Business resource configuration method and system, computer device and storage medium

Abstract

The application provides a service resource configuration method and system, computer equipment and a storage medium, and relates to the technical field of communication. The method comprises the following steps: obtaining a plurality of preset service resource information; obtaining usage information of a plurality of preset users; calculating the service configuration resource amount of each user selecting different service resource information according to the service resource information and the usage information of the plurality of preset users; obtaining the reference value of each user in the plurality of preset users under different configuration resource amount calculation models; obtaining a Loss function according to the reference value of each user in the plurality of preset users under different configuration resource amount calculation models and the configuration resource amount of each user selecting different services; and performing service resource configuration on the users based on the Loss function. The technical scheme provided by the application can improve user experience and enhance user stickiness by performing service resource configuration on the users according to service resource information and user usage information.

Description

Technical Field

[0001] This invention relates to, but is not limited to, the field of communication technology, and in particular to a service resource configuration method, system, computer device, and computer-readable storage medium. Background Technology

[0002] With the continuous advancement of information technologies such as 5G communication, big data, and artificial intelligence, users' demands for mobile phones are also increasing. At the same time, the choice of mobile phone plans has become an extremely important factor.

[0003] The existing packages are mainly designed based on the experience of operators. Many packages are not scientifically designed and provide a poor user experience for rational consumers, resulting in low user stickiness. Summary of the Invention

[0004] This invention was completed in order to at least partially solve the technical problems of poor user experience and low user stickiness in the prior art.

[0005] According to one aspect of the present invention, a method for configuring business resources is provided, comprising: acquiring a plurality of preset business resource information, wherein the business resource information includes at least: a business type and a configuration resource quantity corresponding to the business type;

[0006] Obtain usage information from multiple preset users;

[0007] Based on multiple business resource information and the usage information of multiple preset users, calculate the configuration resource amount for each user among the multiple preset users who select different business resource information;

[0008] Obtain the baseline value for each user among multiple preset users under different resource allocation calculation models;

[0009] The Loss function is obtained by selecting different service resource information configuration amounts for each user among multiple preset users and the baseline values ​​under different configuration resource amount calculation models.

[0010] Based on the Loss function, business resources are configured for users.

[0011] Optionally, the service type includes a first service type, the service resource information also includes resources configured by the service resource information, and the user's usage information includes at least: the first resource f and the second resource t used by the user within a preset time period;

[0012] When the service type is the first service type, the service configuration resource amount for each user among the multiple preset users who select different service resource information is calculated based on multiple service resource information and user usage information, including:

[0013] Based on the first formula, the service configuration resource amount for each user selecting different service resource information is calculated respectively. The first formula includes:

[0014] cost = P + max(0, (fF)*P) F )+max(0,(tT)*P T );

[0015] Where f represents the first resource used by the user within a preset time period, t represents the second resource used by the user within a preset time period, P represents the service configuration resource within a preset time period in the service resource information selected by the current user, T represents the second resource configured within a preset time period in the service resource information selected by the current user, F represents the first resource configured within a preset time period in the service resource information selected by the current user, and P F Among the business resource information selected by the current user, the configured resource quantity of the unit's first resource, P T The configured resource quantity of the second resource in the business resource information selected for the current user within a unit time period, where the current user is one of the preset users.

[0016] Optionally, the service type may also include a second service type, the service resource information may also include resources configured by the service resource information, and the user's usage information may at least include: a third resource c used by the user within a preset time period;

[0017] When the service type is the second service type, the service configuration resource amount for each user among multiple preset users who select different service resource information is calculated based on service resource information and user usage information, including:

[0018] The user's current service configuration resource quantity is calculated according to the second formula, which includes:

[0019] cost = P + c * P F +max(0,(tT)*P T );

[0020] Where P represents the service configuration resource within a preset time period from the service resource information selected by the current user, T represents the second resource configured within the preset time period from the service resource information selected by the current user, and c represents the third resource used by the user within the preset time period. F Among the business resource information selected by the current user, the configured resource quantity of the unit's first resource, P T The configured resource quantity of the second resource in the business resource information selected for the current user within a unit time period, where the current user is one of the preset users.

[0021] Optionally, the resource quantity calculation model includes a rational model and an irrational model, and the user's usage information also includes the user's current business resource information;

[0022] Obtain the baseline value for each user among multiple preset users under different resource allocation calculation models, including:

[0023] Under the rational model, the value with the lowest configured resource amount among multiple business resource information is used as the benchmark value;

[0024] Under the irrational model, for rational users, the minimum value of the configuration resource amount of the user's current business resource information within a preset time period, and the minimum value among the sum of the minimum configuration resource amount among multiple business resource information and the replacement difference m, are selected as the benchmark value.

[0025] For irrational users, the minimum value among the average configuration resource amount of the user's current business resource information within a preset time period and the sum of the average configuration resource amount and the replacement difference m among multiple business resource information is selected as the benchmark value.

[0026] Optionally, the replacement difference m is determined through the following steps:

[0027] Based on the user's usage information, determine the optimal configuration resource amount for the user's business resources;

[0028] The median of the difference between the current configured resource amount of the user's business resource information and the configured resource amount of the optimal business resource information is used as the replacement difference m.

[0029] Optionally, the Loss function is obtained by selecting different service resource amounts for each of the multiple preset users and the baseline values ​​under different configuration resource amount calculation models, including:

[0030] The Loss function is calculated according to the third formula, which includes:

[0031]

[0032] Among them, v ij v is the negative value of the service configuration resource amount for the i-th user selecting the j-th service resource information. ik This is the negative value of the service configuration resource amount for the i-th user selecting the k-th service resource information, where i is one of multiple preset users, r represents a rational user, irr represents an irrational user, a is a constant, and v basei This is the baseline value for user i under different resource allocation calculation models.

[0033] Optionally, configuring business resources for users based on the Loss function includes:

[0034] Based on the Loss function, tensor vectors are generated using MenuNet, and these vectors are used as coefficients.

[0035] Configure business resources for users based on the coefficients.

[0036] According to another aspect of the present invention, a business resource configuration system is provided, comprising: a first acquisition module configured to acquire multiple preset business resource information, wherein the business resource information includes at least: business type and business configuration resource quantity;

[0037] The second acquisition module is configured to acquire usage information from multiple preset users;

[0038] The first calculation module is configured to calculate the amount of business configuration resources for each user who selects different business resource information, based on business resource information and usage information of multiple preset users.

[0039] The third acquisition module is configured to acquire the baseline value of each user among multiple preset users under different resource quantity calculation models;

[0040] The second calculation module is configured to obtain the Loss function based on the configuration resource amount of different services selected by each of the multiple preset users and the baseline value under the different configuration resource amount calculation models.

[0041] The configuration module is set to configure business resources for users based on the Loss function.

[0042] According to another aspect of the present invention, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and when the processor runs the computer program stored in the memory, the processor executes the aforementioned business resource configuration method.

[0043] According to another aspect of the present invention, a computer-readable storage medium is provided, on which a computer program is stored, wherein when the computer program is executed by a processor, the processor performs the aforementioned business resource configuration method.

[0044] The technical solution provided by this invention may include the following beneficial effects:

[0045] The business resource configuration method, system, computer device, and storage medium provided in this invention obtain multiple preset business resource information and usage information of multiple preset users. Based on the multiple business resource information and the usage information of multiple preset users, the configuration resource amount for each user selecting different business resource information is calculated, and the baseline value for each user under different configuration resource amount calculation models is obtained. A Loss function is obtained based on the configuration resource amount for each user selecting different business resource information and the baseline value under different configuration resource amount calculation models. Based on the Loss function, business resource configuration is performed on the user. This can determine the most suitable business resource configuration for the user based on the business resource information and user usage information, thereby improving user experience and increasing user stickiness.

[0046] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the description, claims, and drawings. Attached Figure Description

[0047] The accompanying drawings are provided to further understand the technical solutions of the present invention and constitute a part of the specification. They are used together with the embodiments of the present invention to explain the technical solutions of the present invention, and do not constitute a limitation on the technical solutions of the present invention.

[0048] Figure 1 A flowchart illustrating a business resource configuration method provided in an embodiment of the present invention;

[0049] Figure 2 This is a schematic diagram illustrating the relationship between other app traffic and targeted traffic in a business resource configuration method provided by an embodiment of the present invention.

[0050] Figure 3 A bar chart showing the cost difference between a user's currently selected package and the optimal package, provided by an embodiment of the present invention, for a business resource allocation method;

[0051] Figure 4 This is a schematic diagram of the structure of a business resource configuration system provided in an embodiment of the present invention;

[0052] Figure 5 This is a schematic diagram of the structure of a computer device provided in an embodiment of the present invention. Detailed Implementation

[0053] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the specific implementation methods of the present invention will be described in detail below with reference to the accompanying drawings. It should be understood that the specific implementation methods described herein are for illustration and explanation only and are not intended to limit the present invention.

[0054] Figure 1 This is a flowchart illustrating a business resource configuration method provided in an embodiment of the present invention, as shown below. Figure 1 As shown, this embodiment provides a service resource configuration method, which includes:

[0055] Step 1: Obtain multiple preset business resource information. Each business resource information includes: business type and the amount of configured resources corresponding to the business type.

[0056] Step 2: Obtain usage information from multiple preset users;

[0057] The preset users are those who meet preset conditions. Usage information may include: the first resource f, the second resource t, and the third resource c used by the user within a preset time period.

[0058] Step 3: Based on the business resource information and the usage information of multiple preset users, calculate the amount of business configuration resources for each user among the multiple preset users who select different business resource information.

[0059] Step 4: Obtain the baseline value for each of the multiple preset users under different resource allocation calculation models. Where i is one of a plurality of preset users;

[0060] Step 5: Based on the configuration resource amount of different business resource information selected by each of the multiple preset users, and the baseline value under different configuration resource amount calculation models. Get the Loss function;

[0061] Step 6: Configure business resources for users based on the Loss function.

[0062] Specifically, business resources can be configured for all preset users, or for selected users.

[0063] Therefore, the business resource configuration method in this embodiment can determine the most suitable business resource configuration for the user based on business resource information and user usage information, thereby improving user experience and increasing user stickiness.

[0064] In one embodiment, the service type includes: a first service type, and the service resource information further includes resources configured by the service resource information.

[0065] When the business type is the first business type, step 3 may include:

[0066] Based on the first formula, the service configuration resource amount for each user selecting different service resource information is calculated respectively. The first formula includes:

[0067] cost = P + max(0, (fF)*P) F )+max(0,(tT)*P T );

[0068] Where f represents the first resource used by the user within a preset time period, t represents the second resource used by the user within a preset time period, P represents the service configuration resource within a preset time period in the service resource information selected by the current user, T represents the second resource configured within a preset time period in the service resource information selected by the current user, F represents the first resource configured within a preset time period in the service resource information selected by the current user, and P F For the currently selected business resource information, P represents the configured resource quantity of the first resource within a unit time period. T The configured resource quantity of the second resource in the business resource information selected by the current user within a unit time period; wherein, the business resource information selected by the current user is one of the multiple business resource information selected by the current user, and the current user is one of the preset users.

[0069] Step 3 also includes repeating the above steps until the calculations for all preset users are completed.

[0070] In one embodiment, the business type further includes: a second business type.

[0071] When the service type is the second service type, step 3 may include: calculating the user's current service configuration resource quantity according to a second formula, wherein the second formula includes:

[0072] cost = P + c * P F +max(0,(tT)*P T );

[0073] Where P represents the service configuration resource within a preset time period from the service resource information selected by the current user, T represents the second resource configured within the preset time period from the service resource information selected by the current user, and c represents the third resource used by the user within the preset time period. F For the currently selected business resource information, P represents the configured resource quantity of the first resource within a unit time period. T The resource allocation for the second resource within a unit time period is selected from the business resource information for the current user. The current user is one of the preset users. The preset time period can be one month.

[0074] Step 3 may also include: repeating the above steps until the calculations for all preset users are completed.

[0075] In one embodiment, the configuration resource quantity calculation model includes a rational model and an irrational model, and the user's usage information also includes the user's current business resource information;

[0076] Step 4 includes:

[0077] Under the rational model, the lowest configured resource amount among multiple business resource information is used as the benchmark value.

[0078] Under the irrational model, users are categorized into rational and irrational users. For rational users, the minimum configured resource amount of the user's current business resource information within a preset time period is selected as the baseline value, along with the sum of the minimum configured resource amount among multiple business resource information values ​​and the replacement difference m. For irrational users, the minimum value among the following is selected as the baseline: the average of the configured resource amount of the user's current business resource information within a preset time period, and the sum of the average of the configured resource amount across multiple business resource information sets and the replacement difference m.

[0079] In one implementation, the user's usage information further includes: the service resource information currently selected by the user; the replacement difference m is determined through the following steps: based on the user's usage information, the configured resource quantity of the user's optimal service resource information is determined; the median of the difference between the configured resource quantity of the user's current service resource information and the configured resource quantity of the optimal service resource information is used as the replacement difference m.

[0080] In one implementation, step 5 may include:

[0081] The Loss function is calculated according to the third formula, which includes:

[0082]

[0083] Among them, V ij V is the negative value of the service configuration resource amount for the i-th user selecting the j-th service resource information. ik This is the negative value of the service configuration resource amount for the i-th user selecting the k-th service resource information, where i is one of several preset users, r represents a rational user, irr represents an irrational user, and a is a constant, representing the baseline value of user i under different configuration resource amount calculation models. This is the baseline value for user i under different resource allocation calculation models.

[0084] The following describes the method of this application in detail with specific application examples.

[0085] In this embodiment, the service resource information can be package information, the service type can be package type, the configured resource quantity can be package fee, and the package type includes a first service type and a second service type. The first service type is a regular package, and the second service type is a Wang Card package.

[0086] In this embodiment, step 1 can be specifically as follows:

[0087] The system retrieves information on multiple pre-set mobile phone plans, including plan type and plan fee. These multiple pre-set plans are from the same mobile carrier, and for example, there may be 4-6 plans.

[0088] In this embodiment, step 3 can be: calculating the cost for each user to choose different packages based on the package information and the user's usage information.

[0089] In one implementation, the package type may include: a standard package. The preset time period is monthly. The package information also includes the resources configured in the package information, namely the data allowance F and the voice call time T included in the package; the configured resource amount per unit of the first resource can be understood as the current operator's unit price per GB of data. The configured resource amount per unit time period of the second resource can be understood as the unit price per minute of voice call.

[0090] Specifically, when the package type is a standard package, the parameters of the first formula involved in step 3 can be: the monthly package fee P in the package information, the data allowance F included in the package, the voice call time T included in the package, and the current operator's unit price per GB of data P. F The unit price for a voice call per minute is P. T .

[0091] In one implementation, the package type further includes: the Wang Card package, wherein the Wang Card package includes free data usage.

[0092] In this embodiment, the first resource f, the second resource t, and the third resource c used by the user within the preset time period may include: the data traffic f used by the user in the current month, the outgoing voice duration t, and the sum c of the data traffic used each day of the current month after deducting the targeted data traffic.

[0093] When the package type is the Wang Card package, the parameters of the second formula involved in step 3 can be specifically: the monthly package fee P in the package information, the data allowance F included in the package, the voice call time T included in the package, and the current operator's unit price P per GB of data. F Price per minute of voice call P T .

[0094] In one embodiment, each user's monthly usage can be represented as a triple (f, t, c), where f represents the total data usage for the month, t represents the duration of outgoing voice calls, and c represents the rounded sum of the data usage used each day of the month after deducting targeted data usage.

[0095]

[0096] Where f i This represents the data usage on day i. This represents the targeted traffic on day i.

[0097] In this embodiment, the different resource configuration calculation model in step 4 is used for billing, and the replacement difference m is the cost value m for changing operators. The service resource information currently selected by the user is the service plan currently selected by the user, and the service configuration resource quantity of the user's current service resource information can be understood as the service plan fee of the user's current selected plan.

[0098] Obtaining the baseline value for each of the multiple preset users under different resource allocation calculation models can include:

[0099] Under the rational model, the lowest cost among multiple service packages offered by the operator is used as the benchmark value.

[0100] In the irrational model, for a rational user r, the minimum cost of the user's currently selected plan within a preset time period, and the minimum sum of the lowest cost of a plan among j plans offered by the operator and the cost m of switching operators, are selected as the benchmark value.

[0101] For irrational users (IRR), the minimum of the average cost of the user's currently selected plan within a preset time period, and the sum of the average cost of the plan across j plans from the operator and the cost of switching operators (m) is selected as the benchmark value.

[0102] Repeat the above steps until you obtain the baseline values ​​for all preset users under different resource allocation calculation models.

[0103] In one embodiment, the cost value m for switching operators is determined by the following steps: determining the cost of the user's optimal plan based on the user's usage information; and using the median difference between the cost of the user's currently selected plan and the cost of the optimal plan as the cost value m for switching operators.

[0104] In this embodiment, step 5 can be:

[0105] v ijSpecifically, this could be the benefit for the i-th user choosing package j. Similarly, v... ik Specifically, this could be the benefit for the i-th user choosing package k, where r represents a rational user, irr represents an irrational user, and a is a constant that can be used as a hyperparameter, which can be set to 4. i is one of several preset users. This is the baseline value for user i under different resource allocation calculation models.

[0106] In this embodiment, configuring service resources for the i-th user in step 6 can be to generate a package that matches the usage information of the i-th user.

[0107] Step 6 can be specifically as follows: Based on the Loss function, generate a tensor vector through MenuNet, and use the vector as the coefficient of each service; generate a package that matches the usage information of the i-th user based on the coefficient.

[0108] It should be noted that different packages can be obtained for rational and irrational models, resulting in different packages that match the usage information of the i-th user.

[0109] This allows for better matching of user needs with recommended packages, thus better serving customers.

[0110] Furthermore, this embodiment incorporates game theory to mathematically model the problem of matching mobile phone plans and optimizes it through neural networks, thereby obtaining a new series of plans that are more reasonable and interpretable, thus improving user experience and increasing user stickiness.

[0111] The above embodiments will be described in detail below with specific examples.

[0112] This example considers two types of plans: a standard plan and a King Card plan. Each standard plan can be represented by a quintuple (P, F, T, P). F ,P T This means that the monthly plan fee is P, the plan includes F data and T hours of voice calls, and the unit price per GB of data is P. F The unit price for a voice call per minute is P. T Each King Card package can consist of a quadruple set (P, T, P). F ,P T The ) indicates that the symbol has the same meaning as before. We believe the Wang Card plan will include free data service, meaning free data usage for certain partner apps. Data purchased under a regular plan can be carried over to the next day, while data purchased under the Wang Card plan cannot. For example, if someone uses 0.5GB of data per day, their monthly (30-day) data usage fee under a regular plan would be 15P. F(Assuming no data is included in the package), it would be 30P under the Wang Card plan. F In an extreme case, assuming someone uses only 1MB of data per day for a month, they would still need to pay 30P under the Wang Card plan. F Traffic fees.

[0113] Each person's monthly usage is represented as a triple (f, t, c), where f represents the total data usage for the month, t represents the duration of outgoing voice calls, and c represents the sum of the rounded-down data usage for each day of the month, excluding targeted data usage.

[0114]

[0115] Among them, f i This represents the data usage on day i. This represents the targeted data usage on day i. Therefore, we can calculate the cost of each plan. The monthly cost per person can be divided into three parts: the monthly plan fee, data usage cost, and voice usage cost.

[0116] For a standard package, the cost can be calculated using the following formula:

[0117] cost = P + max(0, (fF)*P) F )+max(0,(tT)*P T )

[0118] For a certain type of King Card plan, the following formula can be used for calculation.

[0119] cost = P + c * P F +max(0,(tT)*P T )

[0120] Each user will choose a plan, and the operator's revenue (rev) will ultimately be the sum of all costs.

[0121] We pre-define six data plans, including four standard plans and two premium plans. The actual rules for these plans might be more complex; for example, data charges might be structured as "5 yuan per GB, 3 yuan per GB after 3 GB usage, and 0.03 yuan / MB for any remaining data." Our designed plans, however, will only consider the price per GB of data, without introducing more complex charging rules. During the competition, we will use two different models to characterize user behavior: a rational model and an irrational model. Under the rational model, users will strictly choose the plan that minimizes their monthly expenses.

[0122] packet = argmin{cost, cost oppo}

[0123] Among them, cost oppo This represents the cost compared to competitors' plans. In this situation, users are very likely to change their plan choices, and if our designed plan is more expensive, customers are easily lost.

[0124] Under the irrational model, we believe there are two types of users: those who are more rational and choose the plan that minimizes their expenses, and those who are irrational and don't research plans at all, paying almost no attention to the cost; they tend to choose plans randomly. Meanwhile, we believe switching carriers inherently incurs a cost m, which may be due to inconveniences of number portability or because users perceive the carrier's call or data services as superior. Therefore, for more rational users, the plan they choose is...

[0125] packet = argmin{cost, cost oppo +m}

[0126] For irrational users, we will use the expected cost, i.e., the average cost of each plan, to represent the expenditure. The plan they choose is...

[0127]

[0128] Through analysis of existing data, we found that irrational models are more in line with reality. In this case, operators can also obtain greater profits, and users are less likely to churn.

[0129] In one embodiment, such as Figure 2 As shown, this is a schematic diagram illustrating the relationship between other app traffic and targeted traffic in a business resource configuration method provided by an embodiment of the present invention. It allows for the analysis of existing data containing free-data traffic. Based on the traffic relationship, the correlation coefficient between the two variables can be calculated.

[0130] Figure 2 The horizontal axis represents traffic from other apps, and the vertical axis represents targeted traffic. For example, we calculated a correlation coefficient of 0.349 between these two variables, indicating a certain degree of correlation. This means that, to some extent, the traffic from free-data apps can be predicted based on the total amount of data used. We chose to use a regression model for training and prediction, and through comparative analysis, we ultimately selected a linear regression model. Furthermore, we assumed that the actual free-data ratio is the same across different Wang Card plans.

[0131] like Figure 3As shown, this is a bar chart illustrating the cost difference between a user's currently selected plan and the optimal plan, provided by an embodiment of the present invention for a service resource allocation method. During data analysis, it was found that many users did not choose the optimal plan. This situation was analyzed, and the corresponding bar chart was created. The horizontal axis represents the difference between each user's currently selected plan and the optimal plan. The median of these differences is used as the cost m value for switching operators.

[0132] Due to the presence of competitors, in order to avoid the gradient vanishing problem, we did not use the softmin function, but instead introduced a new loss calculation method.

[0133]

[0134] in This represents the benchmark value we use for comparison when calculating revenue. In the rational model, it is the lowest-priced value among the competitor's six packages. In the irrational model, it is a comprehensive result considering previous billing data, competitor charges, and the cost m of switching operators.

[0135] Regarding parameter settings

[0136]

[0137] Table 1 shows the information for multiple pre-set packages.

[0138] The cost of switching carriers in this model is set to 20. During training, the loss hyperparameter 'a' is set to 4, the Adam optimizer is used, the learning rate is 0.01, the batch size is 6000, and the training limit is set to 100 epochs. Under the rational model, we generate 4 regular data plans and 1 premium data plan; under the irrational model, we generate 6 regular data plans and 2 premium data plans. The linear regression model for predicting free data usage is as follows:

[0139] y = min(x, 0.309x + 7.879)

[0140] result:

[0141] The meal plans generated by MenuNet are in tensor form, where each component is a four-decimal-digit number. By rounding, they are restored to a meal plan format that conforms to the general public's understanding.

[0142]

[0143]

[0144] Table 2: Packages under the Rational Model

[0145] This package could generate 1,846,500.08 yuan in revenue, representing a 51.78% increase compared to the baseline of the rational model. While this increase may lack tangible value, in the rational model, users will strictly choose the package that minimizes their spending. Therefore, designing packages based solely on experience may fail to cover the usage patterns of some users, leading to their churn. However, using automated design methods can retain as many users as possible, thereby maximizing the company's revenue.

[0146] serial number type Monthly fee Internal flow In-package voice Traffic unit price Voice unit price 1 ordinary 12 0 0 5.6 yuan per GB 0.15 2 ordinary 27 9 100 4.2 yuan per GB 0.15 3 ordinary 39 11 0 2.9 yuan per GB 0.07 4 ordinary 46 16 350 3.2 yuan per GB 0.23 5 ordinary 82 18 90 0.8 yuan per GB 0.15 6 ordinary 90 12 370 2.0 yuan per GB 0.12 7 Wang Ka 51 / 50 0.8 yuan per GB, valid for the day. 0.3 8 Wang Ka 82 / 380 3.7 yuan per GB, valid for the day. 0.03

[0147] Table 3: Packages under the Irrational Model

[0148] Under this package, the revenue could reach 4,716,406.32 yuan, representing an 8.75% increase compared to the baseline of the irrational model. Traditionally, packages designed based on experience typically have strict "tiers," meaning that as the monthly fee increases, the included data and voice minutes gradually increase, while the unit price generally remains unchanged. However, our experiments show that these can be adjusted to meet the needs of different customer types. Packages with more built-in data and voice minutes may have a higher unit price, while those with fewer built-in minutes will have a lower unit price. Packages with more data may have cheaper call minutes… Through our experiments, we also found that the earlier pricing of the Wang Card was somewhat cheap, possibly to promote this package format. However, due to the Wang Card's free data feature, even if the purchased data is only valid for the day, the price can be appropriately increased.

[0149] Furthermore, the resulting packages are not unique; many different combinations can achieve similarly excellent results. A series of packages with basic income guarantees can be automatically generated initially, and then further refined through human experience to design a more market-preferred package series. This algorithm framework can also be extended to other datasets. If 5G packages become mainstream in the future, this method can also be used to automatically design packages.

[0150] Therefore, this example addresses the problem of telecommunications operator package design by mathematically modeling and proposing an automatic design algorithm based on MenuNet. The expected goal of this method is to increase revenue by 1% compared to the original model. According to the experimental results, even under the irrational model, revenue can be increased by 8.75% on the cleaned dataset representing 40% of the total, which can be considered as an increase of 3.5% compared to the original revenue. Therefore, the expected goal has been achieved under both the rational and irrational models.

[0151] Based on real-world data, a mathematical model was developed for the relationship between mobile phone plans and user behavior. An automatic mechanism design algorithm for mobile phone plans was then designed using the MenuNet structure. This algorithm is simple and efficient, capable of quickly generating various plan series with guaranteed income, and possesses strong scalability, easily extending to other datasets. Experimental results demonstrate that the model and method used in this study have achieved the expected goals.

[0152] Figure 4 This is a flowchart illustrating a business resource configuration system provided in an embodiment of the present invention, such as... Figure 4 As shown, this embodiment provides a business resource configuration system 20, which includes:

[0153] The first acquisition module 201 is configured to acquire multiple preset business resource information, wherein the business resource information includes at least: business type and business configuration resource quantity;

[0154] The second acquisition module 202 is configured to acquire usage information of multiple preset users;

[0155] The first calculation module 203 is configured to calculate the amount of business configuration resources for each user who selects different business resource information based on business resource information and usage information of multiple preset users.

[0156] The third acquisition module 204 is configured to acquire the baseline value of each user among multiple preset users under different configuration resource quantity calculation models;

[0157] The second calculation module 205 is configured to obtain the Loss function based on the configuration resource amount of different services selected by each user among multiple preset users and the baseline value under different configuration resource amount calculation models.

[0158] Configuration module 206 is configured to configure business resources for users based on the Loss function.

[0159] Therefore, the business resource configuration system in this embodiment can determine the most suitable business resource configuration for the user based on business resource information and user usage information, thereby improving user experience and increasing user stickiness.

[0160] Based on the same technical concept, embodiments of the present invention also provide a computer device, such as... Figure 5 As shown, the computer device 5 includes a memory 51 and a processor 52. The memory 51 stores a computer program. When the processor 52 runs the computer program stored in the memory 51, the processor 52 executes the aforementioned business resource configuration method.

[0161] Based on the same technical concept, embodiments of the present invention also provide a computer-readable storage medium storing a computer program thereon, wherein when the computer program is executed by a processor, the processor executes the aforementioned business resource configuration method.

[0162] In summary, the business resource configuration method, system, computer device, and storage medium provided in this embodiment of the invention acquire multiple preset business resource information and usage information of multiple preset users. Based on the multiple business resource information and the usage information of multiple preset users, the configuration resource amount for each user selecting different business resource information is calculated, and the baseline value for each user under different configuration resource amount calculation models is obtained. A Loss function is obtained based on the configuration resource amount for each user selecting different business resource information and the baseline value under different configuration resource amount calculation models. Based on the Loss function, business resource configuration is performed on the user. This can determine the most suitable business resource configuration for the user based on the business resource information and user usage information, thereby improving user experience and increasing user stickiness.

[0163] It will be understood by those skilled in the art that all or some of the steps, systems, or apparatuses disclosed above, and their functional modules / units, can be implemented as software, firmware, hardware, or suitable combinations thereof. In hardware implementations, the division between functional modules / units mentioned in the above description does not necessarily correspond to the division of physical components; for example, a physical component may have multiple functions, or a function or step may be performed collaboratively by several physical components. Some or all physical components may be implemented as software executed by a processor, such as a central processing unit, digital signal processor, or microprocessor, or as hardware, or as an integrated circuit, such as an application-specific integrated circuit (ASIC). Such software may be distributed on a computer-readable medium, which may include computer storage media (or non-transitory media) and communication media (or transient media). As is known to those skilled in the art, the term computer storage media includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules, or other data). Computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, digital versatile disc (DVD) or other optical disc storage, magnetic cartridges, magnetic tape, disk storage or other magnetic storage devices, or any other medium that can be used to store desired information and can be accessed by a computer. Furthermore, it is well known to those skilled in the art that communication media typically contain computer-readable instructions, data structures, program modules, or other data in modulated data signals such as carrier waves or other transmission mechanisms, and may include any information delivery medium.

[0164] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for configuring business resources, characterized in that, include: Obtain multiple preset business resource information, wherein the business resource information includes at least: business type and the amount of configuration resources corresponding to the business type; Obtain usage information from multiple preset users; Based on multiple business resource information and the usage information of multiple preset users, calculate the configuration resource amount for each user among the multiple preset users who select different business resource information; Obtain the baseline value for each user among multiple preset users under different resource allocation calculation models; The Loss function is obtained by selecting different service resource information configuration amounts for each user among multiple preset users and the baseline values ​​under different configuration resource amount calculation models. Based on the Loss function, configure business resources for users; The resource allocation calculation model includes both rational and irrational models, and the user's usage information includes the user's current business resource information. Obtain the baseline value for each user among multiple preset users under different resource allocation calculation models, including: Under the rational model, the value with the lowest configured resource amount among multiple business resource information is used as the benchmark value; Under the irrational model, for rational users, the minimum value of the configuration resource amount of the user's current business resource information within a preset time period, and the minimum value among the sum of the minimum configuration resource amount among multiple business resource information and the replacement difference m, are selected as the benchmark value. For irrational users, the minimum value among the average configuration resource amount of the user's current business resource information within a preset time period and the sum of the average configuration resource amount and the replacement difference m among multiple business resource information is selected as the benchmark value.

2. The method according to claim 1, characterized in that, The business type includes a first business type, the business resource information also includes the resources configured by the business resource information, and the user's usage information includes at least: the first resource f and the second resource t used by the user within a preset time period; When the service type is the first service type, the service configuration resource amount for each user among the multiple preset users who select different service resource information is calculated based on multiple service resource information and user usage information, including: Based on the first formula, the service configuration resource amount for each user selecting different service resource information is calculated respectively. The first formula includes: cost=P+max(0,(f-F)*P F )+max(0,(t-T)*P T ); Where f represents the first resource used by the user within a preset time period, t represents the second resource used by the user within a preset time period, P represents the service configuration resource within a preset time period in the service resource information selected by the current user, T represents the second resource configured within a preset time period in the service resource information selected by the current user, F represents the first resource configured within a preset time period in the service resource information selected by the current user, and P F Among the business resource information selected by the current user, the configured resource quantity of the unit's first resource, P T The configured resource quantity of the second resource in the business resource information selected for the current user within a unit time period, where the current user is one of the preset users.

3. The method according to claim 1, characterized in that, The business type also includes a second business type, the business resource information also includes resources configured by the business resource information, and the user's usage information also includes at least: a third resource c used by the user within a preset time period; When the service type is the second service type, the service configuration resource amount for each user among multiple preset users who select different service resource information is calculated based on service resource information and user usage information, including: The user's current service configuration resource quantity is calculated according to the second formula, which includes: cost=P+c*P F +max(0,(t-T)*P T ); Where P represents the service configuration resource within a preset time period from the service resource information selected by the current user, T represents the second resource configured within the preset time period from the service resource information selected by the current user, and c represents the third resource used by the user within the preset time period. F Among the business resource information selected by the current user, the configured resource quantity of the unit's first resource, P T The configured resource quantity of the second resource in the business resource information selected for the current user within a unit time period, where the current user is one of the preset users.

4. The method according to claim 1, characterized in that, The replacement difference m is determined through the following steps: Based on the user's usage information, determine the optimal configuration resource amount for the user's business resources; The median of the difference between the current configured resource amount of the user's business resource information and the configured resource amount of the optimal business resource information is used as the replacement difference m.

5. The method according to claim 1, characterized in that, The configuration of business resources for users based on the Loss function includes: Based on the Loss function, tensor vectors are generated using MenuNet, and these vectors are used as coefficients. Configure business resources for users based on the coefficients.

6. A business resource allocation system, characterized in that, include: The first acquisition module is configured to acquire multiple preset business resource information, wherein the business resource information includes at least: business type and business configuration resource quantity; The second acquisition module is configured to acquire usage information from multiple preset users; The first calculation module is configured to calculate the amount of business configuration resources for each user who selects different business resource information, based on business resource information and usage information of multiple preset users. The third acquisition module is configured to acquire the baseline value of each user among multiple preset users under different resource quantity calculation models; The second calculation module is configured to obtain the Loss function based on the configuration resource amount of different services selected by each of the multiple preset users and the baseline value under the different configuration resource amount calculation models. The configuration module is set to configure business resources for users based on the Loss function; The resource allocation calculation model includes both rational and irrational models, and the user's usage information includes the user's current business resource information. Obtain the baseline value for each user among multiple preset users under different resource allocation calculation models, including: Under the rational model, the value with the lowest configured resource amount among multiple business resource information is used as the benchmark value; Under the irrational model, for rational users, the minimum value of the configuration resource amount of the user's current business resource information within a preset time period, and the minimum value among the sum of the minimum configuration resource amount among multiple business resource information and the replacement difference m, are selected as the benchmark value. For irrational users, the minimum value among the average configuration resource amount of the user's current business resource information within a preset time period and the sum of the average configuration resource amount and the replacement difference m among multiple business resource information is selected as the benchmark value.

7. A computer device, characterized in that, It includes a memory and a processor, wherein the memory stores a computer program, and when the processor runs the computer program stored in the memory, the processor executes the business resource configuration method according to any one of claims 1 to 5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, the processor performs the business resource configuration method according to any one of claims 1 to 5.

Images