parameters;

[0111] The model generation module 40 is used to obtain a target delivery model according to the target delivery parameters, and to optimize the advertisement delivery based on the target delivery model to obtain an advertisement delivery plan.

[0112] The data model processing device provided by the present application adopts the data model processing method in the above embodiment, which can solve the technical problem that the prior art cannot adaptively change the advertising delivery strategy according to user interests, resulting in reduced advertising delivery effect. Compared with the prior art, the beneficial effects of the data model processing device provided by the present application are the same as the beneficial effects of the data model processing method provided by the above embodiment, and other technical features in the data model processing device are the same as the features disclosed in the above embodiment method, which will not be repeated here.

[0113] The present application provides a data model processing device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the data model processing method in the above-mentioned embodiment one.

[0114] Reference below Figure 4 , which shows a schematic diagram of the structure of a data model processing device suitable for implementing the embodiment of the present application. The data model processing device in the embodiment of the present application may include but is not limited to mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions), PMPs (Portable Media Players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 4 The data model processing device shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0115] like Figure 4As shown, the data model processing device may include a processing device 1001 (e.g., a central processing unit, a graphics processor, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM: Read Only Memory) 1002 or a program loaded from a storage device 1003 to a random access memory (RAM: Random Access Memory) 1004. In RAM1004, various programs and data required for the operation of the data model processing device are also stored. The processing device 1001, ROM1002, and RAM1004 are connected to each other through a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Generally, the following systems can be connected to the I / O interface 1006: an input device 1007 including, for example, a touch screen, a touch pad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; an output device 1008 including, for example, a liquid crystal display (LCD: Liquid Crystal Display), a speaker, a vibrator, etc.; a storage device 1003 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 1009. The communication device 1009 can allow the data model processing device to communicate with other devices wirelessly or by wire to exchange data. Although the figure shows a data model processing device with various systems, it should be understood that it is not required to implement or have all the systems shown. More or fewer systems can be implemented or have alternatively.

[0116] In particular, according to the embodiments disclosed in the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present application include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication device, or installed from a storage device 1003, or installed from a ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiment disclosed in the present application are executed.

[0117] The data model processing device provided by the present application adopts the data model processing method in the above embodiment, which can solve the technical problem that the prior art cannot adaptively change the advertising delivery strategy according to user interests, resulting in reduced advertising delivery effect. Compared with the prior art, the beneficial effects of the data model processing device provided by the present application are the same as the beneficial effects of the data model processing method provided by the above embodiment, and the other technical features in the data model processing device are the same as the features disclosed in the method of the previous embodiment, which will not be repeated here.

[0118] It should be understood that the various parts disclosed in this application can be implemented by hardware, software, firmware or a combination thereof. In the description of the above embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0119] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0120] The present application provides a computer-readable storage medium having computer-readable program instructions (ie, computer programs) stored thereon, wherein the computer-readable program instructions are used to execute the data model processing method in the above-mentioned embodiment.

[0121] The computer-readable storage medium provided in the present application may be, for example, a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, systems or devices, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, system or device. The program code contained on the computer-readable storage medium may be transmitted using any appropriate medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination of the above.

[0122] The computer-readable storage medium mentioned above may be included in the data model processing device; or may exist independently without being assembled into the data model processing device.

[0123] The computer-readable storage medium carries one or more programs. When the one or more programs are executed by the data model processing device, the data model processing device:

[0124] Acquire advertisement delivery sample data, and perform data processing on the advertisement delivery sample data to obtain training data;

[0125] Determine the number of current GPUs, determine the number of groups of the training data according to the number of GPUs, and group the training data according to the number of groups to obtain training group data;

[0126] Determine an initial delivery model according to the advertising delivery sample data, determine product relevance based on the initial delivery model according to the advertising delivery sample data, and perform parallel training on the initial delivery model according to the product relevance to obtain target delivery parameters;

[0127] A target delivery model is obtained according to the target delivery parameters, and advertisement delivery optimization is performed based on the target delivery model to obtain an advertisement delivery plan.

[0128] Computer program code for performing the operations of the present application may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of 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., via the Internet using an Internet service provider).

[0129] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present application. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0130] The modules involved in the embodiments described in this application may be implemented by software or hardware, wherein the name of the module does not constitute a limitation on the unit itself in some cases.

[0131] The readable storage medium provided by the present application is a computer-readable storage medium, which stores computer-readable program instructions (i.e., computer programs) for executing the above-mentioned data model processing method, and can solve the technical problem that the prior art cannot adaptively change the advertising delivery strategy according to user interests, resulting in reduced advertising delivery effect. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided by the present application are the same as the beneficial effects of the data model processing method provided by the above-mentioned embodiment, and will not be repeated here.

[0132] The present application also provides a computer program product, including a computer program, which implements the steps of the data model processing method as described above when executed by a processor.

[0133] The computer program product provided by the present application can solve the technical problem that the prior art cannot adaptively change the advertising delivery strategy according to user interests, resulting in reduced advertising delivery effect. Compared with the prior art, the beneficial effects of the computer program product provided by the present application are the same as the beneficial effects of the data model processing method provided by the above embodiment, and will not be repeated here.

[0134] The above descriptions are only some embodiments of the present application, and are not intended to limit the patent scope of the present application. All equivalent structural changes made using the contents of the present application specification and drawings under the technical concept of the present application, or direct / indirect applications in other related technical fields are included in the patent protection scope of the present application.

Claims

1. A data model processing method, characterized in that: The data model processing method comprises: Acquire advertisement delivery sample data, and perform data processing on the advertisement delivery sample data to obtain training data; Determine the number of current GPUs, determine the number of groups of the training data according to the number of GPUs, and group the training data according to the number of groups to obtain training group data; Determine an initial delivery model according to the advertising delivery sample data, determine product relevance based on the initial delivery model according to the advertising delivery sample data, and perform parallel training on the initial delivery model according to the product relevance to obtain target delivery parameters; A target delivery model is obtained according to the target delivery parameters, and advertisement delivery optimization is performed based on the target delivery model to obtain an advertisement delivery plan.

2. The method according to claim 1, characterized in that The step of determining an initial delivery model according to the sample data of advertisement delivery, determining product relevance according to the sample data of advertisement delivery based on the initial delivery model, and performing parallel training on the initial delivery model according to the product relevance to obtain target delivery parameters comprises: Determine the product delivery type, target user label, average interest and maximum interest according to the advertisement delivery sample data, and match the initial delivery model according to the product delivery type; Determine the target user tag according to the training group data, and determine the product association degree according to the target user tag and the product delivery type; The initial delivery model is trained in parallel according to the product association degree, the average interest degree and the maximum interest degree to obtain target delivery parameters.

3. The method according to claim 2, characterized in that The step of determining the target user tag according to the training group data, and determining the product association degree according to the target user tag and the product delivery type comprises: Determine the interest circle of the associated user according to the product delivery type and the target user tag; Obtaining an associated user tag of the associated user according to the associated user's interest circle; The product association degree is obtained based on the associated user tag, the target user tag and the product delivery type.

4. The method according to claim 2, characterized in that The step of training the initial delivery model in parallel according to the product association, the average interest and the maximum interest to obtain target delivery parameters comprises: Obtaining the number of the initial delivery models according to the number of groups, and deploying the initial delivery models on GPUs respectively, wherein each GPU corresponds to one initial delivery model; Inputting the correlation degree and the training group data into the corresponding initial delivery model respectively to obtain initial model parameters; Determine the degree of fit of each of the corresponding initial delivery models according to the initial parameters; Sorting the fitness of each of the initial delivery models to obtain the maximum fitness; Obtaining a maximum fitting degree threshold value according to the average interest degree and the maximum interest degree; When the maximum degree of fit is equal to the maximum degree of fit threshold, the initial parameter corresponding to the maximum degree of fit is determined as the target delivery parameter.

5. The method according to claim 4, characterized in that The step of sorting the fitness of each initial delivery model to obtain the maximum fitness comprises: Summarizing the fitness of each initial delivery model and calculating the gradient value of the fitness; Determining an error value of the initial delivery model according to the gradient value, and when the error value is greater than a preset error value, updating the initial delivery model corresponding to the gradient value to obtain an updated degree of fit; The updated fitness degree and the fitness degrees of each initial delivery model are sorted to obtain the maximum fitness degree.

6. The method according to claim 1, characterized in that The step of obtaining the sample data for advertisement delivery and performing data processing on the sample data for advertisement delivery to obtain training data includes: Acquire advertisement delivery sample data, and extract characteristic fields from the advertisement delivery sample data; The duplicate data in the advertisement delivery sample data is determined according to the characteristic field, the duplicate data is deduplicated, and the deduplicated data is normalized to obtain training data.

7. The method according to any one of claims 1 to 6, characterized in that The step of obtaining a target delivery model according to the target delivery parameters, optimizing advertisement delivery based on the target delivery model, and obtaining an advertisement delivery plan includes: Detecting the output data of the target delivery model, and when abnormal data is detected, detecting the abnormal data; When the abnormal data is preset abnormal data, data processing is performed on the abnormal data and the advertisement delivery sample data to obtain training data; Return to the step of determining the number of current GPUs, determining the number of groups of the training data according to the number of GPUs, and grouping the training data according to the number of groups to obtain training group data.

8. A data model processing device, characterized in that: The device comprises: A data preparation module is used to obtain sample data of advertisement delivery and process the sample data of advertisement delivery to obtain training data; A data loading module, used to determine the number of current GPUs, determine the number of groups of the training data according to the number of GPUs, and group the training data according to the number of groups to obtain training group data; A parameter fitting module, used to determine an initial delivery model according to the advertisement delivery sample data, and to perform parallel training on the initial delivery model according to the training group data to obtain target delivery parameters; The model generation module is used to obtain a target delivery model according to the target delivery parameters, and optimize the advertisement delivery based on the target delivery model to obtain an advertisement delivery plan.

9. A data model processing device, characterized in that: The device comprises: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program is configured to implement the steps of the data model processing method according to any one of claims 1 to 7.

10. A storage medium, characterized in that: The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, the steps of the data model processing method according to any one of claims 1 to 7 are implemented.

Citation Information

Patent Citations

  • Model training method, server and computer readable storage medium

    CN110134636A

  • Model training method and system, nonvolatile storage medium and computer terminal

    CN114463158A

  • Data processing method and device, electronic equipment and storage medium

    CN115292044A

  • Advertisement putting method and advertisement putting system

    CN115907861A

  • Advertisement automatic putting management system based on big data

    CN116362811A