CN115689632A

CN115689632A - Resource distribution method and device, computer equipment and storage medium

Info

Publication number: CN115689632A
Application number: CN202211354324.6A
Authority: CN
Inventors: 孙佳鑫
Original assignee: Douyin Vision Co Ltd
Current assignee: Douyin Vision Co Ltd
Priority date: 2022-11-01
Filing date: 2022-11-01
Publication date: 2023-02-03

Abstract

The present disclosure provides a resource distribution method, apparatus, computer device and storage medium, wherein the method comprises: determining the use conversion rate when different types of virtual resources are distributed to each user to be distributed by using the trained target network model; dividing each user to be distributed into different user groups according to the use conversion rate corresponding to each user to be distributed; determining the virtual resource type to be distributed corresponding to each user group based on the use conversion rate of the user to be distributed in each virtual resource type in each user group, the resource quantity threshold of the virtual resource and the user ratio of using each target service corresponding to the reference group; the users in the reference group refer to users which are not allocated with virtual resources; and distributing the virtual resources to each user to be distributed in the user group according to the type of the virtual resources to be distributed corresponding to each user group.

Description

Resource distribution method and device, computer equipment and storage medium

Technical Field

The present disclosure relates to the field of computer technologies, and in particular, to a resource distribution method and apparatus, a computer device, and a storage medium.

Background

In order to improve the utilization rate of various target services, virtual resources of different virtual resource types are generally distributed to a user, so that the willingness of the user to use the target services is improved by using the distributed virtual resources, and the utilization rate of the service resources is further improved.

However, different users have different usage habits, and the usage requirements of virtual resources of various virtual resource types are different, so how to specifically issue virtual resources for each user becomes a problem to be solved urgently.

Disclosure of Invention

The embodiment of the disclosure at least provides a resource distribution method, a resource distribution device, computer equipment and a storage medium.

In a first aspect, an embodiment of the present disclosure provides a resource distribution method, including:

determining the use conversion rate when distributing different types of virtual resources to each user to be distributed by using the trained target network model; the usage conversion rate is used for indicating the probability of using the target service after the user to be distributed is distributed with the virtual resource;

dividing each user to be distributed into different user groups according to the use conversion rate corresponding to each user to be distributed;

determining the virtual resource type to be distributed corresponding to each user group based on the use conversion rate of the user to be distributed in each virtual resource type in each user group, the resource quantity threshold of the virtual resource and the user ratio of using each target service corresponding to the reference group; the users in the reference group refer to users which are not allocated with virtual resources;

and distributing the virtual resources to each user to be distributed in the user group according to the type of the virtual resources to be distributed corresponding to each user group.

In a possible implementation manner, the dividing, according to the usage conversion rate respectively corresponding to each user to be distributed, each user to be distributed into different user groups includes:

dividing each user to be distributed into different user groups according to the use conversion rate respectively corresponding to each user to be distributed under any virtual resource type; or the like, or, alternatively,

determining a conversion rate average value corresponding to each user to be distributed according to the use conversion rate respectively corresponding to each user to be distributed under each virtual resource type, and dividing each user to be distributed into different user groups by using the conversion rate average value respectively corresponding to each user to be distributed.

In one possible embodiment, the trained target network model is determined according to the following steps:

collecting sample data according to a preset time period, and performing iterative training on the gain network model to be trained by using the collected sample data to obtain a trained gain network model;

determining a model evaluation index of the trained gain network model, and taking the trained gain network model as a trained target network model under the condition that the model evaluation index is larger than the model evaluation index of the verification gain model;

the verification gain model is used for verifying whether the trained gain network model meets the use requirement, and is obtained by training the sample data and has a different network structure from the gain network model.

In a possible implementation manner, the determining, based on the usage conversion rate of the user to be distributed in each virtual resource type in each user group, a resource amount threshold of a virtual resource, and a user ratio of using each target service corresponding to a reference group, a virtual resource type to be distributed corresponding to each user group includes:

for each user group, determining that after different types of virtual resources are distributed to the users to be distributed in the user group, the users to be distributed use the first multiplexing information of the target service corresponding to different types of virtual resources again;

determining the virtual resource distribution gain of different virtual resource types corresponding to the user group according to the first multiplexing information of the user group;

determining conversion rate increment of each user group under each virtual resource type according to the use conversion rate of the user to be distributed in each virtual resource type in each user group and the user ratio corresponding to a reference group;

and determining the type of the virtual resource to be distributed corresponding to each user group according to the maximum virtual resource distribution gain principle, the virtual resource distribution gain, the conversion rate increment and the resource quantity threshold of the virtual resource.

In a possible implementation manner, the determining that, after the virtual resources of different types are distributed to the user to be distributed in the user group, the user to be distributed uses the first multiplexing information of the target service corresponding to the different virtual resource types again includes:

after different types of virtual resources are distributed to the users to be distributed in the user group, the users to be distributed reuse the first multiplexing information of the target service corresponding to different types of virtual resources in different preset time periods;

the determining, according to the first multiplexing information of the user group, a virtual resource distribution gain of the user group corresponding to different virtual resource types includes:

and aiming at each virtual resource type, merging the first multiplexing information of the target service of the user to be distributed in the user group, which uses the virtual resource type again in different preset time periods, so as to obtain the virtual resource distribution gain of the user group under the virtual resource type.

In a possible implementation manner, the determining, according to the usage conversion rate of the user to be distributed in each virtual resource type in each user group and the user ratio corresponding to a reference group, a conversion rate increment of each user group in each virtual resource type includes:

for each user group, determining the average use conversion rate of the user group under each virtual resource type according to the use conversion rate of each user to be distributed in the user group under each virtual resource type;

and determining the average use conversion rate of each user group under each virtual resource type and the difference value of the user rate corresponding to the reference group as the conversion rate increment of each user group under each virtual resource type.

In a possible implementation manner, the determining, according to the principle that the virtual resource distribution gain is the largest, the type of the virtual resource to be distributed corresponding to each user group according to the virtual resource distribution gain, the conversion rate increment, and the resource amount threshold of the virtual resource includes:

determining the virtual resource type to be verified corresponding to each user group according to the maximum virtual resource distribution gain principle, the virtual resource distribution gain, the conversion rate increment and the resource quantity threshold of the virtual resource;

determining second multiplexing information of target services corresponding to different virtual resource types used by users in the reference group;

determining reference resource distribution gains of different virtual resource types corresponding to the reference group according to the second multiplexing information;

and for each virtual resource type to be verified, taking the virtual resource type to be verified as the virtual resource type to be distributed of the corresponding user group under the condition that the virtual resource distribution gain of the user group corresponding to the virtual resource type to be verified is larger than the reference resource distribution gain corresponding to the reference group under the virtual resource type to be verified.

In a second aspect, an embodiment of the present disclosure further provides a resource distribution apparatus, including:

the first determining module is used for determining the use conversion rate when different types of virtual resources are distributed to each user to be distributed by utilizing the trained target network model; the usage conversion rate is used for indicating the probability of using the target service after the user to be distributed is distributed with the virtual resource;

the dividing module is used for dividing each user to be distributed into different user groups according to the use conversion rate respectively corresponding to each user to be distributed;

a second determining module, configured to determine, based on the usage conversion rate of the user to be distributed in each virtual resource type in each user group, a resource amount threshold of a virtual resource, and a user ratio of using each target service corresponding to a reference group, a virtual resource type to be distributed corresponding to each user group; the users in the reference group refer to users which are not allocated with virtual resources;

and the distribution module is used for distributing the virtual resources to each user to be distributed in the user group according to the type of the virtual resources to be distributed corresponding to each user group.

In a possible implementation manner, the dividing module, when dividing each user to be distributed into different user groups according to the usage conversion rate respectively corresponding to each user to be distributed, is configured to:

In one possible embodiment, the apparatus further comprises:

the training module is used for determining a trained target network model according to the following steps:

In a possible implementation manner, the second determining module, when determining the virtual resource type to be distributed corresponding to each user group based on the usage conversion rate of the user to be distributed in each virtual resource type in each user group, a resource amount threshold of a virtual resource, and a user ratio of using each target service corresponding to a reference group, is configured to:

determining conversion rate increment of each user group under each virtual resource type according to the use conversion rate of the user to be distributed in each user group under each virtual resource type and the user ratio corresponding to a reference group;

In a possible implementation manner, the second determining module, when determining that, after the virtual resources of different types are distributed to the user to be distributed in the user group, the user to be distributed uses the first multiplexing information of the target service corresponding to the different virtual resource types again, is configured to:

In a possible implementation manner, the second determining module, when determining, according to the conversion rate of the user to be distributed in each virtual resource type in each user group and the user ratio corresponding to a reference group, a conversion rate increment of each user group in each virtual resource type, is configured to:

In a possible implementation manner, the second determining module, when determining the type of the virtual resource to be distributed corresponding to each user group according to the virtual resource distribution gain, the conversion rate increment, and the resource amount threshold of the virtual resource based on the principle that the virtual resource distribution gain is maximum, is configured to:

determining second multiplexing information of target services corresponding to different virtual resource types used by the users in the reference group;

In a third aspect, this disclosure also provides a computer device, a processor, and a memory, where the memory stores machine-readable instructions executable by the processor, and the processor is configured to execute the machine-readable instructions stored in the memory, and when the machine-readable instructions are executed by the processor, the machine-readable instructions are executed by the processor to perform the steps in the first aspect or any one of the possible implementations of the first aspect.

In a fourth aspect, alternative implementations of the present disclosure also provide a computer-readable storage medium having a computer program stored thereon, where the computer program is executed to perform the steps of the first aspect or any one of the possible implementations of the first aspect.

For the description of the effects of the resource distribution apparatus, the computer device, and the computer-readable storage medium, reference is made to the description of the resource distribution method, and details are not repeated here.

According to the resource distribution method, the resource distribution device, the computer equipment and the storage medium, the use conversion rate when different types of virtual resources are distributed to each user to be distributed can be accurately determined by using the trained gain network model; the users are grouped according to the determined use conversion rate, the distributed decision of each user group obtained by division is optimized by utilizing the resource quantity threshold and the user ratio of each target service used corresponding to the reference group, the type of the virtual resource to be distributed corresponding to each user group can be accurately determined on the basis of restricting the distributed resource quantity, and the virtual resource can be distributed for the users in each user group in a targeted manner, so that the reasonable utilization of the virtual resource can be realized, the utilization efficiency of the virtual resource is improved, and the utilization rate of the related service resource is further improved.

In order to make the aforementioned objects, features and advantages of the present disclosure more comprehensible, preferred embodiments accompanied with figures are described in detail below.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the drawings required in the embodiments will be briefly described below, and the drawings herein incorporated in and forming a part of the specification illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the technical solutions of the present disclosure. It is appreciated that the following drawings depict only certain embodiments of the disclosure and are therefore not to be considered limiting of its scope, for those skilled in the art will be able to derive additional related drawings therefrom without the benefit of the inventive faculty.

Fig. 1 shows a flowchart of a resource distribution method provided by an embodiment of the present disclosure;

FIG. 2 is a schematic diagram illustrating a specific implementation flow of resource distribution provided by an embodiment of the present disclosure;

fig. 3 is a schematic diagram illustrating a resource distribution apparatus provided by an embodiment of the present disclosure;

fig. 4 shows a schematic structural diagram of a computer device provided in an embodiment of the present disclosure.

Detailed Description

In order to make the objects, technical solutions and advantages of the embodiments of the present disclosure more clear, the technical solutions of the embodiments of the present disclosure will be described clearly and completely with reference to the drawings in the embodiments of the present disclosure, and it is obvious that the described embodiments are only a part of the embodiments of the present disclosure, not all of the embodiments. The components of embodiments of the present disclosure, as generally described and illustrated herein, may be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the present disclosure is not intended to limit the scope of the disclosure, as claimed, but is merely representative of selected embodiments of the disclosure. All other embodiments, which can be derived by a person skilled in the art from the embodiments of the disclosure without making creative efforts, shall fall within the protection scope of the disclosure.

Furthermore, the terms "first," "second," and the like in the description and claims of the embodiments of the disclosure and in the drawings described above are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It will be appreciated that the data so used may be interchanged under appropriate circumstances such that the embodiments described herein may be practiced otherwise than as specifically illustrated or described herein.

Reference herein to "a plurality or a number" means two or more. "and/or" describes the association relationship of the associated object, indicating that there may be three relationships, for example, a and/or B, which may indicate: a exists alone, A and B exist simultaneously, and B exists alone. The character "/" generally indicates that the former and latter associated objects are in an "or" relationship.

Research shows that virtual resources corresponding to target services are distributed for users as a conventional means for improving the utilization rate of the target services, and the sending accuracy and pertinence of the types of the virtual resources directly influence the improvement effect of the utilization rate of the target services. However, most conventional virtual resource distribution methods adopt a method of directly distributing virtual resources of various virtual resource types to users, so that the users can autonomously select one of the virtual resources for use. In this way, other virtual resources that are not used by the user are wasted, and the reasonability and pertinence of virtual resource transmission are reduced. Therefore, how to target various user target types of virtual resources becomes a technical pain point.

Based on the research, the resource distribution scheme provided by the disclosure can accurately determine the use conversion rate when different types of virtual resources are distributed to each user to be distributed by using the trained gain network model; the users are grouped according to the determined use conversion rate, the distributed decision of each user group obtained by division is optimized by utilizing the resource quantity threshold and the user ratio of each target service used corresponding to the reference group, the type of the virtual resource to be distributed corresponding to each user group can be accurately determined on the basis of restricting the distributed resource quantity, and the virtual resource can be distributed for the users in each user group in a targeted manner, so that the reasonable utilization of the virtual resource can be realized, the utilization efficiency of the virtual resource is improved, and the utilization rate of the related service resource is further improved.

The above-mentioned drawbacks are the results of the inventor after practical and careful study, and therefore, the discovery process of the above-mentioned problems and the solutions proposed by the present disclosure to the above-mentioned problems should be the contribution of the inventor in the process of the present disclosure.

It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, it need not be further defined or explained in subsequent figures.

To facilitate understanding of the present embodiment, first, a detailed description is given to a resource distribution method disclosed in the embodiments of the present disclosure, an execution main body of the resource distribution method provided in the embodiments of the present disclosure is generally a terminal device or other processing device with certain computing capability, where the terminal device may be a User Equipment (UE), a mobile device, a User terminal, a Personal Digital Assistant (PDA), a handheld device, a computer device, or the like; in some possible implementations, the resource distribution method may be implemented by a processor calling computer readable instructions stored in a memory.

The resource distribution method provided by the embodiment of the present disclosure is described below by taking an execution subject as a server.

As shown in fig. 1, a flowchart of a resource distribution method provided in the embodiment of the present disclosure may include the following steps:

s101: determining the use conversion rate when distributing different types of virtual resources to each user to be distributed by using the trained target network model; the usage conversion rate is used for indicating the probability of the user to be distributed using the target service after the virtual resource is distributed.

Here, the target service may be any one of web services, for example, an application mall service, a game service, a web shopping service, and the like. The virtual resource may be, for example, an application acceleration volume, a coupon volume, a virtual gold coin, or the like. One virtual resource type may correspond to at least one target service, and one target service may correspond to at least one virtual resource type. For example, virtual resource type A corresponds to target service A, virtual resource type B corresponds to target service B, and virtual resource C corresponds to target service A and target service C.

The target network model is a pre-trained network model, and can determine the use conversion rate of the target service of the virtual resource type used by the user after the virtual resource of the virtual resource type is distributed to the user to be distributed according to the input virtual resource type and the attribute information authorized by the user to be distributed. The usage conversion rate is used for indicating the probability that the user uses the target service by using the virtual resource after distributing the virtual resource to the user to be distributed. The user attribute information may include, for example, behavior information authorized by the user, feature information, usage record information of the usage target service, and the like.

In one embodiment, the trained target network model may be determined according to the following steps:

step one, collecting sample data according to a preset time period, and performing iterative training on the gain network model to be trained by using the collected sample data to obtain the trained gain network model.

Here, the preset time period may be set according to an actual application requirement, and the embodiment of the present disclosure is not particularly limited. Illustratively, the predetermined time period may be small enough to allow real-time collection of sample data. Of course, the preset time period may also be set to, for example, 1 hour, twelve hours, one day, two days, etc.

The sample data may include a virtual resource type and user attribute information for the authorized use of the sample user.

In specific implementation, a small-flow random exploration experiment mode can be utilized to randomly select sample users and randomly generate virtual resources of different virtual resource types to the selected sample users, and then the sample use conversion rate corresponding to the sample users can be obtained. Thus, continuous sample data acquisition can be realized by using a random exploration experiment.

After sample data is obtained, the virtual resource type and the user attribute information of the sample user can be spliced, the spliced data is used as the input of a gain (uplift) network model to be trained, then, the gain network model to be trained can be used for carrying out data processing on the input, and the corresponding predicted use conversion rate of the sample user is output. And determining the prediction loss by using the sample use conversion rate and the prediction use conversion rate corresponding to the sample user, and performing iterative training on the gain network model to be trained by using the prediction loss until a preset training cut-off condition is reached to obtain the trained gain network model. The gain network model to be trained is the target network model to be trained.

The preset training cutoff condition may include that the model prediction precision reaches the preset precision and/or the number of rounds of iterative training reaches the preset number of rounds.

And step two, determining a model evaluation index of the trained gain network model, and taking the trained gain network model as the trained target network model under the condition that the model evaluation index is larger than the model evaluation index of the verification gain model.

The verification gain model is used for verifying whether the trained gain network model meets the use requirements or not, is obtained by training sample data and has a different network structure from the gain network model.

The model evaluation index (Area Under vertical cut, AUUC for short) is used for representing the performance of the model. The verification gain model is a model used for verifying the performance of the trained gain network model, and the adopted training data is sample data used for training the gain network model to be trained, but has a network structure different from that of the gain network model.

The use requirement is that the model evaluation index of the trained gain network model is larger than the model evaluation index of the verification gain model, and under the condition of meeting the use requirement, the trained gain network model can be used online, and the use conversion rate corresponding to the user to be distributed can be accurately predicted in the online use process.

In specific implementation, the AUUC of the trained gain network model and the AUUC of the verification gain model may be compared, and the trained gain network model may be used online as the currently trained target network model when it is determined that the AUUC of the trained gain network model is greater than the AUUC index of the verification gain model.

On the contrary, when the AUUC of the trained gain network model is not greater than the AUUC standard of the verification gain model, the newly acquired sample data can be used to continue training the gain network model until the trained target network model is obtained and used online.

Therefore, the gain network model and the verification gain model are trained by using the continuously acquired sample data, so that the problem that the virtual resource distribution gain brought by the trained model is caused by different sample data or the problem that the model is self-caused because different sample data are used for training the gain network model and the verification gain model can be avoided, and the virtual resource distribution gain brought by the trained model is ensured to be brought by the model.

S101, in specific implementation, may splice user attribute information authorized by a user to be distributed and a virtual resource type of a virtual resource to be distributed, and use the spliced data as an input of a target network model, so as to obtain a usage conversion rate output by the target network model when distributing the virtual resource of the type to the user to be distributed. Based on the method, the user attribute information correspondingly authorized by different users to be distributed and different virtual resource types are spliced and input to the target network model, so that the use conversion rate when different types of virtual resources are distributed to each user to be distributed can be obtained.

S102: and dividing the users to be distributed into different user groups according to the use conversion rate respectively corresponding to the users to be distributed.

Here, at least one user to be distributed may be included in one user group.

For example, the number of the user groups to be divided may be determined according to the number of the users to be distributed, and then each user to be distributed is equally divided into different user groups according to the usage conversion rate and the number of the user groups respectively corresponding to each user to be distributed.

Taking the number of the users to be distributed as 100 as an example, the number of the user groups to be divided may be determined to be 10, and then, the 100 users to be distributed may be divided into 10 user groups according to the usage conversion rates corresponding to the 100 users to be distributed, respectively.

In an embodiment, for S102, each user to be distributed may be further divided into different user groups in any one of the following two manners:

according to the first mode, each user to be distributed is divided into different user groups according to the corresponding use conversion rate of each user to be distributed under any virtual resource type.

In specific implementation, any one virtual resource type can be selected from multiple virtual resource types to serve as an allocation resource type, and then, according to the usage conversion rate of each user to be allocated under the allocation resource type and the conversion rate interval corresponding to each preset user group, each user to be allocated is divided into different user groups.

Exemplarily, the virtual resource type a is used as an allocated resource type, a conversion rate interval in which the corresponding conversion rate of each user to be distributed is located is determined according to the corresponding conversion rate of each user to be distributed under the virtual resource type a, and then the user to be distributed can be divided into user groups associated with the conversion rate interval corresponding to the user to be distributed. Wherein one transition rate interval is associated with one user group.

And determining a conversion rate average value corresponding to each user to be distributed according to the corresponding use conversion rate of each user to be distributed under each virtual resource type, and dividing each user to be distributed into different user groups by using the conversion rate average value corresponding to each user to be distributed.

For example, a conversion rate interval corresponding to each user to be distributed may be determined according to a conversion rate average value corresponding to each user to be distributed, and then, the users to be distributed may be divided into user groups associated with the conversion rate intervals.

S103: determining the virtual resource type to be distributed corresponding to each user group based on the use conversion rate of the user to be distributed in each virtual resource type in each user group, the resource quantity threshold of the virtual resource and the user ratio of using each target service corresponding to the reference group; the users in the reference group refer to users to which virtual resources are not allocated.

Here, the users in the reference group are all users to which any type of virtual resource is not allocated, and the number of users in the reference group may be consistent with the number of users to be distributed.

One user group corresponds to one virtual resource type to be distributed, and the virtual resource of the virtual resource type to be distributed is the virtual resource which is most suitable for being sent to the user in the user group corresponding to the virtual resource type to be distributed.

The resource amount threshold for the virtual resource is used to indicate a maximum amount of resources allowed to be distributed. And the ratio of the users using each target service corresponding to the reference group is used for representing the ratio of the number of the users using the target service in the reference group to the number of the users not using the target service.

For example, the resource amount threshold of the virtual resource may be used as a constraint condition, and in a case that it is ensured that the total amount of resources for sending the virtual resource to each user group does not exceed the resource amount threshold, the virtual resource type to be distributed corresponding to each user group is determined according to the usage conversion rate of the user to be distributed in each virtual resource type in each user group and the user ratio corresponding to the reference group, with the difference between the usage conversion rate of the user to be distributed in each virtual resource type in the user group and the user ratio corresponding to the reference group being the maximum target.

In one embodiment, for S103, the following steps may be performed:

s103-1: and aiming at each user group, determining that after different types of virtual resources are distributed to users to be distributed in the user group, the users to be distributed use the first multiplexing information of the target service corresponding to different virtual resource types again.

Here, the first multiplexing information may be, for example, the number of times the target service is reused, the frequency of reusing the target service, and the length of time for reusing the target service. For example, in the case where the virtual resource is a coupon, the first reuse information may include information such as an amount of money for repurchasing an item corresponding to the coupon, and a repurchase rate. When the virtual resource is the skin (or the makeup) of the virtual object in the game, the first multiplexing information may include information such as a time length for playing the game again, the number of times of playing the game using the skin (or the makeup) again, a frequency, a time length, and a gain due to playing the game using the skin (or the makeup) again.

Illustratively, in the case that the user group includes user groups 1 to 3, the virtual resource types include virtual resource types a to C, and the target service includes a target service a corresponding to the virtual resource type a, a target service B corresponding to the virtual resource type B, and a target service C corresponding to the virtual resource type C, it may be determined, for the user group 1, that after the virtual resource of the virtual resource type a is distributed to the user to be distributed in the user group 1, the user to be distributed reuses the first multiplexing information of the target service a, after the virtual resource of the virtual resource type B is distributed to the user to be distributed in the user group 1, the user to be distributed reuses the first multiplexing information of the target service B, and after the virtual resource of the virtual resource type C is distributed to the user to be distributed in the user group 1, the user to be distributed reuses the first multiplexing information of the target service C. Similarly, for both the user group 2 and the user group 3, after different types of virtual resources are distributed to the users to be distributed in the user group, the users to be distributed use the first multiplexing information of the target service corresponding to different types of virtual resources again.

S103-2: and determining the virtual resource distribution gains of different virtual resource types corresponding to the user group according to the first multiplexing information of the user group.

Here, for one user group, one virtual resource transmission gain may exist for each virtual resource type. For example, for the user group 1, the corresponding virtual resource distribution gains may include: a virtual resource distribution gain 1 corresponding to the virtual resource type a, a virtual resource distribution gain 2 corresponding to the virtual resource type B, and a virtual resource distribution gain 3 corresponding to the virtual resource type C. The virtual resource distribution gain is used for representing the gain brought by the target service used by the user to be distributed after the virtual resource is distributed to the user to be distributed in the user group.

In specific implementation, for each user group, according to a preset gain calculation formula, the virtual resource distribution gains of the user group under different virtual resource types are determined according to the first multiplexing information corresponding to the user group under different virtual resource types respectively.

In an embodiment, for the above S103-1, for each user group, it may be determined that after different types of virtual resources are distributed to users to be distributed in the user group, the users to be distributed reuse the first multiplexing information of the target service corresponding to different virtual resource types in different preset time periods.

Here, the number and the time duration of the preset time periods may be set according to actual application requirements, and the disclosed example is not particularly limited. For example, the preset time period may include the first N days (e.g., 30 days) after the different types of virtual resources are distributed to the users to be distributed in the user group, and may further include the M to M +10 days (e.g., 21 to 30 days) after the different types of virtual resources are distributed to the users to be distributed in the user group.

Taking the user group comprising the user group 1 and the user group 2, the virtual resource type comprising the virtual resource types a and B, the target service comprising the target service a corresponding to the virtual resource type a, and the target service B corresponding to the virtual resource type B as an example, for the user group 1, it may be determined that, after the virtual resource corresponding to the virtual resource type a is distributed to the user to be distributed in the user group 1, the user to be distributed in the user group 1 uses the gain of the target service a within the previous 30 days, and it is determined that the frequency of using the target service a by the user to be distributed in the user group 1 on the 21 st to 30 th days. Similarly, after distributing the virtual resource corresponding to the virtual resource type B to the user to be distributed in the user group 1, the gain that the user to be distributed in the user group 1 uses the target service B in the previous 30 days may be determined, and the frequency that the user to be distributed in the user group 1 uses the target service B in the 21 st to 30 th days may be determined.

Similarly, for the user group 2, the gain of using the target service a by the user to be distributed in the user group 2 within the previous 30 days after distributing the virtual resource corresponding to the virtual resource type a to the user to be distributed in the user group 2 may be determined, and the frequency of using the target service a by the user to be distributed in the user group 2 on the 21 st to 30 th days may be determined. Similarly, after distributing the virtual resource corresponding to the virtual resource type B to the user to be distributed in the user group 2, the gain that the user to be distributed in the user group 2 uses the target service B in the previous 30 days may be determined, and the frequency that the user to be distributed in the user group 2 uses the target service B in the 21 st to 30 th days may be determined.

Further, for S103-2, for each virtual resource type, combining the first multiplexing information of the target service that uses the virtual resource type again for the user to be distributed in the user group in different preset time periods, so as to obtain the virtual resource distribution gain of the user group in the virtual resource type.

For example, the virtual resource distribution gain of the user group under the virtual resource type can be determined according to the following formula one:

LTV _ij ＝(RPG30 _ij frequency of using the target service on days 21 to 30 a mean value of the target service on days 21 to 30 a (a 1 a2-a 3)). A4, (formula one)

Wherein, LTV _ij Representing the virtual resource distribution gain of the user group i under the jth virtual resource type, RPG30 _ij And the gain of the target service corresponding to the jth virtual resource type used by the user to be distributed in the user group i in the previous 30 days is represented, and the average value is used for representing the value cost of the target service. a1, a2, a3 and a4 are different values set in advance.

In specific implementation, according to the first multiplexing information corresponding to each user group in different virtual resource types and the first formula, the virtual resource distribution gain of each user group in different virtual resource types can be determined respectively.

S103-3: and determining the conversion rate increment of each user group under each virtual resource type according to the use conversion rate of the user to be distributed in each user group under each virtual resource type and the user ratio corresponding to the reference group.

Here, the conversion increment is used to characterize the difference between the usage conversion and the user ratio. Taking the determination of the conversion rate increment of any user group under any virtual resource type as an example, the usage conversion rate of each user to be distributed in the user group under the virtual resource type may be determined, and the difference between the user rate corresponding to the reference group and the user rate corresponding to each user to be distributed may be determined, and then the average of the differences corresponding to each user to be distributed may be used as the conversion rate increment of the user group under the virtual resource type.

In one embodiment, for S103-3, the following steps may be performed:

s103-3-1: and aiming at each user group, determining the average use conversion rate of the user group under each virtual resource type according to the use conversion rate of each user to be distributed in the user group under each virtual resource type.

For example, for a user group 1 including users to be distributed 1 to 3, an average usage conversion rate of the user group 1 in the virtual resource type a may be determined according to usage conversion rates of the users to be distributed 1 to 3 in the virtual resource type a, respectively. And determining the average use conversion rate of the user group 1 under the virtual resource type B according to the use conversion rates of the users 1-3 to be distributed under the virtual resource type B. And determining the average use conversion rate of the user group 1 under the virtual resource type C according to the use conversion rates of the users 1-3 to be distributed under the virtual resource type C.

S103-3-2: and determining the difference between the average use conversion rate of each user group under each virtual resource type and the user ratio corresponding to the reference group as the conversion rate increment of each user group under each virtual resource type.

For example, the difference between the average usage conversion rate of the user group 1 in the virtual resource type a and the user rate corresponding to the reference group may be used as the conversion rate increment of the user group 1 in the virtual resource type a; taking the difference value of the average use conversion rate of the user group 1 under the virtual resource type B and the user ratio corresponding to the reference group as the conversion rate increment of the user group 1 under the virtual resource type B; and taking the difference value of the average use conversion rate of the user group 1 in the virtual resource type C and the user ratio corresponding to the reference group as the conversion rate increment of the user group 1 in the virtual resource type C.

S103-4: and determining the type of the virtual resources to be distributed corresponding to each user group according to the maximum gain of the virtual resource distribution and the resource quantity threshold of the virtual resources.

For example, the type of virtual resource to be distributed corresponding to each user group may be determined according to the following formula:

max∑ _i，j ltv _ij *Δr _ij *p _ij (formula two)

s.t.∑ _j p _ij ，p _ij E (0, 1), (formula three)

∑ _i (c _ij +Δr _ij )p _ij ≤thres ₂ *ost _base (formula four)

Where max represents the maximization, i represents the ith user group, j represents the jth virtual resource type, ltv _ij Represents the virtual resource distribution gain, Δ r, of the user group i under the virtual resource type j _ij Indicating that user group i corresponds to virtual resource type jIncrement of conversion, p _ij Representing the probability of distributing virtual resources of the jth virtual resource type to user group i, s.t. with \8230;, as constraints, c _ij Indicating the amount of resources, thres, in distributing virtual resources of the jth virtual resource type to user group i ₂ *cost _base Representing a resource amount threshold for the virtual resource.

In specific implementation, the virtual resource distribution gain, the conversion rate increment and the resource amount threshold of the virtual resource can be substituted into the above formulas, and the virtual resource type to be distributed corresponding to each user group is determined by combining the above formulas. For example, by combining the above formulas, p of each user group under each virtual resource type can be determined _ij Then, the maximum p per virtual resource type can be determined according to the user groups _ij And the indicated virtual resource type is used as the virtual resource type to be distributed corresponding to the user group.

In an embodiment, after the virtual resource type to be distributed corresponding to each user group is determined, an association relationship between the user group and the virtual resource type to be distributed may be established and stored. For example, the storage may be performed by means of a distributed Transaction (TCC).

In an embodiment, for the above S103-4, the following steps may be performed:

s103-4-1: and determining the virtual resource type to be verified corresponding to each user group according to the maximum virtual resource distribution gain principle and the virtual resource distribution gain, the conversion rate increment and the resource quantity threshold of the virtual resources.

In specific implementation, the virtual resource type to be verified corresponding to each user group can be determined according to the second to fourth formulas, and then the virtual resource type to be verified corresponding to the user group can be verified according to the following steps.

S103-4-2: and determining second multiplexing information of target services corresponding to different virtual resource types used by the users in the reference group.

In specific implementation, the time point of distributing the different types of virtual resources to each user to be distributed may be used as a time starting point, and the users in the reference group may use the second multiplexing information of the target service corresponding to the different types of virtual resources in different preset time periods.

S103-4-3: and determining the reference resource distribution gain of the reference group corresponding to different virtual resource types according to the second multiplexing information.

For example, the reference resource distribution gain of the reference group under the virtual resource type may be determined according to the following formula five:

wherein the content of the first and second substances,

showing the reference resource distribution gain of the reference group under the jth virtual resource type,

and the gain of the target service corresponding to the jth virtual resource type used by the users in the reference group in the previous 30 days is shown, and the average value is used for representing the value cost of the target service. a1, a2, a3 and a4 are different values set in advance.

In specific implementation, the reference resource distribution gains of the reference groups corresponding to different virtual resource types are determined according to the second multiplexing information of the reference groups corresponding to different virtual resource types and the formula four.

S103-4-4: and aiming at each virtual resource type to be verified, taking the virtual resource type to be verified as the virtual resource type to be distributed of the corresponding user group under the condition that the virtual resource distribution gain of the user group corresponding to the virtual resource type to be verified is larger than the reference resource distribution gain corresponding to the reference group under the virtual resource type to be verified.

For example, for each virtual resource type to be verified, in a case that only one user group is included in the user group corresponding to the virtual resource type to be verified, a virtual resource distribution gain of the unique user group in the virtual resource type to be verified and a reference resource distribution gain of a reference group in the virtual resource type to be verified may be compared, and in a case that the virtual resource distribution gain is greater than the reference resource distribution gain, the virtual resource type to be verified may be used as the virtual resource type to be distributed corresponding to the unique user group. On the contrary, when the virtual resource distribution gain is smaller than the reference resource distribution gain, it indicates that the type of the virtual resource to be verified is currently determined to be unreasonable, and the step S103-4-1 may be returned to be executed to re-determine the type of the virtual resource to be verified corresponding to each user group until the virtual resource distribution gain is larger than the reference resource distribution gain.

For each virtual resource type to be verified, if the user group corresponding to the virtual resource type to be verified only includes a plurality of user groups, for example, the user group corresponding to the virtual resource type a to be verified includes user group 4 and user group 5, the sum of the virtual resource distribution gains may be determined according to the virtual resource distribution gains of the user group 4 and the user group 5 under the virtual resource type to be verified, respectively, and if the sum is greater than the reference resource distribution gain corresponding to the reference group under the virtual resource type to be verified, the virtual resource type a to be verified is used as the virtual resource type to be distributed corresponding to the user group 4 and the user group 5, respectively.

Therefore, the virtual resource types to be verified corresponding to the determined user groups can be verified reasonably by utilizing the reference resource distribution gain corresponding to the reference group, and the reasonable and accurate virtual resource types to be distributed are obtained.

S104: and distributing the virtual resources to each user to be distributed in the user group according to the type of the virtual resources to be distributed corresponding to each user group.

In specific implementation, for each user group, the virtual resources of the virtual resource types to be distributed corresponding to the user group may be distributed to each user to be distributed in the user group. For example, if the virtual resource type to be distributed corresponding to the user group 1 is a virtual resource type B, and the virtual resource type to be distributed corresponding to the user group 2 is a virtual resource type a, the virtual resource of the virtual resource type B may be distributed to each user to be distributed in the user group 1, and the virtual resource of the virtual resource type a may be distributed to each user to be distributed in the user group 2.

Optionally, after determining the usage conversion rate when different types of virtual resources are distributed to each user to be distributed by using the target neural network, each usage conversion rate corresponding to the user to be distributed may be stored. For example, in the distributed storage space Abase. Further, after the association relationship between the user group and the type of the virtual resource to be distributed is stored, if the other server side has a requirement for distributing the virtual resource for the user to be distributed, the use conversion rate of the user to be distributed can be determined from the Abase according to the user identification of the user to be distributed; then, according to the usage conversion rate, determining a user group to which the user to be distributed belongs, and according to the incidence relation corresponding to the user group, determining the type of the virtual resource to be distributed; and finally, sending the virtual resource of the virtual resource type to be distributed to the user to be distributed.

Based on the embodiments, the utilization conversion rate when different types of virtual resources are distributed to each user to be distributed can be accurately determined by using the trained gain network model; the users are grouped according to the determined use conversion rate, the distributed decision of each user group obtained by division is optimized by utilizing the resource quantity threshold and the user ratio of each target service used corresponding to the reference group, the type of the virtual resource to be distributed corresponding to each user group can be accurately determined on the basis of restricting the distributed resource quantity, and the virtual resource can be distributed for the users in each user group in a targeted manner, so that the reasonable utilization of the virtual resource can be realized, the utilization efficiency of the virtual resource is improved, and the utilization rate of the related service resource is further improved.

In an embodiment, as shown in fig. 2, a flowchart of a specific implementation of resource distribution provided for the embodiment of the present disclosure may include three parts, a first part is a model updating part (including real-time updating and/or day-level updating), a second part is a decision optimizing part (which may be optimized in real-time and/or day-level optimizing), and a third part is an online service part, which is used for distributing a virtual resource for each user to be distributed online. Specifically, the model updating part may include: continuously collecting sample data by using a small-flow random exploration experiment mode; the sample data comprises a virtual resource type and user attribute information which is used by a sample user in a corresponding authorization mode. Splicing the virtual resource types and the user attribute information which is authorized to be used by the sample user correspondingly, taking the spliced data as the input of a gain (uplift) network model to be trained, and performing iterative training on the gain (uplift) network model to be trained to obtain a target network model.

The decision optimization part may include: and determining the use conversion rate when different types of virtual resources are distributed to each user to be distributed by using the target network model, and storing the use conversion rate corresponding to each user to be distributed into the Abase. And determining the virtual resource distribution gain corresponding to each user to be distributed after different types of virtual resources are distributed to each user to be distributed by using a pre-trained gain information prediction model, and storing the gain into the Abase. The gain information prediction model can also be obtained by adopting sample data training. And then, according to the use conversion rate respectively corresponding to each user to be distributed, each user to be distributed can be divided into different user groups. And aiming at each user group, determining that after different types of virtual resources are distributed to users to be distributed in the user group, the users to be distributed use the first multiplexing information of the target service corresponding to different virtual resource types again. Then, decision optimization, that is, the above S103, may be performed to determine the type of virtual resource to be distributed corresponding to each user group. And establishing an association relation between the user group and the virtual resource type to be distributed and writing the TCC. The virtual resource distribution gain output by the gain information prediction model may be used to verify whether the optimization result (i.e., the type of virtual resource to be distributed corresponding to the user group) obtained by the decision optimization part is reasonable, and specifically, the optimization result may be determined to be reasonable when the virtual resource distribution gain obtained after distributing the virtual resource of the type of virtual resource to be distributed to the user to be distributed is determined to be greater than the virtual resource distribution gain output by the gain information prediction model.

The online service part may include: any upstream service determines that virtual resources need to be issued to users to be distributed, and the flying disc (server) can obtain the use conversion rate corresponding to the users to be distributed from the Abase according to the user identification corresponding to the users to be distributed, and determine the user group corresponding to the users to be distributed according to the use conversion rate. And acquiring the incidence relation from the TCC, and determining the type of the virtual resource to be distributed corresponding to the user to be distributed according to the incidence relation. Thereafter, the virtual resource of the type may be distributed to the user to be distributed.

It will be understood by those of skill in the art that in the above method of the present embodiment, the order of writing the steps does not imply a strict order of execution and does not impose any limitations on the implementation, as the order of execution of the steps should be determined by their function and possibly inherent logic.

Based on the same inventive concept, a resource distribution device corresponding to the resource distribution method is also provided in the embodiments of the present disclosure, and because the principle of solving the problem of the device in the embodiments of the present disclosure is similar to the resource distribution method in the embodiments of the present disclosure, the implementation of the device may refer to the implementation of the method, and repeated details are not repeated.

As shown in fig. 3, a schematic diagram of a resource distribution apparatus provided for an embodiment of the present disclosure includes:

a first determining module 301, configured to determine, by using the trained target network model, a usage conversion rate when different types of virtual resources are distributed to each user to be distributed; the usage conversion rate is used for indicating the probability of using the target service after the virtual resource is distributed to the user to be distributed;

a dividing module 302, configured to divide each user to be distributed into different user groups according to the usage conversion rate corresponding to each user to be distributed;

a second determining module 303, configured to determine, based on the usage conversion rate of the user to be distributed in each virtual resource type in each user group, a resource amount threshold of a virtual resource, and a user ratio of using each target service corresponding to a reference group, a virtual resource type to be distributed corresponding to each user group; the users in the reference group refer to users which are not allocated with virtual resources;

a distributing module 304, configured to distribute virtual resources to each user to be distributed in the user group according to the type of the virtual resource to be distributed corresponding to each user group.

In a possible implementation manner, when the dividing module 302 divides each user to be distributed into different user groups according to the usage conversion rate respectively corresponding to each user to be distributed, the dividing module is configured to:

In a possible embodiment, the apparatus further comprises:

a training module 305, configured to determine a trained target network model according to the following steps:

the verification gain model is used for verifying whether the trained gain network model meets the use requirement, and is obtained by training the verification gain model by using the sample data and has a different network structure from the gain network model.

In a possible implementation manner, the second determining module 303, when determining the virtual resource type to be distributed corresponding to each user group based on the usage conversion rate of the user to be distributed in each virtual resource type in each user group, the resource amount threshold of the virtual resource, and the user ratio of using each target service corresponding to the reference group, is configured to:

for each user group, determining that after the virtual resources of different types are distributed to the users to be distributed in the user group, the users to be distributed use the first multiplexing information of the target service corresponding to different virtual resource types again;

In a possible implementation manner, the second determining module 303, when determining that, after the virtual resources of different types are distributed to the user to be distributed in the user group, the user to be distributed uses the first multiplexing information of the target service corresponding to different virtual resource types again, is configured to:

In a possible implementation manner, the second determining module 303, when determining, according to the usage conversion rate of the user to be distributed in each virtual resource type in each user group and the user ratio corresponding to a reference group, a conversion rate increment of each virtual resource type in each user group, is configured to:

In a possible implementation manner, the second determining module 303, when determining, according to the principle that the virtual resource distribution gain is the largest, the type of the virtual resource to be distributed corresponding to each user group according to the virtual resource distribution gain, the conversion rate increment, and the resource amount threshold of the virtual resource, is configured to:

The description of the processing flow of each module in the device and the interaction flow between the modules may refer to the related description in the above method embodiments, and will not be described in detail here.

Based on the same technical concept, the embodiment of the application also provides computer equipment. Referring to fig. 4, a schematic structural diagram of a computer device provided in an embodiment of the present application includes:

a processor 41, a memory 42, and a bus 43. Wherein the memory 42 stores machine-readable instructions executable by the processor 41, the processor 41 is configured to execute the machine-readable instructions stored in the memory 42, and when the machine-readable instructions are executed by the processor 41, the processor 41 performs the following steps: s101: determining the use conversion rate when different types of virtual resources are distributed to each user to be distributed by using the trained target network model; the use conversion rate is used for indicating the probability of using the target service after the virtual resource is distributed to the user to be distributed; s102: dividing each user to be distributed into different user groups according to the use conversion rate corresponding to each user to be distributed; s103: determining the virtual resource type to be distributed corresponding to each user group based on the use conversion rate of the user to be distributed in each virtual resource type in each user group, the resource quantity threshold of the virtual resource and the user ratio of using each target service corresponding to the reference group; the users in the reference group are users to which virtual resources are not allocated, and S104: and distributing the virtual resources to each user to be distributed in the user group according to the type of the virtual resources to be distributed corresponding to each user group.

The storage 42 includes a memory 421 and an external storage 422; the memory 421 is also referred to as an internal memory, and is used for temporarily storing the operation data in the processor 41 and the data exchanged with the external storage 422 such as a hard disk, the processor 41 exchanges data with the external storage 422 through the memory 421, and when the computer device is operated, the processor 41 communicates with the storage 42 through the bus 43, so that the processor 41 executes the execution instructions mentioned in the above method embodiments.

The embodiments of the present disclosure further provide a computer-readable storage medium, where a computer program is stored, and when the computer program is executed by a processor, the computer program performs the steps of the resource distribution method in the foregoing method embodiments. The storage medium may be a volatile or non-volatile computer-readable storage medium.

The computer program product of the resource distribution method provided in the embodiments of the present disclosure includes a computer-readable storage medium storing a program code, where instructions included in the program code may be used to execute steps of the resource distribution method described in the above method embodiments, which may be referred to specifically for the above method embodiments, and are not described herein again.

The computer program product may be embodied in hardware, software or a combination thereof. In an alternative embodiment, the computer program product is embodied in a computer storage medium, and in another alternative embodiment, the computer program product is embodied in a Software product, such as a Software Development Kit (SDK), or the like.

It can be clearly understood by those skilled in the art that, for convenience and simplicity of description, the specific working process of the system and the apparatus described above may refer to the corresponding process in the foregoing method embodiment, and details are not described herein again. In the several embodiments provided in the present disclosure, it should be understood that the disclosed system, apparatus, and method may be implemented in other ways. The above-described apparatus embodiments are merely illustrative, and for example, the division of the units into only one type of logical function may be implemented in other ways, and for example, multiple units or components may be combined or integrated into another system, or some features may be omitted, or not implemented. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection of devices or units through some communication interfaces, and may be in an electrical, mechanical or other form.

The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.

In addition, functional units in the embodiments of the present disclosure may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit.

The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a non-volatile computer-readable storage medium executable by a processor. Based on such understanding, the technical solution of the present disclosure may be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present disclosure. And the aforementioned storage medium includes: various media capable of storing program codes, such as a usb disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.

If the technical scheme of the application relates to personal information, a product applying the technical scheme of the application clearly informs personal information processing rules before processing the personal information, and obtains personal independent consent. If the technical scheme of the application relates to sensitive personal information, a product applying the technical scheme of the application obtains individual consent before processing the sensitive personal information, and simultaneously meets the requirement of 'express consent'. For example, at a personal information collection device such as a camera, a clear and significant identifier is set to inform that the personal information collection range is entered, the personal information is collected, and if the person voluntarily enters the collection range, the person is regarded as agreeing to collect the personal information; or on the device for processing the personal information, under the condition of informing the personal information processing rule by using obvious identification/information, obtaining personal authorization by modes of popping window information or asking a person to upload personal information of the person by himself, and the like; the personal information processing rule may include information such as a personal information processor, a personal information processing purpose, a processing method, and a type of personal information to be processed.

Finally, it should be noted that: the above-mentioned embodiments are merely specific embodiments of the present disclosure, which are used for illustrating the technical solutions of the present disclosure and not for limiting the same, and the scope of the present disclosure is not limited thereto, and although the present disclosure is described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: any person skilled in the art can modify or easily conceive of the technical solutions described in the foregoing embodiments or equivalent technical features thereof within the technical scope of the present disclosure; such modifications, changes or substitutions do not depart from the spirit and scope of the embodiments of the present disclosure, and should be construed as being included therein. Therefore, the protection scope of the present disclosure shall be subject to the protection scope of the claims.

Claims

1. A method for resource distribution, comprising:

determining the use conversion rate when different types of virtual resources are distributed to each user to be distributed by using the trained target network model; the usage conversion rate is used for indicating the probability of using the target service after the virtual resource is distributed to the user to be distributed;

determining the virtual resource type to be distributed corresponding to each user group based on the usage conversion rate of the user to be distributed in each virtual resource type, the resource quantity threshold of the virtual resource and the user ratio of using each target service corresponding to the reference group; the users in the reference group refer to users which are not allocated with virtual resources;

2. The method according to claim 1, wherein the dividing the users to be distributed into different user groups according to the usage conversion rates respectively corresponding to the users to be distributed comprises:

3. The method of claim 1, wherein the trained target network model is determined according to the following steps:

4. The method according to claim 1, wherein the determining the virtual resource type to be distributed for each user group based on the usage conversion rate of the user to be distributed in each virtual resource type, the resource amount threshold of the virtual resource, and the user ratio using each target service corresponding to a reference group for each user group comprises:

and determining the type of the virtual resource to be distributed corresponding to each user group according to the maximum virtual resource distribution gain principle and the virtual resource distribution gain, the conversion rate increment and the resource quantity threshold of the virtual resource.

5. The method according to claim 4, wherein the determining that the user to be distributed reuses first multiplexing information of the target service corresponding to different virtual resource types after distributing different types of virtual resources to the user to be distributed in the user group comprises:

after determining that different types of virtual resources are distributed to the users to be distributed in the user group, the users to be distributed reuse the first multiplexing information of the target service corresponding to different types of virtual resources within different preset time periods;

and aiming at each virtual resource type, combining the first multiplexing information of the target service which uses the virtual resource type again for the user to be distributed in the user group in different preset time periods to obtain the virtual resource distribution gain of the user group under the virtual resource type.

6. The method according to claim 4, wherein the determining, according to the usage conversion rate of the user to be distributed in each virtual resource type in each user group and the user ratio corresponding to a reference group, a conversion rate increment of each user group in each virtual resource type comprises:

7. The method according to claim 4, wherein the determining, according to a principle that a virtual resource distribution gain is maximum, a type of virtual resources to be distributed corresponding to each user group according to the virtual resource distribution gain, the conversion rate increment, and a resource amount threshold of the virtual resources comprises:

determining the type of the virtual resource to be verified corresponding to each user group according to the maximum virtual resource distribution gain principle, the virtual resource distribution gain, the conversion rate increment and the resource quantity threshold of the virtual resource;

8. A resource distribution apparatus, comprising:

the first determining module is used for determining the use conversion rate when different types of virtual resources are distributed to each user to be distributed by utilizing the trained target network model; the usage conversion rate is used for indicating the probability of using the target service after the virtual resource is distributed to the user to be distributed;

the dividing module is used for dividing the users to be distributed into different user groups according to the use conversion rate respectively corresponding to the users to be distributed;

9. A computer device, comprising: a processor, a memory storing machine-readable instructions executable by the processor, the processor to execute the machine-readable instructions stored in the memory, the processor to perform the steps of the resource distribution method of any one of claims 1 to 7 when the machine-readable instructions are executed by the processor.

10. A computer-readable storage medium, characterized in that a computer program is stored on the computer-readable storage medium, which computer program, when executed by a computer device, performs the steps of the resource distribution method according to any one of claims 1 to 7.