Media information publishing method and device, storage medium and electronic equipment

By obtaining and optimizing the parameter values ​​of multiple media information release methods and generating target publishing methods, the problem of low efficiency of media information release in the existing technology is solved, the click-through rate and conversion rate are improved, and the automatic operation and publishing system is realized. Adjustment.

CN119027183APending Publication Date: 2024-11-26TENCENT TECHNOLOGY (SHENZHEN) CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202310595923.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2023-05-23
Publication Date
2024-11-26

AI Technical Summary

Technical Problem

The publishing method of media information in the prior art leads to a low number of click-through rates, conversion rates and resource transfers, and manual development of publishing methods wastes human resources and is inefficient.

Method used

By obtaining the publishing method of multiple media information, determine the value of each publishing method in terms of click-through rate, conversion rate and resource transfer quantity, calculate the fitness value, generate the target publishing method, and select and optimize according to the fitness value. Release method, dynamically adjust the release strategy to improve results.

Benefits of technology

It improves the click-through rate, conversion rate and resource transfer of media information, optimizes the efficiency of advertising, reduces human resource waste, and realizes automatic adjustment of the automated operation and publishing system.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119027183A_ABST
    Figure CN119027183A_ABST
Patent Text Reader

Abstract

The invention discloses a media information publishing method and device, a storage medium and electronic equipment. The method comprises the following steps: acquiring N publishing modes of media information; determining values of each publishing mode in the N publishing modes on the plurality of target parameters to obtain N parameter value sets; obtaining N fitness values through the N parameter value sets; and generating a target publishing mode through the N fitness values, and publishing the media information according to the target publishing mode. The technical problem that the click rate, the conversion rate and the resource transfer quantity of the media information are low due to a media information publishing mode in the prior art is solved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the computer field, and in particular to a method and device for publishing media information, a storage medium, and an electronic device. Background Art

[0002] In the related field, before publishing an advertisement, it is usually necessary to formulate a publishing method of the advertisement, for example, the publishing time, publishing location, publishing object, etc. of the advertisement. The publishing method of the advertisement is usually formulated by experienced staff.

[0003] The effect of advertising is affected by many environmental factors. Even experienced staff cannot take all environmental factors into consideration when formulating the publishing method. Moreover, the environment will change, and experienced staff cannot predict the changing trend of the environment. This makes the click-through rate, conversion rate and advertising revenue of the published ads unsatisfactory. In addition, manually formulating the publishing method of advertising wastes human resources and is inefficient.

[0004] To address the above-mentioned problems, no effective solution has been proposed yet. Summary of the invention

[0005] The embodiments of the present application provide a method and device for publishing media information, a storage medium and an electronic device, so as to at least solve the technical problems of low click rate, conversion rate and resource transfer quantity of media information caused by the media information publishing method in the prior art.

[0006] According to one aspect of an embodiment of the present application, a method for publishing media information is provided, comprising: obtaining N publishing methods for media information, wherein N is a positive integer greater than 1; determining the values ​​of each publishing method in the N publishing methods on multiple target parameters to obtain N parameter value sets, wherein each of the parameter value sets includes the values ​​of the publishing method corresponding to the parameter value set on the multiple target parameters, and the multiple target parameters include: click-through rate, conversion rate and resource transfer quantity, the click-through rate represents the probability that the media information is clicked when the media information is published, the conversion rate represents the probability that the target event corresponding to the media information is executed when the media information is published, and the resource transfer quantity represents the number of resources triggered to transfer when the media information is published; obtaining N fitness values ​​through the N parameter value sets, wherein each fitness value is used to represent the fitness of the publishing method corresponding to the fitness value on the multiple target parameters; generating a target publishing method through the N fitness values, and publishing the media information according to the target publishing method.

[0007] Optionally, the above-mentioned device is also used to perform a selection operation on the N publishing methods according to the N fitness values, and select M publishing methods from the N publishing methods, wherein M is less than N; when the fitness values ​​of the M publishing methods do not reach a preset fitness threshold, perform a crossover operation and a mutation operation on the M publishing methods to obtain a sub-publishing method set, wherein the sub-publishing method set includes multiple sub-publishing methods; and obtain the target publishing method through the sub-publishing method set.

[0008] Optionally, the above-mentioned device is also used to determine the probability of each of the N publishing methods being selected according to the N fitness values, and obtain N probability values, wherein each of the N probability values ​​is used to represent the probability of the publishing method corresponding to the probability value being selected; perform an accumulation operation on each of the N probability values ​​in turn; when the cumulative sum obtained when performing the accumulation operation on the i-th probability value is greater than or equal to a preset cumulative sum threshold, select the i-th publishing method corresponding to the i-th probability value, wherein i is greater than or equal to 1 and less than or equal to N, and the M publishing methods include the i-th publishing method.

[0009] Optionally, the above-mentioned device is also used to sort the N publishing methods in order of the fitness values ​​from large to small, to obtain sorted N publishing methods; assign rank values ​​to the sorted N publishing methods in sequence, wherein a publishing method that is ranked higher is assigned a larger rank value; and determine the ratio of the rank value of each of the N publishing methods to the sum of the rank values ​​of the N publishing methods as the probability of the publishing method being selected.

[0010] Optionally, the above-mentioned device is also used to randomly select k fitness values ​​from the N fitness values, where k is an integer greater than or equal to 2; and determine the publishing method corresponding to the maximum value of the k fitness values ​​as the publishing method among the M publishing methods.

[0011] Optionally, the above-mentioned device is also used to perform a clustering operation on the sub-publication methods in the sub-publication method set according to the fitness value of each sub-publication method in the sub-publication method set to obtain a target cluster cluster, wherein the fitness value of the sub-publication method in the target cluster cluster is greater than the fitness value of other sub-publication methods in the sub-publication method set, and the other sub-publication methods are sub-publication methods in the sub-publication method set except the sub-publication methods in the target cluster cluster; when there is a sub-publication method in the target cluster cluster whose fitness value is greater than or equal to the preset fitness threshold, the sub-publication method in the target cluster cluster whose fitness value is greater than or equal to the preset fitness threshold is determined as the target release method.

[0012] Optionally, the above-mentioned device is also used to obtain the weight value of each target parameter in the multiple target parameters; perform function value calculation on the multiple weight values ​​corresponding to the multiple target parameters and the N parameter value sets to obtain the N fitness values.

[0013] Optionally, the above-mentioned device is also used to perform the following operations on each parameter value set in the N parameter value sets, and the parameter value set when performing the following operations is called the current parameter value set: multiplying each parameter value in the current parameter value set by the corresponding weight value to obtain multiple product values; and determining the sum of the multiple product values ​​as the fitness value of the current parameter value set.

[0014] Optionally, the above-mentioned device is also used to obtain multidimensional feature data when the media information is released in accordance with the target release mode after the media information is released in accordance with the target release mode, wherein the multidimensional feature data includes the release time of the media information, the release location of the media information, the release object of the media information, and the type of the media information; determine target feature data in the multidimensional feature data, wherein, when the media information is released in accordance with the target feature data, the click-through rate of the media information is greater than or equal to a preset click-through rate threshold, or, the conversion rate of the media information is greater than or equal to a preset conversion rate threshold, or, the resource transfer quantity is greater than or equal to a preset quantity threshold; adjust the target release mode according to the target feature data.

[0015] Optionally, the above-mentioned device is also used to adjust the release time of the media information in the target release mode to the target time period when the target characteristic data indicates that the media information is released in a target time period and the click-through rate of the media information is greater than or equal to the preset click-through rate threshold; or, when the target characteristic data indicates that the media information is released at a target location and the conversion rate of the media information is greater than or equal to the preset conversion rate threshold, adjust the release location of the media information in the target release mode to the target location; or, when the target characteristic data indicates that the target type of media information is released to the target object and the resource transfer quantity of the media information is greater than or equal to the preset quantity threshold, adjust the target release mode to release the target type of media information to the target object.

[0016] According to another aspect of an embodiment of the present application, a media information publishing device is further provided, comprising: an acquisition module, configured to acquire N publishing modes of the media information, wherein N is a positive integer greater than 1; a determination module, configured to determine the values ​​of each of the N publishing modes on multiple target parameters, and obtain N parameter value sets, wherein each of the parameter value sets includes the values ​​of the publishing mode corresponding to the parameter value set on the multiple target parameters, and the multiple target parameters include: a click-through rate, a conversion rate, and a resource transfer quantity, wherein the click-through rate represents the probability that the media information is clicked when the media information is published, the conversion rate represents the probability that the target event corresponding to the media information is executed when the media information is published, and the resource transfer quantity represents the number of resources triggered to be transferred when the media information is published; a processing module, configured to obtain N fitness values ​​through the N parameter value sets, wherein each fitness value is used to represent the fitness of the publishing mode corresponding to the fitness value on the multiple target parameters; and a generation module, configured to generate a target publishing mode through the N fitness values, and publish the media information according to the target publishing mode.

[0017] According to another aspect of the embodiments of the present application, a computer-readable storage medium is provided, in which a computer program is stored, wherein the computer program is configured to execute the above-mentioned method for publishing media information when running.

[0018] According to another aspect of the embodiment of the present application, a computer program product or a computer program is provided, the computer program product or the computer program includes computer instructions, the computer instructions are stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes the above method for publishing media information.

[0019] According to another aspect of the embodiments of the present application, there is further provided an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor is configured to execute the above-mentioned method for publishing media information through the computer program.

[0020] In the embodiment of the present application, the click rate, conversion rate and the number of virtual resource transfers are taken as target parameters, the values ​​of the target parameters of N publishing methods of media information are determined, and the fitness value of each publishing method is obtained through the values ​​of the target parameters of the N publishing methods. The target publishing method is generated through the fitness value, and since the fitness value represents the fitness of the publishing method in terms of click rate, conversion rate and the number of virtual resource transfers, the generated target publishing method has a higher fitness in terms of click rate, conversion rate and the number of virtual resource transfers. Media resources are published according to the target publishing method, achieving the technical effect of improving the click rate, conversion rate and the number of resource transfers of media information, thereby solving the technical problem of low click rate, conversion rate and the number of resource transfers of media information caused by the publishing method of media information in the prior art. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0022] Figure 1 is a schematic diagram of an application environment of an optional method for publishing media information according to an embodiment of the present application;

[0023] Figure 2 It is a flowchart of an optional method for publishing media information according to an embodiment of the present application;

[0024] Figure 3 is a schematic diagram of another optional method for publishing media information according to an embodiment of the present application;

[0025] Figure 4 is a schematic diagram of another optional method for publishing media information according to an embodiment of the present application;

[0026] Figure 5 is a schematic diagram of a single-point crossover according to an embodiment of the present application;

[0027] Figure 6 is a schematic diagram of a multi-point crossover according to an embodiment of the present application;

[0028] Figure 7 It is a process and schematic diagram of the K-means clustering algorithm according to an embodiment of the present application;

[0029] Figure 8 It is a schematic diagram of the overall process according to an embodiment of the present application;

[0030] Fig. 9 is a schematic structural diagram of an optional media information publishing device according to an embodiment of the present application;

[0031] Fig.10 is a block diagram of a computer system structure of an optional electronic device according to an embodiment of the present application;

[0032] Fig.11 It is a schematic diagram of the structure of an optional electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0033] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.

[0034] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0035] First, some nouns or terms that appear in the description of the embodiments of the present application are subject to the following interpretation:

[0036] Multi-objective collaborative optimization: Multi-objective collaborative optimization means that when solving multi-objective optimization problems, it is necessary not only to balance the weights of various objectives, but also to consider the mutual influence and synergy between various objectives.

[0037] Automated operation and publishing system: The automated operation and publishing system is an intelligent advertising publishing system that aims to automatically adjust the advertising publishing method based on data mining and optimization algorithms to improve advertising effectiveness and reduce publishing costs.

[0038] Genetic algorithm: Genetic algorithm is a heuristic search algorithm that simulates the evolutionary process in nature to optimize and solve problems. Genetic algorithm mainly includes steps such as population initialization, selection operation, crossover operation and mutation operation.

[0039] Deep learning model: A deep learning model is a machine learning model based on a multi-layer neural network that can automatically learn feature representations from a large amount of data. Common deep learning models include convolutional neural networks (CNN), recurrent neural networks (RNN), and autoencoders (AE).

[0040] Model ensemble: Model ensemble is a technique to improve model performance by combining multiple models with different structures or parameters to reduce prediction errors and improve generalization capabilities. Common model ensemble methods include bagging, boosting, and stacking.

[0041] According to one aspect of an embodiment of the present invention, a method for publishing media information is provided. Optionally, as an optional implementation, the method for publishing media information can be applied to, but is not limited to, Figure 1 The application environment shown. The application environment includes a user device 102, a network 110 and a server 112, wherein the user device 102 includes a memory 104, a processor 106 and a display 108, wherein the memory is used to store data, including but not limited to storing the publishing method of media information. The processor is used to process the data, including but not limited to determining the values ​​of each publishing method in N publishing methods on multiple target parameters. The display includes but is not limited to being used to display the published media information.

[0042] Optionally, in this embodiment, the user device may be a terminal device configured with a target client, which may include but is not limited to at least one of the following: a mobile phone (such as an Android phone, an iOS phone, etc.), a laptop, a tablet computer, a PDA, a MID (Mobile Internet Devices), a PAD, a desktop computer, a smart TV, etc. The target client may be a video client, an instant messaging client, a browser client, a game client, an audio client, etc., and advertisements may be published to users on the target client.

[0043] Optionally, the network 110 may include but is not limited to: a wired network, a wireless network, wherein the wired network includes: a local area network, a metropolitan area network and a wide area network, and the wireless network includes: Bluetooth, WIFI and other networks that implement wireless communication.

[0044] The server 112 includes a database 114 and a processing engine 116. The database 114 is used to store data, including but not limited to storing media information and media information publishing methods. The processing engine 116 can be used to process data, including but not limited to determining the values ​​of each of N publishing methods on multiple target parameters, and obtaining N fitness values ​​through N parameter value sets.

[0045] The server may be a single server, a server cluster consisting of multiple servers, or a cloud server. The above is only an example and is not limited in this embodiment.

[0046] The above is only an example and is not specifically limited in this embodiment.

[0047] Optionally, as an optional implementation, as Figure 2 As shown, the above-mentioned method for publishing media information includes:

[0048] Step S202, obtaining N publishing modes of the media information, wherein N is a positive integer greater than 1;

[0049] The above media information may be an advertisement, and the publishing method of the advertisement includes but is not limited to the publishing time, publishing location, publishing object, etc. The above N publishing methods may be randomly generated publishing methods or manually pre-defined publishing methods. Each publishing method may be represented by a coding method, such as binary coding, real number coding, etc. Different codes represent different parameters in the publishing method (for example, publishing time, publishing location, publishing channel, etc.).

[0050] In order to optimize the advertising publishing method based on multiple target parameters, it is first necessary to clarify the target parameters that need to be optimized during the advertising publishing process, such as click-through rate, conversion rate, and the number of virtual resource transfers.

[0051] The click-through rate CTR indicates the probability of the media information being clicked when the media information is released. For example, the ratio of the number of times an advertisement is clicked to the number of times the advertisement is released can be determined as the click-through rate of the advertisement, and the click-through rate reflects the ability of the published advertisement to attract users to click.

[0052] The conversion rate CR indicates the probability that the target event corresponding to the media information is executed when the media information is released. The ratio of the number of times the advertisement guides users to complete the predetermined goal (such as purchase, registration, etc.) to the number of times the advertisement is clicked reflects the ability of the advertisement to lead to effective actions. For example, if the media information is a game advertisement, the conversion rate indicates the probability that the target account participates in the game after the game advertisement is released to the target account. For another example, if the media information is an advertisement for selling items, the conversion rate indicates the probability that the target account purchases the item after the advertisement is released to the target account.

[0053] The resource transfer quantity indicates the quantity of resources transferred when the media information is released. The resource transfer quantity may indicate the revenue of the advertising platform after the platform releases the advertisement to the target account, including but not limited to the fees paid by the advertiser to the platform.

[0054] Step S204, determining the values ​​of each of the N publishing methods on multiple target parameters, and obtaining N parameter value sets, wherein each of the parameter value sets includes the values ​​of the publishing method corresponding to the parameter value set on the multiple target parameters, and the multiple target parameters include: click-through rate, conversion rate and resource transfer quantity, the click-through rate represents the probability that the media information is clicked when the media information is published, the conversion rate represents the probability that the target event corresponding to the media information is executed when the media information is published, and the resource transfer quantity represents the quantity of resources triggered to be transferred when the media information is published;

[0055] For example, take the target parameters of click-through rate, conversion rate and resource transfer quantity as an example, where the click-through rate is the first target parameter among multiple target parameters, the conversion rate is the second target parameter, and the resource transfer quantity is the third target parameter. Assume that the above N is 4, and the randomly generated N publishing methods are A, B, C, and D.

[0056] Assume that x is the code corresponding to the publishing method (such as binary code), F i (x) represents the value of the release mode x on the i-th target parameter. For example, F 1 (A) represents the click rate of publishing method A, F 2 (A) represents the conversion rate of publishing method A, F 3 (A) represents the number of virtual resource transfers with publishing mode A. Similarly, we can get the parameter value set T1 = {F 1 (A), F 2 (A), F 3 (A)}、T2={F 1 (B), F 2 (B), F 3 (B)}、T3={F 1 (C), F2 (C), F 3 (C)}、T4={F 1 (D), F 2 (D), F 3 (D)}, where T1 is the parameter value set corresponding to publishing method A, including the click rate, conversion rate and number of virtual resource transfers of publishing method A; T2 is the parameter value set corresponding to publishing method B, including the click rate, conversion rate and number of virtual resource transfers of publishing method B, and so on.

[0057] Step S206, obtaining N fitness values ​​through the N parameter value sets, wherein each fitness value is used to represent the fitness of the publishing method corresponding to the fitness value on the multiple target parameters;

[0058] In order to achieve collaborative optimization among multiple target parameters, the weight value of each target parameter in the multiple target parameters is obtained. The target parameters can be weighted. These weights can be determined according to actual business needs, or they can be determined by fuzzy comprehensive evaluation method, hierarchical analysis method, etc. Weight values ​​can be assigned to each target parameter in advance, such as Figure 3 The flowchart shown may specifically include the following steps:

[0059] Step S301, determining target parameters: clarifying the target parameters that need to be optimized during the advertisement publishing process, such as click-through rate, conversion rate, number of virtual resource transfers, etc.

[0060] Step S302, normalize each target parameter: Since each target parameter may have different dimensions and numerical ranges, direct comparison and summation may lead to distorted calculation results. Therefore, it is necessary to normalize the values ​​of each target parameter and convert them into dimensionless data within the same numerical range.

[0061] Step S303, assigning weights to each target parameter: The weight value of each target parameter can be obtained in the following manner:

[0062] Subjective weighting method: This method assigns weights to each target parameter based on actual needs and experience through expert scoring, analytic hierarchy process (AHP), etc. For example, experts in the advertising field can be invited to score each target parameter, and then the weights are calculated based on the scoring results.

[0063] Objective weighting method: Using the principles of statistics and information theory, weights are assigned to target parameters based on objective indicators such as historical data and correlation analysis. For example, the entropy weight method, principal component analysis (PCA) and other methods can be used to calculate the weights of each target.

[0064] Hybrid weighting method: Consider the advantages of subjective weighting method and objective weighting method, and assign weights to each target parameter. For example, you can first use the subjective weighting method to get the preliminary weight, and then use the objective weighting method to modify the preliminary weight, and finally get the comprehensive weight.

[0065] Step S304, normalize the weights assigned to each target parameter: To ensure that the sum of the weights of all target parameters is 1, the calculated weights need to be normalized. The normalized weight distribution can be expressed as:

[0066] ω=(ω 1 ,ω 2 ,…,ω n )

[0067] where ω i represents the weight of the i-th target parameter, and ∑ω i =1, i=1,2,...,n.

[0068] Through the above steps, appropriate weights can be assigned to each objective parameter in the collaborative optimization model of multi-objective parameters.

[0069] The following formula can be used to calculate the function values ​​of multiple weight values ​​corresponding to multiple target parameters and N parameter value sets to obtain N fitness values:

[0070] f(x)=∑ω i *F i (x)

[0071] Where f(x) represents the fitness value of release mode x, F i (x) represents the value of release mode x on the i-th target parameter.

[0072] For example, the following operations are performed on each parameter value set in the N parameter value sets, and the parameter value set when the following operations are performed is called the current parameter value set: each parameter value in the current parameter value set is multiplied by the corresponding weight value to obtain multiple product values; and the sum of the multiple product values ​​is determined as the fitness value of the current parameter value set.

[0073] The above parameter value set T1 = {F 1 (A), F 2 (A), F 3 (A)} is an example of the current parameter value set, assuming that the weight assigned to the click rate is ω 1 , the weight of conversion rate distribution is ω 2 , the weight of resource transfer quantity allocation is ω 3 , then the fitness value of the release mode A corresponding to the parameter value set T1 is:

[0074] f(x)=ω 1 *F 1 (A)+ω 2 *F 2 (A)+ω 3 *F 3 (A)

[0075] Among them, F 1 (A), F 2 (A), F 3 (A) are the click rate, conversion rate and number of virtual resource transfers of publishing method A respectively.

[0076] Step S208: generating a target publishing mode through the N fitness values, and publishing the media information according to the target publishing mode.

[0077] Genetic algorithms can be used to solve the collaborative optimization of multiple target parameters. Genetic algorithms achieve the transmission and evolution of excellent genes through crossover, mutation and selection operations. In order to improve the search ability of genetic algorithms, machine learning technology can be combined to guide the population. Figure 4 As shown, the following steps may be specifically included:

[0078] Step S401, performing a selection operation on the N publishing modes according to N fitness values, and selecting M publishing modes from the N publishing modes, where M is less than N;

[0079] For the selection operation in the genetic algorithm, the selection operation is performed according to the fitness value of each publishing method, which can ensure that the publishing method with high fitness has a higher probability of being selected and retained.

[0080] Method 1: Determine the probability of each of the N publishing modes being selected according to the N fitness values ​​to obtain N probability values, wherein each of the N probability values ​​is used to represent the probability of the publishing mode corresponding to the probability value being selected; perform accumulation operation on each of the N probability values ​​in turn;

[0081] The probability of each publishing method being selected is calculated based on the fitness value of each publishing method. This can be achieved in the following ways:

[0082] P(j)=f(j) / Σf(.)

[0083] Among them, P(j) is the probability value of the j-th release method being selected, f(j) is the fitness value of the j-th release method, and Σf(.) is the sum of the fitness values ​​of N release methods.

[0084] The probability of each publishing method being selected is calculated based on the fitness value of each publishing method. It can also be achieved by sorting N publishing methods according to the fitness value, and then assigning the selection probability according to the sorting result. The specific operation is as follows:

[0085] (1) Sort the N publishing methods according to the size of their fitness values, either from large to small or from small to large.

[0086] (2) A rank value is assigned to each publishing method. The publishing method with a larger fitness value has a larger rank value. For example, the rank value of the individual with the highest fitness is 2, the rank value of the individual with the second highest fitness is 1, and so on.

[0087] (3) Calculate the probability of each publishing method being selected based on the rank value: P(i) = Rank(i) / ΣRank(.), where Rank(i) is the rank value of the i-th publishing method, and ΣRank(.) is the sum of the rank values ​​of all publishing methods.

[0088] For example, the fitness values ​​are sorted in descending order to obtain the sorted N publishing methods; rank values ​​are assigned to the sorted N publishing methods in sequence, wherein a publishing method that is ranked higher is assigned a larger rank value; and the ratio of the rank value of each of the N publishing methods to the total rank values ​​of the N publishing methods is determined as the probability of the publishing method being selected.

[0089] Generate a random number r (0<=r<=1), accumulate the selected probability value of each release method, and when the cumulative sum is greater than or equal to r, use the release method corresponding to the currently added selected probability value as the selected release method. For example, when the cumulative sum obtained when performing the accumulation operation on the i-th probability value is greater than or equal to the preset cumulative sum threshold, select the i-th release method corresponding to the i-th probability value, where i is greater than or equal to 1 and less than or equal to N, and the M release methods include the i-th release method. Repeat this process until a sufficient number of individuals are selected.

[0090] Method 2: Randomly select a certain number of publishing methods from N publishing methods, and select the publishing method with the highest fitness value from the selected publishing methods. The specific operations are as follows:

[0091] The specific operation is as follows: randomly select k fitness values ​​from the N fitness values, where k is an integer greater than or equal to 2; determine the publishing method corresponding to the maximum value of the k fitness values ​​as the publishing method in the M publishing methods. Repeat the above process until a sufficient number of publishing methods are selected.

[0092] Step S402: When the fitness values ​​of the selected M publishing methods do not reach a preset fitness threshold, a crossover operation and a mutation operation are performed on the M publishing methods to obtain a sub-publishing method set, wherein the sub-publishing method set includes multiple sub-publishing methods;

[0093] The selected M publishing methods are cross-operated to generate new publishing methods (sub-publishing methods). The cross-operation can be performed using single-point cross-operation, multi-point cross-operation, uniform cross-operation, etc. The cross-operation helps to generate new combinations of advertising publishing methods and promotes the diversity and innovation of publishing methods.

[0094] For example, the following operations are performed for any two of the M publishing modes. The two publishing modes when performing the following operations are called a parent publishing mode and a parent publishing mode, which include:

[0095] Determine S intersection points in the parent publishing mode and the parent publishing mode, where S is an integer greater than or equal to 1;

[0096] A cross operation is performed on the parent publishing mode and the parent publishing mode at the S intersection points to obtain a first child publishing mode and a second child publishing mode, wherein the child publishing mode set includes the first child publishing mode and the second child publishing mode.

[0097] When S=1, it is a single-point crossover. Figure 5 The figure shows a schematic diagram of a single-point crossover, where a and b are the parent release mode and the parent release mode (the parent release mode and the parent release mode are represented by a code string). An intersection point R is randomly selected from the parent release mode a and the parent release mode b, and the parent release mode a and the parent release mode b are divided into two parts. The first child release mode ( Figure 5 The publishing method P1 in the second sub-publishing method ( Figure 5 Release method P2).

[0098] When S=2, it is a multi-point crossover. Figure 6 The figure shows a schematic diagram of multi-point crossover, where a and b are the parent release mode and the parent release mode (the parent release mode and the parent release mode are represented by a code string). Two crossover points R1 and R2 are randomly selected in the parent release mode a and the parent release mode b, and the parent release mode a and the parent release mode b are divided into three parts. The first child release mode ( Figure 6 The publishing method P1 in the second sub-publishing method ( Figure 6 Release method P2).

[0099] The new release mode obtained after the crossover operation is mutated to increase the diversity of the population. The mutation operation can be performed using single-point mutation, multi-point mutation, uniform mutation and other methods. The mutation operation helps to prevent the genetic algorithm from falling into the local optimal solution too early and improves the global search capability.

[0100] Step S403, obtaining the target publishing method through the sub-publishing method set.

[0101] The set of sub-release methods obtained by the genetic algorithm can be guided by machine learning technology. For example, cluster analysis is used to divide the sub-release methods in the sub-release method set, separating high-fitness individuals from low-fitness sub-release methods. Then, the high-fitness sub-release methods are further optimized to guide the evolutionary direction of the low-fitness release methods. This step helps to improve the search efficiency of the genetic algorithm and the quality of the solution.

[0102] For example, according to the fitness value of each sub-publishing method in the sub-publishing method set, a clustering operation is performed on the sub-publishing methods in the sub-publishing method set to obtain a target cluster, wherein the fitness value of the sub-publishing method in the target cluster is greater than the fitness value of other sub-publishing methods in the sub-publishing method set, and the other sub-publishing methods are sub-publishing methods in the sub-publishing method set except the sub-publishing methods in the target cluster; when there is a sub-publishing method in the target cluster whose fitness value is greater than or equal to the preset fitness threshold, the sub-publishing method in the target cluster whose fitness value is greater than or equal to the preset fitness threshold is determined as the target publishing method. The above fitness threshold can be set according to actual conditions.

[0103] The clustering algorithm can generally adopt the K-means algorithm, and its calculation formula is as follows:

[0104] WCSS=Σ(Σ||x i -c j || 2 )

[0105] Among them, x i is the i-th publishing method among M publishing methods, c j is the centroid of the jth cluster, ||x i -c j || 2 is the i-th release method x i and its cluster centroid c j The square of the Euclidean distance between them. Figure 7 Shown is the process and schematic diagram of the K-means clustering algorithm.

[0106] The termination conditions of the selection operation, mutation operation and clustering operation in the above embodiment may be the number of iterations, the fitness value reaching the above preset fitness threshold, etc. If the termination condition is met, the target publishing mode is output; if not, the selection operation, mutation operation and clustering operation are continued.

[0107] Since the market environment will change in real time, fixed advertising publishing methods cannot adapt to market changes. After publishing media information according to the target publishing method, the publishing method can be dynamically adjusted according to real-time data to achieve automated optimization of operational publishing.

[0108] For example, after the media information is published according to the target publishing method, multidimensional feature data when the media information is published according to the target publishing method is obtained, wherein the multidimensional feature data includes the publishing time of the media information, the publishing location of the media information, the publishing object of the media information, and the type of the media information; target feature data is determined in the multidimensional feature data, wherein, when the media information is published according to the target feature data, the click-through rate of the media information is greater than or equal to a preset click-through rate threshold, or the conversion rate of the media information is greater than or equal to a preset conversion rate threshold, or the resource transfer quantity is greater than or equal to a preset quantity threshold; the target publishing method is adjusted according to the target feature data.

[0109] The target feature data may be determined in the multi-dimensional feature data by a deep machine learning model, which may specifically include the following steps:

[0110] Step S1, after publishing the advertisement according to the target publishing method, collect multi-dimensional feature data when the advertisement is published, including but not limited to: the publishing location, publishing time, publishing object of the advertisement, and advertisement type, etc.;

[0111] Step S2, preprocessing the collected multidimensional feature data, including but not limited to data cleaning, missing value processing, outlier processing, etc.

[0112] Step S3, performing feature engineering on the collected multi-dimensional feature data, including but not limited to feature selection, feature dimensionality reduction, feature transformation, etc. The purpose of feature engineering is to extract key information and reduce the calculation dimension;

[0113] Step S4: input the multidimensional feature data obtained in step S3 into a deep learning model, and extract target feature data from the multidimensional feature data through the deep learning model.

[0114] The deep learning model can be a model structure such as a multi-layer perceptron (MLP), a convolutional neural network (CNN), or a long short-term memory network (LSTM). The number of layers, number of neurons, activation function, and connection method of the deep learning model can be determined according to actual conditions.

[0115] The deep learning model is obtained through multiple iterations of training using the training data set. The model parameters can be updated using optimization algorithms such as gradient descent and stochastic gradient descent. The model is evaluated and verified using methods such as cross-validation and leave-one-out to ensure that the model has good generalization capabilities.

[0116] To further improve model performance, you can use ensemble learning methods (such as bagging, boosting, stacking, etc.) to combine multiple deep learning models with different structures or parameters. Model ensemble helps improve the generalization ability of the model and reduce overfitting:

[0117] Bagging (Bootstrap Aggregating) generates multiple training subsets by sampling the original data set with replacement (bootstrap sampling), and then trains multiple base learners separately. Finally, the prediction results of the base learners are voted or averaged to obtain the final prediction. The formula of bagging is as follows:

[0118]

[0119] Among them, Y(x) is the final prediction result, N is the number of base learners, and y_i(x) is the prediction result of the i-th base learner.

[0120] Y(x)=sign(∑α i *h i (x)

[0121] Among them, Y(x) is the final prediction result, α i is the weight of the ith base learner, h i (x) is the prediction result of the i-th base learner, and sign is the sign function.

[0122] Stacking: Stacking combines the prediction results of multiple base learners as input features by training a meta-learner (such as logistic regression, support vector machine, etc.). The formula for stacking is as follows:

[0123] Y(x)=f(Σw i *y i (x)

[0124] Among them, Y(x) is the final prediction result, f is the meta-learner, and w iis the weight of the ith base learner, y i (x) is the prediction result of the i-th base learner.

[0125] Collect data in the process of advertising release in real time, including key indicators such as click-through rate, cost, conversion rate, etc. Based on real-time data, build release method adjustment indicators, such as response rate, adjustment range, etc. Release method adjustment indicators are used to measure the urgency and range of method adjustment. Dynamically adjust the release method based on the release method adjustment indicators. For example, when the response rate is fast and the adjustment range is large, increase the release budget; otherwise, reduce the release budget. Evaluate the release effect after the method adjustment, such as calculating the adjusted click-through rate, cost, conversion rate and other key indicators. Based on the evaluation results, continuously iterate and optimize the release method to better adapt to market changes and user needs.

[0126] For example, when the target characteristic data indicates that the media information is to be published in a target time period, and the click-through rate of the media information is greater than or equal to the preset click-through rate threshold, the publishing time of the media information in the target publishing method is adjusted to the target time period; or, when the target characteristic data indicates that the media information is to be published at a target location, and the conversion rate of the media information is greater than or equal to the preset conversion rate threshold, the publishing location of the media information in the target publishing method is adjusted to the target location; or, when the target characteristic data indicates that media information of a target type is to be published to a target object, and the quantity of resource transfers of the media information is greater than or equal to the preset quantity threshold, the target publishing method is adjusted to publish media information of the target type to the target object.

[0127] like Figure 8 The overall flow chart shown may include the following steps:

[0128] Step S801, determining a plurality of target parameters, and assigning a weight to each of the plurality of target parameters;

[0129] Step S802, obtaining N publishing methods, which may be randomly generated or formulated by staff;

[0130] Step S803, calculating the fitness value of each of the N publishing modes;

[0131] Step S804, selecting M publishing methods from N publishing methods according to the fitness values ​​of the various publishing methods, where M is a positive integer less than N;

[0132] Step S805, performing a crossover operation and a mutation operation on the M publishing methods to obtain a sub-publishing method set, wherein the sub-publishing method set includes multiple sub-publishing methods obtained after the crossover and mutation operations;

[0133] Step S806: clustering the sub-publishing mode set, selecting a publishing mode with a higher fitness value from the sub-publishing mode set, and obtaining a target cluster.

[0134] Step S807, determining whether there is a publishing mode in the target cluster whose fitness value is greater than or equal to a preset fitness threshold, if so, executing step S808, otherwise executing step S809;

[0135] Step S808, after performing a crossover mutation operation on the release modes in the target cluster, a clustering operation is performed again until a release mode having a fitness value greater than or equal to a preset fitness threshold exists in the obtained cluster;

[0136] Step S809, determining a publishing mode with a fitness value greater than or equal to a preset fitness threshold as a target publishing mode, and publishing the advertisement according to the target publishing mode;

[0137] Step S810, performing preprocessing, feature engineering and other operations on the multi-dimensional feature data of the advertisement release to extract key information;

[0138] Step S811, using a deep learning model to mine and analyze multi-dimensional feature data to provide a basis for optimizing advertising publishing strategies.

[0139] Step S812: During the advertisement publishing process, the publishing strategy is dynamically adjusted according to real-time data to achieve automated optimization of operation publishing.

[0140] Step S813, evaluate and verify the optimized publishing system to ensure continuous optimization of the system.

[0141] For example, suppose an advertising company wants to optimize its automated operation publishing system. The company has multiple advertising goals, such as increasing click-through rate, reducing costs, and increasing conversion rate. At the same time, advertising publishing involves multiple features, such as user age, gender, geographic location, and interests.

[0142] By assigning weights to multiple objectives, a multi-objective collaborative optimization model is constructed. In this example, the weights of click-through rate, cost, and conversion rate are set to 0.4, 0.3, and 0.3, respectively. The model is optimized using genetic algorithms combined with machine learning techniques. In each iteration, new publishing methods are generated through crossover, mutation, and selection operations. At the same time, multi-dimensional feature data of advertising releases are mined and analyzed through deep learning and other technologies to extract key information and improve the publishing effect. The publishing strategy is dynamically adjusted according to real-time data to achieve automated optimization of operational releases. During the advertising release process, the system adjusts the advertising strategy in real time according to market changes and user feedback. Finally, the optimized publishing system is evaluated to verify its performance in improving click-through rate, reducing costs, and increasing conversion rate.

[0143] In this way, the automated operation publishing system can achieve collaborative optimization under multiple objectives and multiple features, improve the advertising publishing effect, and reduce the publishing cost. The technical solution of the present invention has a wide range of application prospects and can provide effective solutions for the advertising industry, e-commerce platforms and other fields.

[0144] It is understandable that in the specific implementation of this application, related data such as user information is involved. When the above embodiments of this application are applied to specific products or technologies, user permission or consent is required, and the collection, use and processing of relevant data need to comply with relevant laws, regulations and standards of relevant countries and regions.

[0145] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present application.

[0146] According to another aspect of the embodiments of the present application, a media information publishing device for implementing the above-mentioned media information publishing method is also provided. Fig. 9As shown, the device includes: an acquisition module 92, which is used to acquire N publishing methods of media information, wherein N is a positive integer greater than 1; a determination module 94, which is used to determine the values ​​of each publishing method in the N publishing methods on multiple target parameters, and obtain N parameter value sets, wherein each of the parameter value sets includes the values ​​of the publishing method corresponding to the parameter value set on the multiple target parameters, and the multiple target parameters include: click-through rate, conversion rate and resource transfer quantity, wherein the click-through rate represents the probability that the media information is clicked when the media information is published, the conversion rate represents the probability that the target event corresponding to the media information is executed when the media information is published, and the resource transfer quantity represents the number of resources triggered to transfer when the media information is published; a processing module 96, which is used to obtain N fitness values ​​through the N parameter value sets, wherein each fitness value is used to represent the fitness of the publishing method corresponding to the fitness value on the multiple target parameters; a generation module 98, which is used to generate a target publishing method through the N fitness values, and publish the media information according to the target publishing method.

[0147] Optionally, the above-mentioned device is also used to perform a selection operation on the N publishing methods according to the N fitness values, and select M publishing methods from the N publishing methods, wherein M is less than N; when the fitness values ​​of the M publishing methods do not reach a preset fitness threshold, perform a crossover operation and a mutation operation on the M publishing methods to obtain a sub-publishing method set, wherein the sub-publishing method set includes multiple sub-publishing methods; and obtain the target publishing method through the sub-publishing method set.

[0148] Optionally, the above-mentioned device is also used to determine the probability of each of the N publishing methods being selected according to the N fitness values, and obtain N probability values, wherein each of the N probability values ​​is used to represent the probability of the publishing method corresponding to the probability value being selected; perform an accumulation operation on each of the N probability values ​​in turn; when the cumulative sum obtained when performing the accumulation operation on the i-th probability value is greater than or equal to a preset cumulative sum threshold, select the i-th publishing method corresponding to the i-th probability value, wherein i is greater than or equal to 1 and less than or equal to N, and the M publishing methods include the i-th publishing method.

[0149] Optionally, the above-mentioned device is also used to sort the N publishing methods in order of the fitness values ​​from large to small, to obtain sorted N publishing methods; assign rank values ​​to the sorted N publishing methods in sequence, wherein a publishing method that is ranked higher is assigned a larger rank value; and determine the ratio of the rank value of each of the N publishing methods to the sum of the rank values ​​of the N publishing methods as the probability of the publishing method being selected.

[0150] Optionally, the above-mentioned device is also used to randomly select k fitness values ​​from the N fitness values, where k is an integer greater than or equal to 2; and determine the publishing method corresponding to the maximum value of the k fitness values ​​as the publishing method among the M publishing methods.

[0151] Optionally, the above-mentioned device is also used to perform a clustering operation on the sub-publication methods in the sub-publication method set according to the fitness value of each sub-publication method in the sub-publication method set to obtain a target cluster cluster, wherein the fitness value of the sub-publication method in the target cluster cluster is greater than the fitness value of other sub-publication methods in the sub-publication method set, and the other sub-publication methods are sub-publication methods in the sub-publication method set except the sub-publication methods in the target cluster cluster; when there is a sub-publication method in the target cluster cluster whose fitness value is greater than or equal to the preset fitness threshold, the sub-publication method in the target cluster cluster whose fitness value is greater than or equal to the preset fitness threshold is determined as the target release method.

[0152] Optionally, the above-mentioned device is also used to obtain the weight value of each target parameter in the multiple target parameters; perform function value calculation on the multiple weight values ​​corresponding to the multiple target parameters and the N parameter value sets to obtain the N fitness values.

[0153] Optionally, the above-mentioned device is also used to perform the following operations on each parameter value set in the N parameter value sets, and the parameter value set when performing the following operations is called the current parameter value set: multiplying each parameter value in the current parameter value set by the corresponding weight value to obtain multiple product values; and determining the sum of the multiple product values ​​as the fitness value of the current parameter value set.

[0154] Optionally, the above-mentioned device is also used to obtain multidimensional feature data when the media information is released in accordance with the target release mode after the media information is released in accordance with the target release mode, wherein the multidimensional feature data includes the release time of the media information, the release location of the media information, the release object of the media information, and the type of the media information; determine target feature data in the multidimensional feature data, wherein, when the media information is released in accordance with the target feature data, the click-through rate of the media information is greater than or equal to a preset click-through rate threshold, or, the conversion rate of the media information is greater than or equal to a preset conversion rate threshold, or, the resource transfer quantity is greater than or equal to a preset quantity threshold; adjust the target release mode according to the target feature data.

[0155] Optionally, the above-mentioned device is also used to adjust the release time of the media information in the target release mode to the target time period when the target characteristic data indicates that the media information is released in a target time period and the click-through rate of the media information is greater than or equal to the preset click-through rate threshold; or, when the target characteristic data indicates that the media information is released at a target location and the conversion rate of the media information is greater than or equal to the preset conversion rate threshold, adjust the release location of the media information in the target release mode to the target location; or, when the target characteristic data indicates that the target type of media information is released to the target object and the resource transfer quantity of the media information is greater than or equal to the preset quantity threshold, adjust the target release mode to release the target type of media information to the target object.

[0156] According to one aspect of the present application, a computer program product is provided, the computer program product comprising a computer program / instruction, the computer program / instruction comprising 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 the communication part 1009, and / or installed from a removable medium 1011. When the computer program is executed by the central processor 1001, various functions provided in the embodiments of the present application are executed.

[0157] The serial numbers of the above-mentioned embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0158] Fig.10 The structure block diagram of a computer system for implementing an electronic device according to an embodiment of the present application is schematically shown.

[0159] It should be noted that Fig.10 The computer system 1000 of the electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0160] like Fig.10 As shown, the computer system 1000 includes a central processing unit 1001 (CPU), which can perform various appropriate actions and processes according to the program stored in the read-only memory 1002 (ROM) or the program loaded from the storage part 1008 to the random access memory 1003 (RAM). Various programs and data required for system operation are also stored in the random access memory 1003. The central processing unit 1001, the read-only memory 1002 and the random access memory 1003 are connected to each other through a bus 1004. The input / output interface 1005 (Input / Output interface, i.e., I / O interface) is also connected to the bus 1004.

[0161] The following components are connected to the input / output interface 1005: an input section 1006 including a keyboard, a mouse, etc.; an output section 1007 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 1008 including a hard disk, etc.; and a communication section 1009 including a network interface card such as a LAN card, a modem, etc. The communication section 1009 performs communication processing via a network such as the Internet. A drive 1010 is also connected to the input / output interface 1005 as needed. A removable medium 1011, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 1010 as needed, so that a computer program read therefrom is installed into the storage section 1008 as needed.

[0162] In particular, according to an embodiment of the present application, the process described in each method flow chart can be implemented as a computer software program. For example, an embodiment of the present application includes a computer program product, which includes a computer program carried on a computer readable medium, and the computer program contains a program code for executing the method shown in the flow chart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 1009, and / or installed from the removable medium 1011. When the computer program is executed by the central processor 1001, various functions defined in the system of the present application are executed.

[0163] According to another aspect of the embodiment of the present application, an electronic device for implementing the above-mentioned method for publishing media information is also provided. The electronic device may be Figure 1 The terminal device or server shown in the figure. This embodiment is described by taking the electronic device as a terminal device as an example. Fig.11As shown, the electronic device includes a memory 1102 and a processor 1104. The memory 1102 stores a computer program, and the processor 1104 is configured to execute the steps in any of the above method embodiments through the computer program.

[0164] Optionally, in this embodiment, the electronic device may be located in at least one network device among a plurality of network devices of a computer network.

[0165] Optionally, in this embodiment, the processor may be configured to perform the following steps through a computer program:

[0166] S1, obtaining N publishing modes of media information, where N is a positive integer greater than 1;

[0167] S2, determining the value of each of the N publishing methods on multiple target parameters, and obtaining N parameter value sets, wherein each of the parameter value sets includes the values ​​of the publishing method corresponding to the parameter value set on the multiple target parameters, and the multiple target parameters include: click-through rate, conversion rate and resource transfer quantity, the click-through rate represents the probability that the media information is clicked when the media information is published, the conversion rate represents the probability that the target event corresponding to the media information is executed when the media information is published, and the resource transfer quantity represents the quantity of resources triggered to be transferred when the media information is published;

[0168] S3, obtaining N fitness values ​​through the N parameter value sets, wherein each fitness value is used to represent the fitness of the publishing method corresponding to the fitness value on the multiple target parameters;

[0169] S4, generating a target publishing mode according to the N fitness values, and publishing the media information according to the target publishing mode.

[0170] Alternatively, a person skilled in the art may understand that: Fig.11 The structure shown is for illustration only, and the electronic device may also be a smart phone (such as an Android phone, an iOS phone, etc.), a tablet computer, a PDA, a mobile Internet device (MID), a PAD, and other terminal devices. Fig.11 The electronic device and the electronic equipment described above are not limited in structure. Fig.11 More or fewer components (such as network interfaces, etc.) as shown in, or with Fig.11 Different configurations shown.

[0171] Among them, the memory 1102 can be used to store software programs and modules, such as the program instructions / modules corresponding to the media information publishing method and device in the embodiments of the present application. The processor 1104 executes various functional applications and data processing by running the software programs and modules stored in the memory 1102, that is, to implement the above-mentioned media information publishing method. The memory 1102 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 1102 may further include a memory remotely located relative to the processor 1104, and these remote memories may be connected to the terminal via a network. Examples of the above-mentioned networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof. Among them, the memory 1102 can be specifically, but not limited to, used to store media information, as well as information such as the method of publishing media information. As an example, such as Fig.11 As shown, the memory 1102 may include but is not limited to the acquisition module 92, determination module 94, processing module 96 and generation module 98 in the media information publishing device. In addition, it may also include but is not limited to other module units in the media information publishing device, which will not be repeated in this example.

[0172] Optionally, the transmission device 1106 is used to receive or send data via a network. Specific examples of the network may include a wired network and a wireless network. In one example, the transmission device 1106 includes a network adapter (Network Interface Controller, NIC), which can be connected to other network devices and routers via a network cable so as to communicate with the Internet or a local area network. In one example, the transmission device 1106 is a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0173] In addition, the electronic device further includes: a display 1108 for displaying the published media information; and a connection bus 1110 for connecting various module components in the electronic device.

[0174] In other embodiments, the terminal device or server may be a node in a distributed system, wherein the distributed system may be a blockchain system, and the blockchain system may be a distributed system formed by connecting the multiple nodes through network communication. Among them, the nodes may form a peer-to-peer (P2P, Peer To Peer) network, and any form of computing device, such as a server, terminal and other electronic devices, may become a node in the blockchain system by joining the peer-to-peer network.

[0175] According to one aspect of the present application, a computer-readable storage medium is provided, and a processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes the media information publishing method provided in the above-mentioned various optional implementations.

[0176] Optionally, in this embodiment, the computer-readable storage medium may be configured to store a computer program for performing the following steps:

[0177] S1, obtaining N publishing modes of media information, where N is a positive integer greater than 1;

[0178] S2, determining the value of each of the N publishing methods on multiple target parameters, and obtaining N parameter value sets, wherein each of the parameter value sets includes the values ​​of the publishing method corresponding to the parameter value set on the multiple target parameters, and the multiple target parameters include: click-through rate, conversion rate and resource transfer quantity, the click-through rate represents the probability that the media information is clicked when the media information is published, the conversion rate represents the probability that the target event corresponding to the media information is executed when the media information is published, and the resource transfer quantity represents the quantity of resources triggered to be transferred when the media information is published;

[0179] S3, obtaining N fitness values ​​through the N parameter value sets, wherein each fitness value is used to represent the fitness of the publishing method corresponding to the fitness value on the multiple target parameters;

[0180] S4, generating a target publishing mode according to the N fitness values, and publishing the media information according to the target publishing mode.

[0181] Optionally, in this embodiment, a person of ordinary skill in the art may understand that all or part of the steps in the various methods of the above embodiments may be completed by instructing hardware related to the terminal device through a program, and the program may be stored in a computer-readable storage medium, and the storage medium may include: a flash drive, a read-only memory (ROM), a random access memory (RAM), a disk or an optical disk, etc.

[0182] The serial numbers of the above-mentioned embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0183] If the integrated units in the above embodiments are implemented in the form of software functional units and sold or used as independent products, they can be stored in the above computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for enabling one or more computer devices (which may be personal computers, servers, or network devices, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present application.

[0184] In the above embodiments of the present application, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, please refer to the relevant description of other embodiments.

[0185] In the several embodiments provided in the present application, it should be understood that the disclosed client can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0186] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0187] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0188] The above is only a preferred implementation of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A method for publishing media information, characterized in that: include: Obtain N publishing methods of media information, where N is a positive integer greater than 1; Determine the value of each of the N publishing methods on multiple target parameters, and obtain N parameter value sets, wherein each of the parameter value sets includes the values ​​of the publishing method corresponding to the parameter value set on the multiple target parameters, and the multiple target parameters include: click-through rate, conversion rate and resource transfer quantity, the click-through rate represents the probability that the media information is clicked when the media information is published, the conversion rate represents the probability that the target event corresponding to the media information is executed when the media information is published, and the resource transfer quantity represents the quantity of resources triggered to be transferred when the media information is published; Obtaining N fitness values ​​through the N parameter value sets, wherein each fitness value is used to represent the fitness of the publishing method corresponding to the fitness value on the multiple target parameters; A target publishing mode is generated by using the N fitness values, and the media information is published according to the target publishing mode.

2. The method according to claim 1, characterized in that The method of generating a target publishing mode by taking the N fitness values ​​includes: Performing a selection operation on the N publishing modes according to the N fitness values, and selecting M publishing modes from the N publishing modes, where M is less than N; When the fitness values ​​of the M publishing methods do not reach a preset fitness threshold, performing a crossover operation and a mutation operation on the M publishing methods to obtain a sub-publishing method set, wherein the sub-publishing method set includes multiple sub-publishing methods; The target publishing method is obtained through the sub-publishing method set.

3. The method according to claim 2, characterized in that The performing a selection operation on the N publishing modes according to the N fitness values ​​to select M publishing modes from the N publishing modes includes: Determine the probability of each of the N publishing modes being selected according to the N fitness values, and obtain N probability values, wherein each of the N probability values ​​is used to represent the probability of the publishing mode corresponding to the probability value being selected; Performing an accumulation operation on each of the N probability values ​​in sequence; When the cumulative sum obtained when performing the accumulation operation on the i-th probability value is greater than or equal to a preset cumulative sum threshold, the i-th publication method corresponding to the i-th probability value is selected, where i is greater than or equal to 1 and less than or equal to N, and the M publication methods include the i-th publication method.

4. The method according to claim 3, characterized in that The determining the probability of each of the N publishing modes being selected according to the N fitness values ​​to obtain N probability values ​​also includes: Sorting the N publishing methods according to the fitness values ​​from large to small to obtain sorted N publishing methods; Assigning rank values ​​to the sorted N publishing methods in sequence, wherein a publishing method with a higher ranking is assigned a larger rank value; The ratio of the rank value of each of the N publishing modes to the total rank values ​​of the N publishing modes is determined as the probability of the publishing mode being selected.

5. The method according to claim 2, characterized in that: The performing a selection operation on the N publishing modes according to the N fitness values ​​to select M publishing modes from the N publishing modes also includes: Randomly select k fitness values ​​from the N fitness values, where k is an integer greater than or equal to 2; The publishing mode corresponding to the maximum value among the k fitness values ​​is determined as the publishing mode among the M publishing modes.

6. The method according to claim 2, characterized in that The obtaining the target publishing mode through the sub-publishing mode set includes: According to the fitness value of each sub-publishing method in the sub-publishing method set, a clustering operation is performed on the sub-publishing methods in the sub-publishing method set to obtain a target cluster, wherein the fitness value of the sub-publishing method in the target cluster is greater than the fitness value of other sub-publishing methods in the sub-publishing method set, and the other sub-publishing methods are sub-publishing methods in the sub-publishing method set except the sub-publishing methods in the target cluster; In the case that there is a sub-release mode with a fitness value greater than or equal to the preset fitness threshold in the target cluster, the sub-release mode with a fitness value greater than or equal to the preset fitness threshold in the target cluster is determined as the target release mode.

7. The method according to any one of claims 1 to 6, characterized in that The obtaining of N fitness values ​​through the N parameter value sets includes: Obtaining a weight value of each of the target parameters; Function value calculation is performed on multiple weight values ​​corresponding to the multiple target parameters and the N parameter value sets to obtain the N fitness values.

8. The method according to claim 7, characterized in that The performing function value calculation on the multiple weight values ​​corresponding to the multiple target parameters and the N parameter value sets to obtain the N fitness values ​​includes: The following operations are performed on each parameter value set in the N parameter value sets. The parameter value set when performing the following operations is called the current parameter value set: Multiply each parameter value in the current parameter value set by the corresponding weight value to obtain multiple product values; The sum of the multiple product values ​​is determined as the fitness value corresponding to the current parameter value set.

9. The method according to claim 1, characterized in that: After publishing the media information in the target publishing manner, the method further includes: Acquire multidimensional feature data when the media information is released in the target release mode, wherein the multidimensional feature data includes the release time of the media information, the release location of the media information, the release object of the media information, and the type of the media information; Determining target feature data in the multidimensional feature data, wherein when the media information is published according to the target feature data, the click rate of the media information is greater than or equal to a preset click rate threshold, or the conversion rate of the media information is greater than or equal to a preset conversion rate threshold, or the resource transfer quantity is greater than or equal to a preset quantity threshold; The target publishing method is adjusted according to the target characteristic data.

10. The method according to claim 9, characterized in that Adjusting the target publishing method according to the target characteristic data includes: When the target characteristic data indicates that the media information is to be released in a target time period and the click rate of the media information is greater than or equal to the preset click rate threshold, adjusting the release time of the media information in the target release mode to the target time period; or When the target characteristic data indicates that the media information is to be published at the target location, and the conversion rate of the media information is greater than or equal to the preset conversion rate threshold, adjusting the publishing location of the media information in the target publishing mode to the target location; or When the target characteristic data indicates publishing the target type of media information to the target object, and the resource transfer quantity of the media information is greater than or equal to the preset quantity threshold, the target publishing mode is adjusted to publishing the target type of media information to the target object.

11. A device for publishing media information, characterized in that: include: An acquisition module, used to acquire N publishing modes of media information, where N is a positive integer greater than 1; A determination module, used to determine the value of each of the N publishing methods on multiple target parameters, and obtain N parameter value sets, wherein each of the parameter value sets includes the values ​​of the publishing method corresponding to the parameter value set on the multiple target parameters, and the multiple target parameters include: click-through rate, conversion rate and resource transfer quantity, the click-through rate represents the probability that the media information is clicked when the media information is published, the conversion rate represents the probability that the target event corresponding to the media information is executed when the media information is published, and the resource transfer quantity represents the quantity of resources triggered to be transferred when the media information is published; A processing module, used to obtain N fitness values ​​through the N parameter value sets, wherein each fitness value is used to represent the fitness of the publishing method corresponding to the fitness value on the multiple target parameters; A generating module is used to generate a target publishing mode according to the N fitness values, and publish the media information according to the target publishing mode.

12. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored program, wherein the program can be executed by a terminal device or a computer to execute the method described in any one of claims 1 to 10.

13. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instructions are executed by a processor, the steps of the method described in any one of claims 1 to 10 are implemented.

14. An electronic device comprising a memory and a processor, characterized in that: A computer program is stored in the memory, and the processor is configured to execute the method according to any one of claims 1 to 10 through the computer program.