A resource distribution method, apparatus, computer device, and storage medium

Abstract

This disclosure provides a resource distribution method, apparatus, computer device, and storage medium. The method includes: using a trained target network model to determine the usage conversion rate when distributing different types of virtual resources to each user to be distributed; dividing each user to be distributed into different user groups based on the usage conversion rate corresponding to each user; determining the type of virtual resource to be distributed for each user group based on the usage conversion rate of each user group under each type of virtual resource, the resource quantity threshold of the virtual resource, and the user ratio of each target service in the reference group; users in the reference group refer to users who have not been allocated virtual resources; and distributing virtual resources to each user in the user group according to the type of virtual resource to be distributed for each user group.

Description

Technical Field

[0001] This disclosure relates to the field of computer technology, and more specifically, to a resource distribution method, apparatus, computer device, and storage medium. Background Technology

[0002] To increase the utilization rate of various target services, it is common practice to distribute virtual resources of different types to users. This increases users' willingness to use the target services by using the distributed virtual resources, thereby improving the utilization rate of service resources.

[0003] However, since different users have different usage habits and different needs for various types of virtual resources, how to distribute virtual resources to each user in a targeted manner has become an urgent problem to be solved. Summary of the Invention

[0004] This disclosure provides at least one resource distribution method, apparatus, computer device, and storage medium.

[0005] In a first aspect, embodiments of this disclosure provide a resource distribution method, including:

[0006] Using a trained target network model, the usage conversion rate is determined when different types of virtual resources are distributed to each user to be distributed; the usage conversion rate is used to indicate the probability that the user to be distributed will use the target service after the virtual resources are distributed.

[0007] Based on the usage conversion rate corresponding to each of the users to be distributed, the users to be distributed are divided into different user groups;

[0008] Based on the usage conversion rate of the users to be distributed in each user group under each virtual resource type, the resource quantity threshold of the virtual resource, and the ratio of users using each target service in the reference group, the virtual resource type to be distributed for each user group is determined; the users in the reference group refer to users who have not been allocated virtual resources.

[0009] Based on the type of virtual resource to be distributed for each user group, virtual resources are distributed to each user within that user group.

[0010] In one possible implementation, dividing the users to be distributed into different user groups based on the usage conversion rate corresponding to each user includes:

[0011] Based on the usage conversion rate of each user to be distributed under any of the virtual resource types, the users to be distributed are divided into different user groups; or,

[0012] Based on the usage conversion rate of each user to be distributed under each virtual resource type, the average conversion rate of each user to be distributed is determined, and the average conversion rate of each user to be distributed is used to divide each user to be distributed into different user groups.

[0013] In one possible implementation, the trained target network model is determined according to the following steps:

[0014] According to the preset time period, sample data is collected, and the collected sample data is used to iteratively train the gain network model to be trained, so as to obtain the trained gain network model.

[0015] Determine the model evaluation index of the trained gain network model, and if the model evaluation index is greater than the model evaluation index of the validation gain model, use the trained gain network model as the trained target network model.

[0016] The verification gain model is used to verify whether the trained gain network model meets the usage requirements. The verification gain model is trained using the sample data and has a different network structure from the gain network model.

[0017] In one possible implementation, determining the virtual resource type to be distributed for each user group based on the usage conversion rate of the users to be distributed in each user group under each virtual resource type, the resource quantity threshold of the virtual resource, and the user ratio of the reference group corresponding to each target service includes:

[0018] For each user group, it is determined that after distributing different types of virtual resources to the users to be distributed in that user group, the users to be distributed will use the target service corresponding to the different virtual resource types again.

[0019] Based on the first reuse information of the user group, determine the virtual resource distribution gain corresponding to different virtual resource types for the user group;

[0020] Based on the usage conversion rate of the users to be distributed in each user group under each virtual resource type and the user ratio corresponding to the reference group, determine the conversion rate increment of each user group under each virtual resource type;

[0021] Based on the principle of maximizing virtual resource distribution gain, the type of virtual resource to be distributed for each user group is determined according to the virtual resource distribution gain, the conversion rate increment, and the resource quantity threshold of the virtual resource.

[0022] In one possible implementation, determining the first reuse information for the target service corresponding to the different virtual resource types that the user to be distributed to will use again after distributing different types of virtual resources to the user to be distributed in the user group includes:

[0023] After determining that different types of virtual resources are distributed to the users to be distributed in the user group, the first reuse information of the target service corresponding to different virtual resource types is used again by the users to be distributed in different preset time periods.

[0024] The step of determining the virtual resource distribution gain for different virtual resource types corresponding to the user group based on the first multiplexing information of the user group includes:

[0025] For each virtual resource type, the first reuse information of the target service of the virtual resource type is merged when the user to be distributed in the user group uses the virtual resource type again in different preset time periods to obtain the virtual resource distribution gain of the user group under the virtual resource type.

[0026] In one possible implementation, determining the conversion rate increment for each user group under each virtual resource type based on the usage conversion rate of the users to be distributed in each user group under each virtual resource type and the user ratio corresponding to the reference group includes:

[0027] For each user group, the average usage conversion rate of the user group under each virtual resource type is determined based on the usage conversion rate of each user to be distributed in the user group under each virtual resource type;

[0028] The difference between the average usage conversion rate of each user group under each virtual resource type and the user ratio corresponding to the reference group is determined as the conversion rate increment of each user group under each virtual resource type.

[0029] In one possible implementation, determining the type of virtual resource to be distributed for each user group based on the principle of maximizing virtual resource distribution gain, the conversion rate increment, and the resource quantity threshold of the virtual resource, includes:

[0030] Based on the principle of maximizing virtual resource distribution gain, the virtual resource type to be verified for each user group is determined according to the virtual resource distribution gain, the conversion rate increment, and the virtual resource quantity threshold.

[0031] Determine the second reuse information of the target services corresponding to different virtual resource types used by users in the reference group;

[0032] Based on the second reuse information, determine the reference resource distribution gain corresponding to different virtual resource types for the reference group;

[0033] For each virtual resource type to be verified, if the virtual resource distribution gain of the user group corresponding to the virtual resource type to be verified is greater than the reference resource distribution gain of the reference group under the virtual resource type to be verified, then the virtual resource type to be verified shall be regarded as the virtual resource type to be distributed for the corresponding user group.

[0034] Secondly, embodiments of this disclosure also provide a resource distribution device, comprising:

[0035] The first determining module is used to determine the usage conversion rate when distributing different types of virtual resources to each user to be distributed, using a trained target network model; the usage conversion rate is used to indicate the probability that the user to be distributed will use the target service after being distributed the virtual resources.

[0036] The segmentation module is used to divide each user to be distributed into different user groups based on the usage conversion rate corresponding to each user to be distributed.

[0037] The second determining module is used to determine the virtual resource type to be distributed for each user group based on the usage conversion rate of the users to be distributed in each user group under each virtual resource type, the resource quantity threshold of the virtual resource, and the ratio of users using each target service in the reference group; the users in the reference group refer to users who have not been allocated virtual resources.

[0038] The distribution module is used to distribute virtual resources to each user in the user group according to the type of virtual resource to be distributed corresponding to each user group.

[0039] In one possible implementation, the segmentation module, when dividing the users to be distributed into different user groups according to the usage conversion rate corresponding to each user, is used to:

[0040] Based on the usage conversion rate of each user to be distributed under any of the virtual resource types, the users to be distributed are divided into different user groups; or,

[0041] Based on the usage conversion rate of each user to be distributed under each virtual resource type, the average conversion rate of each user to be distributed is determined, and the average conversion rate of each user to be distributed is used to divide each user to be distributed into different user groups.

[0042] In one possible implementation, the device further includes:

[0043] The training module is used to determine the trained target network model according to the following steps:

[0044] According to the preset time period, sample data is collected, and the collected sample data is used to iteratively train the gain network model to be trained, so as to obtain the trained gain network model.

[0045] Determine the model evaluation index of the trained gain network model, and if the model evaluation index is greater than the model evaluation index of the validation gain model, use the trained gain network model as the trained target network model.

[0046] The verification gain model is used to verify whether the trained gain network model meets the usage requirements. The verification gain model is trained using the sample data and has a different network structure from the gain network model.

[0047] In one possible implementation, the second determining module, when determining the virtual resource type to be distributed for each user group based on the usage conversion rate of the users to be distributed in each user group under each virtual resource type, the resource quantity threshold of the virtual resource, and the user ratio of the reference group corresponding to each target service, is configured to:

[0048] For each user group, it is determined that after distributing different types of virtual resources to the users to be distributed in that user group, the users to be distributed will use the target service corresponding to the different virtual resource types again.

[0049] Based on the first reuse information of the user group, determine the virtual resource distribution gain corresponding to different virtual resource types for the user group;

[0050] Based on the usage conversion rate of the users to be distributed in each user group under each virtual resource type and the user ratio corresponding to the reference group, determine the conversion rate increment of each user group under each virtual resource type;

[0051] Based on the principle of maximizing virtual resource distribution gain, the type of virtual resource to be distributed for each user group is determined according to the virtual resource distribution gain, the conversion rate increment, and the resource quantity threshold of the virtual resource.

[0052] In one possible implementation, the second determining module, when determining the first reuse information of the target service corresponding to the different virtual resource types after distributing different types of virtual resources to the users to be distributed in the user group, is configured to:

[0053] After determining that different types of virtual resources are distributed to the users to be distributed in the user group, the first reuse information of the target service corresponding to different virtual resource types is used again by the users to be distributed in different preset time periods.

[0054] The step of determining the virtual resource distribution gain for different virtual resource types corresponding to the user group based on the first multiplexing information of the user group includes:

[0055] For each virtual resource type, the first reuse information of the target service of the virtual resource type is merged when the user to be distributed in the user group uses the virtual resource type again in different preset time periods to obtain the virtual resource distribution gain of the user group under the virtual resource type.

[0056] In one possible implementation, the second determining module, when determining the conversion rate increment of each user group under each virtual resource type based on the usage conversion rate of the users to be distributed in each user group under each virtual resource type and the user ratio corresponding to the reference group, is configured to:

[0057] For each user group, the average usage conversion rate of the user group under each virtual resource type is determined based on the usage conversion rate of each user to be distributed in the user group under each virtual resource type;

[0058] The difference between the average usage conversion rate of each user group under each virtual resource type and the user ratio corresponding to the reference group is determined as the conversion rate increment of each user group under each virtual resource type.

[0059] In one possible implementation, the second determining module, when determining the type of virtual resource to be distributed for each user group based on the principle of maximizing virtual resource distribution gain, the virtual resource distribution gain, the conversion rate increment, and the resource quantity threshold of the virtual resource, is configured to:

[0060] Based on the principle of maximizing virtual resource distribution gain, the virtual resource type to be verified for each user group is determined according to the virtual resource distribution gain, the conversion rate increment, and the virtual resource quantity threshold.

[0061] Determine the second reuse information of the target services corresponding to different virtual resource types used by users in the reference group;

[0062] Based on the second reuse information, determine the reference resource distribution gain corresponding to different virtual resource types for the reference group;

[0063] For each virtual resource type to be verified, if the virtual resource distribution gain of the user group corresponding to the virtual resource type to be verified is greater than the reference resource distribution gain of the reference group under the virtual resource type to be verified, then the virtual resource type to be verified shall be regarded as the virtual resource type to be distributed for the corresponding user group.

[0064] Thirdly, an optional implementation of this disclosure also provides a computer device, a processor, and a memory, wherein 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. When the machine-readable instructions are executed by the processor, the steps of the first aspect above, or any possible implementation of the first aspect, are performed.

[0065] Fourthly, an optional implementation of this disclosure also provides a computer-readable storage medium storing a computer program that, when run, performs the steps of the first aspect or any possible implementation of the first aspect.

[0066] For a description of the effects of the aforementioned resource distribution device, computer equipment, and computer-readable storage medium, please refer to the description of the resource distribution method above; it will not be repeated here.

[0067] The resource distribution method, apparatus, computer equipment, and storage medium provided in this disclosure utilize a trained gain network model to accurately determine the usage conversion rate when distributing different types of virtual resources to each user. Based on the determined usage conversion rate, users are grouped, and then the distribution decision for each user group is optimized using a resource quantity threshold and the user ratio of each target service corresponding to a reference group. This allows for the accurate determination of the type of virtual resource to be distributed to each user group while constraining the amount of resources to be distributed. This enables targeted distribution of virtual resources to users in each user group, thereby achieving rational utilization of virtual resources, improving the utilization efficiency of virtual resources, and ultimately increasing the utilization rate of related service resources.

[0068] To make the above-mentioned objects, features and advantages of this disclosure more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0069] To more clearly illustrate the technical solutions of the embodiments of this disclosure, the accompanying drawings used in the embodiments will be briefly described below. These drawings are incorporated in and constitute a part of this specification. They illustrate embodiments conforming to this disclosure and, together with the specification, serve to explain the technical solutions of this disclosure. It should be understood that the following drawings only show some embodiments of this disclosure and should not be considered as limiting the scope. Those skilled in the art can obtain other related drawings based on these drawings without creative effort.

[0070] Figure 1 A flowchart of a resource distribution method provided by an embodiment of this disclosure is shown;

[0071] Figure 2 This illustration shows a schematic diagram of a specific implementation process for resource distribution provided in an embodiment of this disclosure;

[0072] Figure 3 A schematic diagram of a resource distribution apparatus provided in an embodiment of this disclosure is shown;

[0073] Figure 4 A schematic diagram of the structure of a computer device provided in an embodiment of this disclosure is shown. Detailed Implementation

[0074] To make the objectives, technical solutions, and advantages of the embodiments of this disclosure clearer, the technical solutions of the embodiments of this disclosure will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this disclosure, and not all of them. The components of the embodiments of this disclosure described and shown herein can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this disclosure is not intended to limit the scope of the claimed disclosure, but merely represents selected embodiments of this disclosure. All other embodiments obtained by those skilled in the art based on the embodiments of this disclosure without inventive effort are within the scope of protection of this disclosure.

[0075] Furthermore, the terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this disclosure are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in a sequence other than that illustrated or described herein.

[0076] In this article, "multiple or several" refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone. The character " / " generally indicates that the preceding and following related objects have an "or" relationship.

[0077] Research has found that distributing virtual resources corresponding to target services to users is a common method to improve the utilization rate of those services. The accuracy and relevance of the virtual resource distribution directly impacts the effectiveness of this improvement. However, conventional virtual resource distribution methods often involve directly distributing virtual resources of various types to users, allowing them to choose one. This wastes unused virtual resources, reducing the rationality and relevance of virtual resource distribution. Therefore, how to specifically distribute virtual resources of various target types to different users has become a key technical challenge.

[0078] Based on the above research, this disclosure provides a resource distribution scheme. Utilizing a trained gain network model, the usage conversion rate when distributing different types of virtual resources to each user can be accurately determined. Users are grouped according to the determined usage conversion rate. Then, by using a resource quantity threshold and the ratio of users using each target service in the reference group, the distribution decision for each user group is optimized. This allows for the accurate determination of the type of virtual resource to be distributed to each user group while constraining the amount of resources to be distributed. This enables targeted distribution of virtual resources to users in each user group, thereby achieving rational utilization of virtual resources, improving the efficiency of virtual resource utilization, and ultimately increasing the utilization rate of related service resources.

[0079] The shortcomings of the above solutions are the result of the inventor's practical experience and careful research. Therefore, the discovery process of the above problems and the solutions proposed in this disclosure below should be considered as the inventor's contribution to this disclosure.

[0080] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0081] To facilitate understanding of this embodiment, a resource distribution method disclosed in this disclosure will first be described in detail. The execution subject of the resource distribution method provided in this disclosure is generally a terminal device or other processing device with certain computing power. The terminal device can be a user equipment (UE), a mobile device, a user terminal, a terminal, a personal digital assistant (PDA), a handheld device, a computer device, etc. In some possible implementations, the resource distribution method can be implemented by the processor calling computer-readable instructions stored in the memory.

[0082] The resource distribution method provided in this disclosure embodiment will be described below using a server as the execution subject as an example.

[0083] like Figure 1 The flowchart shown is a resource distribution method provided in an embodiment of this disclosure, which may include the following steps:

[0084] S101: Using the trained target network model, determine the usage conversion rate when distributing different types of virtual resources to each user to be distributed; use the conversion rate to indicate the probability that the user to be distributed will use the target service after being distributed virtual resources.

[0085] Here, the target service can be any type of online service, such as an app store service, a game service, or an online shopping service. Virtual resources can be, for example, app acceleration coupons, discount coupons, or virtual coins. One type of virtual resource can correspond to at least one target service, and one target service can correspond to at least one type of virtual resource. 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 both target service A and target service C.

[0086] The target network model is a pre-trained network model that, based on the input virtual resource type and the attribute information authorized by the user to be distributed, determines the conversion rate of a user's use of the target service after distributing virtual resources of that type to the user. The conversion rate indicates the probability that a user will use the target service after the virtual resource is distributed to them. User attribute information may include, for example, user authorization behavior information, feature information, and usage records of the target service.

[0087] In one embodiment, the trained target network model can be determined according to the following steps:

[0088] Step 1: Collect sample data according to the preset time period, and use the collected sample data to iteratively train the gain network model to be trained, so as to obtain the trained gain network model.

[0089] Here, the preset time period can be set according to actual application needs, and this embodiment does not impose specific limitations. For example, the preset time period can be small enough to achieve real-time sample data collection. Of course, the preset time period can also be set to, for example, 1 hour, 12 hours, 1 day, 2 days, etc.

[0090] Sample data may include virtual resource types and user attribute information corresponding to the authorized use of sample users.

[0091] In practice, a small-scale random exploratory experiment can be used to randomly select sample users and randomly send different types of virtual resources to them. The conversion rate of each sample user can then be obtained. In this way, random exploratory experiments can be used to continuously collect sample data.

[0092] After obtaining the sample data, the virtual resource type and the user attribute information of the sample users can be concatenated. This concatenated data is then used as input to the uplift network model to be trained. The uplift network model then processes the input data, outputting the predicted usage conversion rate for each sample user. Using the sample usage conversion rate and the predicted usage conversion rate, the prediction loss is determined. This prediction loss is then used to iteratively train the uplift network model until a preset training cutoff condition is met, resulting in a trained uplift network model. This uplift network model is the target network model to be trained.

[0093] The preset training cutoff conditions may include the model prediction accuracy reaching a preset accuracy and / or the number of iterations of training reaching a preset number of iterations.

[0094] Step 2: Determine the model evaluation metric of the trained gain network model, and if the model evaluation metric is greater than the model evaluation metric of the validation gain model, use the trained gain network model as the trained target network model.

[0095] Among them, the validation gain model is used to verify whether the trained gain network model meets the usage requirements. The validation gain model is trained using sample data and has a different network structure from the gain network model.

[0096] The Area Under Uplift Curve (AUUC) is used to characterize the performance of a model. A validation gain model is used to validate the performance of a trained gain network model. It uses the same training data as the gain network model to be trained, but has a different network structure.

[0097] The requirement for use is that the evaluation metric of the trained gain network model is greater than that of the validation gain model. If the requirement is met, it can be said that the trained gain network model can be used online and can accurately predict the conversion rate of users to be distributed during online use.

[0098] In practice, the AUUC of the trained gain network model can be compared with the AUUC of the validation gain model. If the AUUC of the trained gain network model is greater than the AUUC of the validation gain model, the trained gain network model can be used as the target network model and deployed online.

[0099] Conversely, if the AUUC of the trained gain network model is not greater than the AUUC of the validation gain model, the newly collected sample data can be used to continue training the gain network model until the trained target network model is obtained and put into online use.

[0100] In this way, by using continuously collected sample data to train the gain network model and the validation gain model, we can avoid using different sample data to train the gain network model and the validation gain model. This avoids the inability to determine whether the virtual resource distribution gain brought by the trained model is caused by different sample data or by a problem with the model itself, thus ensuring that the virtual resource distribution gain brought by the trained model is brought by the model itself.

[0101] In specific implementation, S101 can concatenate the authorized user attribute information of the user to be distributed and the virtual resource type of the virtual resource to be distributed. The concatenated data is then used as input to the target network model. This allows the target network model to output the usage conversion rate when distributing that type of virtual resource to the user to be distributed. Based on this, by concatenating the authorized user attribute information of different users to be distributed and different virtual resource types, and inputting this data into the target network model, the usage conversion rate when distributing different types of virtual resources to each user to be distributed can be obtained.

[0102] S102: Based on the usage conversion rate of each user to be distributed, divide each user to be distributed into different user groups.

[0103] Here, a user group may include at least one user to be distributed.

[0104] For example, the number of user groups to be divided can be determined based on the number of users to be distributed to, and then each user to be distributed can be evenly divided into different user groups based on the usage conversion rate of each user and the number of user groups.

[0105] Taking 100 users to be distributed as an example, we can first determine that the number of user groups to be divided is 10. Then, based on the usage conversion rate of each of the 100 users to be distributed, we can divide the 100 users to be distributed into 10 user groups.

[0106] In one embodiment, for S102, the users to be distributed can also be divided into different user groups in either of the following two ways:

[0107] Method 1: Divide each user to be distributed into different user groups based on the usage conversion rate of each user under any virtual resource type.

[0108] In practice, any virtual resource type can be selected from a variety of virtual resource types as the allocation resource type. Then, based on the usage conversion rate of each user to be allocated under the allocation resource type and the preset conversion rate range for each user group, each user to be allocated can be divided into different user groups.

[0109] For example, virtual resource type A is used as the allocation resource type. Based on the usage conversion rate of each user to be allocated under virtual resource type A, the conversion rate interval of each user to be allocated is determined. Then, the users to be allocated can be assigned to the user groups associated with the corresponding conversion rate interval. Here, one conversion rate interval is associated with one user group.

[0110] Method 2: Based on the usage conversion rate of each user to be distributed under each virtual resource type, determine the average conversion rate of each user to be distributed, and use the average conversion rate of each user to be distributed to divide each user to be distributed into different user groups.

[0111] For example, the conversion rate range for each user to be distributed can be determined based on the average conversion rate for each user to be distributed. Then, the users to be distributed can be assigned to user groups associated with the conversion rate range.

[0112] S103: Based on the usage conversion rate of users to be distributed in each user group under each virtual resource type, the resource quantity threshold of virtual resources, and the ratio of users using each target service in the reference group, determine the virtual resource type to be distributed for each user group; users in the reference group refer to users who have not been allocated virtual resources.

[0113] Here, the users in the reference group are all users who have not been assigned any type of virtual resource, and the number of users in the reference group can be the same as the number of users to be assigned.

[0114] A user group corresponds to a virtual resource type to be distributed. The virtual resources of the virtual resource type to be distributed are the most suitable virtual resources to be sent to users in the user group corresponding to that virtual resource type.

[0115] The virtual resource quantity threshold indicates the maximum amount of resources that can be distributed. The user ratio for each target service in the reference group represents the ratio of the number of users using the target service to the number of users not using the target service in the reference group.

[0116] For example, a virtual resource quantity threshold can be used as a constraint. While ensuring that the total amount of virtual resources sent to each user group does not exceed the resource quantity threshold, the objective is to maximize the difference between the usage conversion rate of the users to be distributed in each virtual resource type in the user group and the user ratio of the corresponding reference group. Based on the usage conversion rate of the users to be distributed in each virtual resource type in each user group and the user ratio of the corresponding reference group for each target service, the virtual resource type to be distributed for each user group can be determined.

[0117] In one embodiment, S103 can be implemented according to the following steps:

[0118] S103-1: For each user group, determine the first reuse information of the target service corresponding to the different virtual resource types when the user to be distributed uses different types of virtual resources again after distributing different types of virtual resources to the users to be distributed in the user group.

[0119] Here, the first reuse information includes, for example, the number of times the target service can be reused, the frequency of reuse, and the duration of reuse. For instance, if the virtual resource is a coupon, the first reuse information could include the amount of money spent on purchasing the item corresponding to the coupon, the repurchase rate, etc. If the virtual resource is a skin (or outfit) of a virtual object in a game, the first reuse information could include the duration of playing the game again, the number of times, frequency, and duration of playing the game again using the skin (or outfit), and the benefits gained from playing the game again using the skin (or outfit), etc.

[0120] For example, in a user group comprising user groups 1 to 3, virtual resource types comprising virtual resource types A to C, and target services comprising target service A corresponding to virtual resource type A, target service B corresponding to virtual resource type B, and target service C corresponding to virtual resource type C, for user group 1, it can be determined that after distributing virtual resource type A to the users to be distributed in user group 1, the users to be distributed will reuse the first reuse information of target service A again; after distributing virtual resource type B to the users to be distributed in user group 1, the users to be distributed will reuse the first reuse information of target service B again; and after distributing virtual resource type C to the users to be distributed in user group 1, the users to be distributed will reuse the first reuse information of target service C again. Similarly, for user groups 2 and 3, it can be determined that after distributing different types of virtual resources to the users to be distributed in user groups, the users to be distributed will reuse the first reuse information of the target service corresponding to different virtual resource types.

[0121] S103-2: Based on the first multiplexing information of the user group, determine the virtual resource distribution gain corresponding to different virtual resource types for the user group.

[0122] Here, for a user group, there can be a corresponding virtual resource distribution gain for each virtual resource type. For example, for user group 1 mentioned above, its corresponding virtual resource distribution gain may include: virtual resource distribution gain 1 under virtual resource type A, virtual resource distribution gain 2 under virtual resource type B, and virtual resource distribution gain 3 under virtual resource type C. The virtual resource distribution gain is used to characterize the gain brought to the users in the user group by the users using the target service after the virtual resources are distributed to them.

[0123] In practice, for each user group, the virtual resource distribution gain of that user group under different virtual resource types can be determined according to the preset gain calculation formula and the first reuse information corresponding to that user group under different virtual resource types.

[0124] In one embodiment, for S103-1 above, for each user group, it can be determined that after distributing different types of virtual resources to the users to be distributed in the user group, the users to be distributed will use the target service corresponding to different virtual resource types again within different preset time periods.

[0125] Here, the number and duration of the preset time periods can be set according to actual application needs, and this disclosed example does not impose specific limitations. For example, the preset time periods may include the first N days (e.g., 30 days) after different types of virtual resources are distributed to the users in the user group, or they may include the M to M+10 days (e.g., the 21st to 30th days) after different types of virtual resources are distributed to the users in the user group.

[0126] Taking a user group consisting of user group 1 and user group 2, virtual resource types including virtual resource types A and B, and target services including target service A corresponding to virtual resource type A and target service B corresponding to virtual resource type B as an example, for user group 1, we can determine the gain of users in user group 1 using target service A within the first 30 days after distributing virtual resources corresponding to virtual resource type A to the users to be distributed, and the frequency of users in user group 1 using target service A from day 21 to day 30. Similarly, we can determine the gain of users in user group 1 using target service B within the first 30 days after distributing virtual resources corresponding to virtual resource type B to the users to be distributed, and the frequency of users in user group 1 using target service B from day 21 to day 30.

[0127] Similarly, for user group 2, we can determine the gain of users in user group 2 using target service A within the first 30 days after distributing virtual resources of type A to the users to be distributed in user group 2, and the frequency of users in user group 2 using target service A from day 21 to day 30. Likewise, we can determine the gain of users in user group 2 using target service B within the first 30 days after distributing virtual resources of type B to the users to be distributed in user group 2, and the frequency of users in user group 2 using target service B from day 21 to day 30.

[0128] Furthermore, for S103-2, for each virtual resource type, the first reuse information of the target service of the virtual resource type can be merged when the users to be distributed in the user group use the virtual resource type again in different preset time periods to obtain the virtual resource distribution gain of the user group under the virtual resource type.

[0129] For example, the virtual resource distribution gain for a user group under a virtual resource type can be determined according to the following formula:

[0130] LTV ij =(RPG30) ij + Frequency of use of the target service from day 21 to 30 * Average value of the target service from day 21 to 30 * (a1 * a2 - a3)) * a4, (Formula 1)

[0131] Among them, LTV ij RPG30 represents the virtual resource distribution gain for user group i under the j-th virtual resource type. ij This represents the gain of the target service corresponding to the j-th virtual resource type for the users in user group i who are to be assigned the service within the first 30 days. The average value is used to characterize the value cost of the target service. a1, a2, a3, and a4 are different pre-set values.

[0132] In practice, based on the first reuse information corresponding to each user group under different virtual resource types and Formula 1 above, the virtual resource distribution gain of each user group under different virtual resource types can be determined respectively.

[0133] S103-3: Based on the conversion rate of users to be distributed in each user group under each virtual resource type and the user ratio of the reference group, determine the conversion rate increment of each user group under each virtual resource type.

[0134] Here, the conversion rate increment is used to characterize the difference between the usage conversion rate and the user ratio. Taking the determination of the conversion rate increment for any user group under any virtual resource type as an example, we can determine the difference between the usage conversion rate of each user to be distributed in the user group under that virtual resource type and the user ratio corresponding to the reference group. Then, the average of the differences corresponding to each user to be distributed can be used as the conversion rate increment of the user group under that virtual resource type.

[0135] In one embodiment, S103-3 can be implemented according to the following steps:

[0136] S103-3-1: For each user group, determine the average usage conversion rate of the user group under each virtual resource type based on the usage conversion rate of each user to be distributed in the user group under each virtual resource type.

[0137] For example, for user group 1, which includes users 1 to 3 to be distributed, the average usage conversion rate of user group 1 under virtual resource type A can be determined based on the usage conversion rates of users 1 to 3 under virtual resource type A. The average usage conversion rate of user group 1 under virtual resource type B can be determined based on the usage conversion rates of users 1 to 3 under virtual resource type B. Finally, the average usage conversion rate of user group 1 under virtual resource type C can be determined based on the usage conversion rates of users 1 to 3 under virtual resource type C.

[0138] S103-3-2: The difference between the average usage conversion rate of each user group under each virtual resource type and the user ratio of the corresponding reference group is determined as the conversion rate increment of each user group under each virtual resource type.

[0139] For example, the difference between the average conversion rate of user group 1 under virtual resource type A and the user ratio of the corresponding reference group can be used as the conversion rate increment of user group 1 under virtual resource type A; the difference between the average conversion rate of user group 1 under virtual resource type B and the user ratio of the corresponding reference group can be used as the conversion rate increment of user group 1 under virtual resource type B; and the difference between the average conversion rate of user group 1 under virtual resource type C and the user ratio of the corresponding reference group can be used as the conversion rate increment of user group 1 under virtual resource type C.

[0140] S103-4: Based on the principle of maximizing virtual resource distribution gain, determine the type of virtual resource to be distributed for each user group according to the virtual resource distribution gain, conversion rate increment, and virtual resource quantity threshold.

[0141] For example, the type of virtual resource to be distributed for each user group can be determined according to the following formula:

[0142] max∑ i，j ltv ij *Δr ij *p ij (Formula 2)

[0143] st∑ j p ij p ij ∈(0,1), (Formula 3)

[0144] ∑ i (c ij +Δr ij )p ij ≤thres2*ost base (Formula 4)

[0145] Where max represents maximization, i represents the i-th user group, j represents the j-th virtual resource type, and ltv ij Δr represents the virtual resource distribution gain for user group i under virtual resource type j. ij This represents the conversion rate increment (p) for user group i under virtual resource type j. ij Let st represent the probability of distributing the j-th virtual resource type to user group i, where st represents the constraint ..., and c ij This represents the amount of resources needed to distribute virtual resources of the j-th virtual resource type to user group i, where thres2*cost is the maximum amount of resources required.base This represents the threshold for the amount of virtual resources.

[0146] In practical implementation, the virtual resource distribution gain, conversion rate increment, and virtual resource quantity threshold can be substituted into the above formulas, and the type of virtual resource to be distributed for each user group can be determined by combining the above formulas. For example, by combining the above formulas, the p value for each user group under each virtual resource type can be determined. ij Then, based on the maximum p value for each user group under each virtual resource type, ij The indicated virtual resource type serves as the virtual resource type to be distributed to this user group.

[0147] In one implementation, after determining the virtual resource types to be distributed for each user group, an association between the user groups and the virtual resource types to be distributed can be established and stored. For example, this storage can be performed using a distributed transaction (try-confirm-cancel, TCC) approach.

[0148] In one embodiment, S103-4 described above can be implemented according to the following steps:

[0149] S103-4-1: Based on the principle of maximizing virtual resource distribution gain, determine the type of virtual resource to be verified for each user group according to the virtual resource distribution gain, conversion rate increment, and virtual resource quantity threshold.

[0150] In practice, the virtual resource type to be verified for each user group can be determined according to Formulas 2 to 4 above. Then, the virtual resource type to be verified for each user group can be verified according to the following steps. If the verification is successful, the virtual resource type to be verified for each user group can be used as the virtual resource type to be distributed for each user group.

[0151] S103-4-2: Determine the second reuse information for users in the reference group who use target services corresponding to different virtual resource types.

[0152] In practice, the starting point can be the time point when different types of virtual resources are distributed to each user to be distributed. The second reuse information can be used, which is the time point when users in the reference group use the target services corresponding to different types of virtual resources in different preset time periods.

[0153] S103-4-3: Based on the second reuse information, determine the reference resource distribution gain corresponding to different virtual resource types in the reference group.

[0154] For example, the reference resource distribution gain of the reference group under the virtual resource type can be determined according to the following formula five:

[0155]

[0156] in, The reference resource distribution gain of the reference group under the j-th virtual resource type is shown. This represents the gain of users in the reference group using the target service corresponding to the j-th virtual resource type within the first 30 days. The average value is used to characterize the value cost of the target service. a1, a2, a3, and a4 are different pre-set values.

[0157] In practice, the reference resource distribution gain for different virtual resource types is determined based on the second reuse information corresponding to the reference group under different virtual resource types and Formula 4 above.

[0158] S103-4-4: For each virtual resource type to be verified, if the virtual resource distribution gain of the user group corresponding to the virtual resource type to be verified is greater than the reference resource distribution gain of the reference group under the virtual resource type to be verified, then the virtual resource type to be verified shall be used as the virtual resource type to be distributed for the corresponding user group.

[0159] For example, for each virtual resource type to be verified, if there is only one user group corresponding to the virtual resource type to be verified, the virtual resource distribution gain of that unique user group under the virtual resource type to be verified can be compared with the reference resource distribution gain of the reference group under the virtual resource type to be verified. If the virtual resource distribution gain is greater than the reference resource distribution gain, the virtual resource type to be verified can be used as the virtual resource type to be distributed for that unique user group. Conversely, if the virtual resource distribution gain is less than the reference resource distribution gain, it indicates that the currently determined virtual resource type to be verified is unreasonable, and the process can return to execute S103-4-1 above to redetermine the virtual resource types to be verified for each user group until the virtual resource distribution gain is greater than the reference resource distribution gain.

[0160] For each virtual resource type to be verified, if there are only a few user groups corresponding to the virtual resource type to be verified, for example, if the user groups corresponding to the virtual resource type A to be verified include user group 4 and user group 5, then the sum of the virtual resource distribution gains can be determined based on the virtual resource distribution gains of user group 4 and user group 5 under the virtual resource type to be verified. If the sum is greater than the reference resource distribution gain corresponding to the reference group under the virtual resource type to be verified, then the virtual resource type A to be verified can be used as the virtual resource type to be distributed for user group 4 and user group 5 respectively.

[0161] In this way, by utilizing the reference resource distribution gain corresponding to the reference group, the rationality of the virtual resource types to be verified for each user group can be verified, thus obtaining reasonable and accurate virtual resource types to be distributed.

[0162] S104: Distribute virtual resources to each user in the user group according to the type of virtual resource to be distributed.

[0163] In practice, for each user group, the virtual resources of the corresponding virtual resource type to be distributed can be distributed to each user in that user group. For example, if the virtual resource type to be distributed for user group 1 is virtual resource type B, and the virtual resource type to be distributed for user group 2 is virtual resource type A, then virtual resources of virtual resource type B can be distributed to each user in user group 1, and virtual resources of virtual resource type A can be distributed to each user in user group 2.

[0164] Optionally, after determining the usage conversion rate when distributing different types of virtual resources to each user using the target neural network, the usage conversion rates corresponding to each user can be stored. For example, they can be stored in the distributed storage space Abase. Further, after storing the association between user groups and the types of virtual resources to be distributed, if other servers have a need to distribute virtual resources to users, they can first determine the usage conversion rate of the user from Abase based on the user's identifier; then, based on the usage conversion rate, determine the user group to which the user belongs; and based on the association corresponding to the user group, determine the type of virtual resource to be distributed; finally, the virtual resource of that type can be sent to the user.

[0165] Based on the above embodiments, by utilizing the trained gain network model, the usage conversion rate when distributing different types of virtual resources to each user can be accurately determined. Users are grouped according to the determined usage conversion rate, and then the distribution decision for each user group is optimized using a resource quantity threshold and the user ratio of each target service corresponding to a reference group. This allows for the accurate determination of the type of virtual resource to be distributed to each user group while constraining the amount of resources to be distributed. This enables targeted distribution of virtual resources to users in each user group, thereby achieving rational utilization of virtual resources, improving the efficiency of virtual resource utilization, and ultimately increasing the utilization rate of related service resources.

[0166] In one embodiment, such as Figure 2The diagram illustrates a specific implementation process for resource distribution according to an embodiment of this disclosure. It can include three parts: a first part is a model update part (including real-time updates and / or daily updates); a second part is a decision optimization part (which can be optimized in real-time and / or daily); and a third part is an online service part, used to distribute virtual resources online to each user to be distributed. Specifically, the model update part can include: continuously collecting sample data using a low-volume random exploratory experiment; wherein, the sample data includes the virtual resource type and the user attribute information corresponding to the authorized use of the sample user. The virtual resource type and the user attribute information corresponding to the authorized use of the sample user are concatenated, and the concatenated data is used as input to the uplift network model to be trained. The uplift network model to be trained is iteratively trained to obtain the target network model.

[0167] The decision optimization section may include: using a target network model to determine the usage conversion rate when distributing different types of virtual resources to each user, and storing the usage conversion rate corresponding to each user in Abase. Using a pre-trained gain information prediction model, determine the virtual resource distribution gain for each user after distributing different types of virtual resources, and store it in Abase. The gain information prediction model can also be trained using sample data. Then, based on the usage conversion rate corresponding to each user, the users to be distributed are divided into different user groups. For each user group, determine the first reuse information of the target service corresponding to different virtual resource types after distributing different types of virtual resources to the users in that user group. Afterwards, decision optimization can be performed, i.e., performing S103 above, to determine the virtual resource type to be distributed for each user group. The association between user groups and virtual resource types to be distributed is established and written into the TCC. The virtual resource distribution gain output by the gain information prediction model can be used to verify whether the optimization result obtained by the decision optimization part (i.e., the type of virtual resource to be distributed corresponding to the user group) is reasonable. Specifically, if the virtual resource distribution gain after determining that the virtual resource type to be distributed to the user to be distributed is greater than the virtual resource distribution gain output by the gain information prediction model, the optimization result is determined to be reasonable.

[0168] The online service component may include: If any upstream service determines that virtual resources need to be distributed to users, the frisbee (server) can retrieve the usage conversion rate corresponding to that user from Abase based on the user identifier, and determine the user group corresponding to that user based on the usage conversion rate. It then retrieves the association relationship from TCC and determines the type of virtual resource to be distributed to that user based on the association relationship. Finally, it can distribute that type of virtual resource to the user.

[0169] Those skilled in the art will understand that, in the above-described method of the specific implementation, the order in which each step is written does not imply a strict execution order and does not constitute any limitation on the implementation process. The specific execution order of each step should be determined by its function and possible internal logic.

[0170] Based on the same inventive concept, this disclosure also provides a resource distribution device corresponding to the resource distribution method. Since the principle of the device in this disclosure for solving the problem is similar to that of the resource distribution method described above, the implementation of the device can refer to the implementation of the method, and the repeated parts will not be described again.

[0171] like Figure 3 The diagram shown is a schematic representation of a resource distribution device provided in an embodiment of this disclosure, comprising:

[0172] The first determining module 301 is used to determine the usage conversion rate when distributing different types of virtual resources to each user to be distributed using a trained target network model; the usage conversion rate is used to indicate the probability that the user to be distributed will use the target service after being distributed the virtual resources.

[0173] The segmentation module 302 is used to divide each user to be distributed into different user groups according to the usage conversion rate corresponding to each user to be distributed;

[0174] The second determining module 303 is used to determine the virtual resource type to be distributed for each user group based on the usage conversion rate of the users to be distributed in each user group under each virtual resource type, the resource quantity threshold of the virtual resource, and the user ratio of each target service in the reference group; the users in the reference group refer to users who have not been allocated virtual resources.

[0175] The distribution module 304 is used to distribute virtual resources to each user in the user group according to the type of virtual resource to be distributed corresponding to each user group.

[0176] In one possible implementation, the segmentation module 302, when dividing the users to be distributed into different user groups according to the usage conversion rate corresponding to each user, is used to:

[0177] Based on the usage conversion rate of each user to be distributed under any of the virtual resource types, the users to be distributed are divided into different user groups; or,

[0178] Based on the usage conversion rate of each user to be distributed under each virtual resource type, the average conversion rate of each user to be distributed is determined, and the average conversion rate of each user to be distributed is used to divide each user to be distributed into different user groups.

[0179] In one possible implementation, the device further includes:

[0180] Training module 305 is used to determine the trained target network model according to the following steps:

[0181] According to the preset time period, sample data is collected, and the collected sample data is used to iteratively train the gain network model to be trained, so as to obtain the trained gain network model.

[0182] Determine the model evaluation index of the trained gain network model, and if the model evaluation index is greater than the model evaluation index of the validation gain model, use the trained gain network model as the trained target network model.

[0183] The verification gain model is used to verify whether the trained gain network model meets the usage requirements. The verification gain model is trained using the sample data and has a different network structure from the gain network model.

[0184] In one possible implementation, the second determining module 303, when determining the virtual resource type to be distributed for each user group based on the usage conversion rate of the users to be distributed in each user group under each virtual resource type, the resource quantity threshold of the virtual resource, and the user ratio of the reference group corresponding to each target service, is configured to:

[0185] For each user group, it is determined that after distributing different types of virtual resources to the users to be distributed in that user group, the users to be distributed will use the target service corresponding to the different virtual resource types again.

[0186] Based on the first reuse information of the user group, determine the virtual resource distribution gain corresponding to different virtual resource types for the user group;

[0187] Based on the usage conversion rate of the users to be distributed in each user group under each virtual resource type and the user ratio corresponding to the reference group, determine the conversion rate increment of each user group under each virtual resource type;

[0188] Based on the principle of maximizing virtual resource distribution gain, the type of virtual resource to be distributed for each user group is determined according to the virtual resource distribution gain, the conversion rate increment, and the resource quantity threshold of the virtual resource.

[0189] In one possible implementation, the second determining module 303, when determining the first reuse information of the target service corresponding to the different virtual resource types after distributing different types of virtual resources to the users to be distributed in the user group, is used to:

[0190] After determining that different types of virtual resources are distributed to the users to be distributed in the user group, the first reuse information of the target service corresponding to different virtual resource types is used again by the users to be distributed in different preset time periods;

[0191] The step of determining the virtual resource distribution gain for different virtual resource types corresponding to the user group based on the first multiplexing information of the user group includes:

[0192] For each virtual resource type, the first reuse information of the target service of the virtual resource type is merged when the user to be distributed in the user group uses the virtual resource type again in different preset time periods to obtain the virtual resource distribution gain of the user group under the virtual resource type.

[0193] In one possible implementation, the second determining module 303, when determining the conversion rate increment of each user group under each virtual resource type based on the usage conversion rate of the users to be distributed in each user group under each virtual resource type and the user ratio corresponding to the reference group, is configured to:

[0194] For each user group, the average usage conversion rate of the user group under each virtual resource type is determined based on the usage conversion rate of each user to be distributed in the user group under each virtual resource type;

[0195] The difference between the average usage conversion rate of each user group under each virtual resource type and the user ratio corresponding to the reference group is determined as the conversion rate increment of each user group under each virtual resource type.

[0196] In one possible implementation, the second determining module 303, when determining the type of virtual resource to be distributed for each user group according to the principle of maximizing virtual resource distribution gain, based on the virtual resource distribution gain, the conversion rate increment, and the resource quantity threshold of the virtual resource, is configured to:

[0197] Based on the principle of maximizing virtual resource distribution gain, the virtual resource type to be verified for each user group is determined according to the virtual resource distribution gain, the conversion rate increment, and the virtual resource quantity threshold.

[0198] Determine the second reuse information of the target services corresponding to different virtual resource types used by users in the reference group;

[0199] Based on the second reuse information, determine the reference resource distribution gain corresponding to different virtual resource types for the reference group;

[0200] For each virtual resource type to be verified, if the virtual resource distribution gain of the user group corresponding to the virtual resource type to be verified is greater than the reference resource distribution gain of the reference group under the virtual resource type to be verified, then the virtual resource type to be verified shall be regarded as the virtual resource type to be distributed for the corresponding user group.

[0201] The processing flow of each module in the device and the interaction flow between each module can be referred to the relevant descriptions in the above method embodiments, and will not be detailed here.

[0202] Based on the same technical concept, embodiments of this application also provide a computer device. (Refer to...) Figure 4 The diagram shown is a structural schematic of a computer device provided in an embodiment of this application, comprising:

[0203] The system includes a processor 41, a memory 42, and a bus 43. The memory 42 stores machine-readable instructions executable by the processor 41. The processor 41 executes these machine-readable instructions, performing the following steps: S101: Using a trained target network model, determine the usage conversion rate when distributing different types of virtual resources to each user to be assigned; the usage conversion rate indicates the probability that a user will use the target service after being assigned virtual resources; S102: Based on the usage conversion rate corresponding to each user to be assigned, divide each user to be assigned into different user groups; S103: Based on the usage conversion rate of each user to be assigned under each virtual resource type, the resource quantity threshold of the virtual resources, and the ratio of users using each target service in the reference group, determine the type of virtual resource to be distributed for each user group; users in the reference group refer to users who have not been assigned virtual resources; and S104: Distribute virtual resources to each user to be assigned within each user group according to the type of virtual resource to be distributed for that user group.

[0204] The aforementioned memory 42 includes a main memory 421 and an external memory 422. The main memory 421, also known as internal memory, is used to temporarily store the computational data in the processor 41, as well as the data exchanged with external memory such as a hard disk. The processor 41 exchanges data with the external memory 422 through the main memory 421. When the computer device is running, the processor 41 and the memory 42 communicate through the bus 43, so that the processor 41 executes the execution instructions mentioned in the above method embodiments.

[0205] This disclosure also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, performs the steps of the resource distribution method described in the above-described method embodiments. The storage medium may be a volatile or non-volatile computer-readable storage medium.

[0206] The computer program product of the resource distribution method provided in this disclosure includes a computer-readable storage medium storing program code. The instructions included in the program code can be used to execute the steps of the resource distribution method described in the above method embodiments. For details, please refer to the above method embodiments, which will not be repeated here.

[0207] The computer program product can be implemented specifically through hardware, software, or a combination thereof. In one alternative embodiment, the computer program product is specifically embodied in a computer storage medium; in another alternative embodiment, the computer program product is specifically embodied in a software product, such as a software development kit (SDK), etc.

[0208] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems and devices described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here. In the several embodiments provided in this disclosure, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of units is only a logical functional division; in actual implementation, there may be other division methods. Furthermore, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Another point is that the displayed or discussed mutual coupling or direct coupling or communication connection may be through some communication interfaces; the indirect coupling or communication connection of devices or units may be electrical, mechanical, or other forms.

[0209] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0210] In addition, the functional units in the various embodiments of this disclosure can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0211] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a processor-executable, non-volatile, computer-readable storage medium. Based on this understanding, the technical solution of this disclosure, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this disclosure. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0212] If the technical solution of this application involves personal information, the product using this technical solution has clearly informed the user of the personal information processing rules and obtained the user's voluntary consent before processing the personal information. If the technical solution of this application involves sensitive personal information, the product using this technical solution has obtained the user's separate consent before processing the sensitive personal information, and also meets the requirement of "express consent". For example, at personal information collection devices such as cameras, clear and prominent signs are set up to inform users that they have entered the scope of personal information collection and that personal information will be collected. If an individual voluntarily enters the collection scope, it is deemed that they have agreed to the collection of their personal information; or on the personal information processing device, with clear signs / information informing users of the personal information processing rules, authorization is obtained from the user through pop-up information or by asking the user to upload their personal information; wherein, the personal information processing rules may include information such as the personal information processor, the purpose of personal information processing, the processing method, and the types of personal information processed.

[0213] Finally, it should be noted that the above-described embodiments are merely specific implementations of this disclosure, used to illustrate the technical solutions of this disclosure, and not to limit it. The protection scope of this disclosure is not limited thereto. Although this disclosure has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments, or make equivalent substitutions for some of the technical features, within the scope of the technology disclosed in this disclosure. Such modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this disclosure, and should all be covered within the protection scope of this disclosure. Therefore, the protection scope of this disclosure should be determined by the protection scope of the claims.

Claims

1. A resource distribution method, characterized in that, include: Using a trained target network model, determine the usage conversion rate when distributing different types of virtual resources to each user; The conversion rate is used to indicate the probability that the user to be distributed will use the target service after the virtual resource is distributed; Based on the usage conversion rate corresponding to each of the users to be distributed, the users to be distributed are divided into different user groups; Based on the usage conversion rate of the users to be distributed in each user group under each virtual resource type, the resource quantity threshold of the virtual resource, and the ratio of users using each target service in the reference group, the virtual resource type to be distributed for each user group is determined; the users in the reference group refer to users who have not been allocated virtual resources, wherein one user group corresponds to one virtual resource type to be distributed; Based on the type of virtual resource to be distributed for each user group, virtual resources are distributed to each user within that user group.

2. The method according to claim 1, characterized in that, The step of dividing the users to be distributed into different user groups based on the usage conversion rate corresponding to each user includes: Based on the usage conversion rate of each user to be distributed under any of the virtual resource types, the users to be distributed are divided into different user groups; or, Based on the usage conversion rate of each user to be distributed under each virtual resource type, the average conversion rate of each user to be distributed is determined, and the average conversion rate of each user to be distributed is used to divide each user to be distributed into different user groups.

3. The method according to claim 1, characterized in that, Determine the trained target network model according to the following steps: According to the preset time period, sample data is collected, and the collected sample data is used to iteratively train the gain network model to be trained, so as to obtain the trained gain network model. Determine the model evaluation index of the trained gain network model, and if the model evaluation index is greater than the model evaluation index of the validation gain model, use the trained gain network model as the trained target network model. The verification gain model is used to verify whether the trained gain network model meets the usage requirements. The verification gain model is trained using the sample data and has a different network structure from the gain network model.

4. The method according to claim 1, characterized in that, The step of determining the virtual resource type to be distributed for each user group based on the usage conversion rate of the users to be distributed in each user group under each virtual resource type, the resource quantity threshold of the virtual resource, and the user ratio of the reference group corresponding to each target service includes: For each user group, it is determined that after distributing different types of virtual resources to the users to be distributed in that user group, the users to be distributed will use the target service corresponding to the different virtual resource types again. Based on the first reuse information of the user group, determine the virtual resource distribution gain corresponding to different virtual resource types for the user group; Based on the usage conversion rate of the users to be distributed in each user group under each virtual resource type and the user ratio corresponding to the reference group, determine the conversion rate increment of each user group under each virtual resource type; Based on the principle of maximizing virtual resource distribution gain, the type of virtual resource to be distributed for each user group is determined according to 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, characterized in that, The first reuse information, which determines that after distributing different types of virtual resources to the users to be distributed in the user group, the users to be distributed can reuse the target service corresponding to the different virtual resource types, includes: After determining that different types of virtual resources are distributed to the users to be distributed in the user group, the first reuse information of the target service corresponding to different virtual resource types is used again by the users to be distributed in different preset time periods. The step of determining the virtual resource distribution gain for different virtual resource types corresponding to the user group based on the first multiplexing information of the user group includes: For each virtual resource type, the first reuse information of the target service of the virtual resource type is merged when the user to be distributed in the user group uses the virtual resource type again 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, characterized in that, The step of determining the conversion rate increment of each user group under each virtual resource type based on the usage conversion rate of the users to be distributed in each user group under each virtual resource type and the user ratio corresponding to the reference group includes: For each user group, the average usage conversion rate of the user group under each virtual resource type is determined based on the usage conversion rate of each user to be distributed in the user group under each virtual resource type; The difference between the average usage conversion rate of each user group under each virtual resource type and the user ratio corresponding to the reference group is determined as the conversion rate increment of each user group under each virtual resource type.

7. The method according to claim 4, characterized in that, The process of determining the type of virtual resource to be distributed for each user group, based on the principle of maximizing virtual resource distribution gain, the conversion rate increment, and the virtual resource quantity threshold, includes: Based on the principle of maximizing virtual resource distribution gain, the virtual resource type to be verified for each user group is determined according to the virtual resource distribution gain, the conversion rate increment, and the virtual resource quantity threshold. Determine the second reuse information of the target services corresponding to different virtual resource types used by users in the reference group; Based on the second reuse information, determine the reference resource distribution gain corresponding to different virtual resource types for the reference group; For each virtual resource type to be verified, if the virtual resource distribution gain of the user group corresponding to the virtual resource type to be verified is greater than the reference resource distribution gain of the reference group under the virtual resource type to be verified, then the virtual resource type to be verified shall be regarded as the virtual resource type to be distributed for the corresponding user group.

8. A resource distribution device, characterized in that, include: The first determination module is used to determine the usage conversion rate when distributing different types of virtual resources to each user to be distributed, using a trained target network model. The conversion rate is used to indicate the probability that the user to be distributed will use the target service after the virtual resource is distributed; The segmentation module is used to divide each user to be distributed into different user groups based on the usage conversion rate corresponding to each user to be distributed. The second determining module is used to determine the virtual resource type to be distributed for each user group based on the usage conversion rate of the users to be distributed in each user group under each virtual resource type, the resource quantity threshold of the virtual resource, and the ratio of users using each target service in the reference group; the users in the reference group refer to users who have not been allocated virtual resources, wherein one user group corresponds to one virtual resource type to be distributed; The distribution module is used to distribute virtual resources to each user in the user group according to the type of virtual resource to be distributed corresponding to each user group.

9. A computer device, characterized in that, include: A processor and a memory, the memory storing machine-readable instructions executable by the processor, the processor executing the machine-readable instructions stored in the memory, and when the machine-readable instructions are executed by the processor, the processor performing the steps of the resource distribution method as described in any one of claims 1 to 7.

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

Images