Resource allocation method and device, computer equipment and computer readable storage medium

Abstract

The invention relates to a resource allocation method and device, computer equipment and a computer readable storage medium. The method comprises the following steps: for each application platform, fitting to obtain a relation parameter group corresponding to the application platform according to a preset relation function, historical release resources, historical exposure, historical click rate and historical installation quality parameters; the relation parameter group comprises a first relation parameter between the release resource and the exposure, a second relation parameter between the exposure and the click rate, and a third relation parameter between the click rate and the installation quality parameter; according to the relation function and the relation parameter group, determining a target relation function between the putting resource corresponding to the application platform and the installation quality parameter; and according to the target relation function corresponding to each application platform and a preset total resource amount, determining resource allocation data corresponding to the maximum installation quality parameter, the resource allocation data being used for resource allocation processing of the target application. By adopting the method, application putting resources can be reasonably distributed.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, and in particular to a resource allocation method, apparatus, computer device, and computer-readable storage medium. Background Art

[0002] Application platforms, also known as app stores, are important channels for users to download games and applications. The platform model of app stores is that third-party developers upload their own applications (Application, APP) to the app store, the app store provides traffic for the application, and users download it from the app store and use the services provided by the application.

[0003] However, the resource delivery standards and user quality of each application platform are different. Therefore, third-party developers need to deliver different resources on different application platforms. Therefore, a resource allocation method that can reasonably allocate application delivery resources is urgently needed. Summary of the invention

[0004] Based on this, it is necessary to provide a resource allocation method, device, computer equipment and computer-readable storage medium for reasonably allocating application delivery resources in response to the above technical problems.

[0005] In a first aspect, the present application provides a resource allocation method, comprising:

[0006] For each application platform, obtain the historical delivery resources, historical exposure, historical click volume and historical installation quality parameters corresponding to each application on the application platform within a preset historical time period;

[0007] According to the preset relationship function, the historical delivery resources, the historical exposure, the historical click volume and the historical installation quality parameter, a relationship parameter group corresponding to the application platform is fitted; the relationship parameter group includes a first relationship parameter between delivery resources and exposure, a second relationship parameter between exposure and click volume, and a third relationship parameter between click volume and installation quality parameter;

[0008] Determine, according to the relationship function and the relationship parameter group, a target relationship function between the delivery resources corresponding to the application platform and the installation quality parameters;

[0009] According to the target relationship function corresponding to each of the application platforms and the preset total amount of resources, resource allocation data corresponding to the maximum installed quality parameter is determined, and the resource allocation data is used to perform resource allocation processing for the target application.

[0010] In one embodiment, the relationship parameter group corresponding to the application platform is obtained by fitting according to the preset relationship function, the historical delivery resources, the historical exposure, the historical click volume and the historical installation quality parameter, including:

[0011] Fitting a first relationship curve between delivery resources and exposure amounts according to a preset first function model, the historical delivery resources, and the historical exposure amounts, and determining a first relationship parameter corresponding to the application platform according to the first relationship curve;

[0012] Fitting a second relationship curve between exposure and click volume according to a preset second function model, the historical exposure volume, and the historical click volume, and determining a second relationship parameter corresponding to the application platform according to the second relationship curve;

[0013] According to the preset third function model, the historical click volume, and the historical installation quality parameters, a third relationship curve between the click volume and the installation quality parameters is fitted, and according to the third relationship curve, a third relationship parameter corresponding to the application platform is determined.

[0014] In one embodiment, the relationship function includes a first relationship function between delivery resources and exposure, a second relationship function between exposure and click volume, and a third relationship function between click volume and installation quality parameters, and determining the target relationship function between delivery resources corresponding to the application platform and installation quality parameters according to the relationship function and the relationship parameter group includes:

[0015] Determine a target parameter-free function between the delivery resources corresponding to the application platform and the installation quality parameter according to the first relationship function, the second relationship function and the third relationship function;

[0016] The relationship parameter group is input into the target parameter-free function to obtain the target relationship function between the deployment resources corresponding to the application platform and the installation quality parameters.

[0017] In one embodiment, the third relationship function includes a first sub-function between click volume and installed volume, a second sub-function between click volume and login volume, and a third sub-function between click volume and retention volume, and determining the target parameter-free function between the delivery resources corresponding to the application platform and the installed volume quality parameters according to the first relationship function, the second relationship function, and the third relationship function includes:

[0018] Determine a first parameter-free function between the deployed resources and the installed capacity according to the first relationship function, the second relationship function and the first sub-function;

[0019] Determine a second parameter-free function between delivery resources and login quantity according to the first relationship function, the second relationship function and the second sub-function;

[0020] Determine a third parameter-free function between delivery resources and retention amount according to the first relationship function, the second relationship function and the third sub-function;

[0021] The first parameter-free function, the second parameter-free function and the third parameter-free function are added together to obtain a target parameter-free function between the deployment resources corresponding to the application platform and the installation quality parameters.

[0022] In one embodiment, after determining the resource allocation data corresponding to the maximum installed quality parameter according to the target relationship function corresponding to each of the application platforms and the preset total amount of resources, the method further includes:

[0023] When the preset resource allocation update conditions are met, for each application platform, the delivery resources, exposure, click volume and installation quality parameters of the target application on the application platform within a preset time period are obtained;

[0024] Determine an exposure prediction value, a click prediction value, and an installation quality parameter prediction value according to the relationship function, the relationship parameter group, the delivery resources, the exposure, and the click volume;

[0025] Calculate the error corresponding to each of the relationship functions according to a preset cost function, the delivery resources, the exposure, the click volume, the installation quality parameter, the exposure prediction value, the click volume prediction value and the installation quality parameter prediction value;

[0026] For each relationship function, according to the error corresponding to the relationship function, a preset parameter updating algorithm is used to update the relationship parameters corresponding to the relationship function, and the step of determining the target relationship function between the deployment resources corresponding to the application platform and the installation quality parameters based on the relationship function and the relationship parameter group is returned.

[0027] In one embodiment, the updating of the relationship parameters corresponding to the relationship function using a preset parameter updating algorithm according to the error corresponding to the relationship function includes:

[0028] For each relationship parameter, according to the error corresponding to the relationship function, calculate the gradient of the cost function with respect to the relationship parameter;

[0029] The relationship parameters are updated according to a preset learning rate and the gradient.

[0030] In a second aspect, the present application further provides a resource allocation device, including:

[0031] The first acquisition module is used to acquire, for each application platform, historical delivery resources, historical exposure, historical click volume, and historical installation quality parameters corresponding to each application on the application platform within a preset historical time period;

[0032] A fitting module, used for fitting a relationship parameter group corresponding to the application platform according to a preset relationship function, the historical delivery resources, the historical exposure, the historical click volume and the historical installation quality parameter; the relationship parameter group includes a first relationship parameter between delivery resources and exposure, a second relationship parameter between exposure and click volume, and a third relationship parameter between click volume and installation quality parameter;

[0033] A first determination module, configured to determine a target relationship function between the delivery resources corresponding to the application platform and the installation quality parameters according to the relationship function and the relationship parameter group;

[0034] The second determination module is used to determine the resource allocation data corresponding to the maximum installed quality parameter according to the target relationship function corresponding to each of the application platforms and the preset total amount of resources, and the resource allocation data is used to perform resource allocation processing on the target application.

[0035] In one embodiment, the fitting module is specifically used for:

[0036] Fitting a first relationship curve between delivery resources and exposure amounts according to a preset first function model, the historical delivery resources, and the historical exposure amounts, and determining a first relationship parameter corresponding to the application platform according to the first relationship curve;

[0037] Fitting a second relationship curve between exposure and click volume according to a preset second function model, the historical exposure volume, and the historical click volume, and determining a second relationship parameter corresponding to the application platform according to the second relationship curve;

[0038] According to the preset third function model, the historical click volume, and the historical installation quality parameters, a third relationship curve between the click volume and the installation quality parameters is fitted, and according to the third relationship curve, a third relationship parameter corresponding to the application platform is determined.

[0039] In one embodiment, the relationship function includes a first relationship function between delivery resources and exposure, a second relationship function between exposure and click volume, and a third relationship function between click volume and installation quality parameter, and the first determination module is specifically used to:

[0040] Determine a target parameter-free function between the delivery resources corresponding to the application platform and the installation quality parameters according to the first relationship function, the second relationship function and the third relationship function;

[0041] The relationship parameter group is input into the target parameter-free function to obtain the target relationship function between the deployment resources corresponding to the application platform and the installation quality parameters.

[0042] In one embodiment, the third relationship function includes a first sub-function between click volume and installed volume, a second sub-function between click volume and login volume, and a third sub-function between click volume and retention volume, and the first determination module is specifically used to:

[0043] Determine a first parameter-free function between the deployed resources and the installed capacity according to the first relationship function, the second relationship function and the first sub-function;

[0044] Determine a second parameter-free function between delivery resources and login quantity according to the first relationship function, the second relationship function and the second sub-function;

[0045] Determine a third parameter-free function between delivery resources and retention amount according to the first relationship function, the second relationship function and the third sub-function;

[0046] The first parameter-free function, the second parameter-free function and the third parameter-free function are added together to obtain a target parameter-free function between the deployment resources corresponding to the application platform and the installation quality parameters.

[0047] In one embodiment, after determining the resource allocation data corresponding to the maximum installed quality parameter according to the target relationship function corresponding to each of the application platforms and the preset total amount of resources, the device further includes:

[0048] The second acquisition module is used to acquire, for each application platform, the delivery resources, exposure, click volume and installation quality parameters of the target application on the application platform within a preset time period when the preset resource allocation update condition is met;

[0049] A third determination module is used to determine the exposure prediction value, the click prediction value and the installation quality parameter prediction value according to the relationship function, the relationship parameter group, the delivery resources, the exposure and the click volume;

[0050] A calculation module, used for calculating the error corresponding to each of the relationship functions according to a preset cost function, the delivery resources, the exposure, the click volume, the installation quality parameter, the exposure prediction value, the click volume prediction value and the installation quality parameter prediction value;

[0051] The updating module is used to update the relationship parameters corresponding to each relationship function according to the error corresponding to the relationship function using a preset parameter updating algorithm, and return to the step of determining the target relationship function between the deployment resources corresponding to the application platform and the installation quality parameters according to the relationship function and the relationship parameter group.

[0052] In one embodiment, the update module is specifically used to:

[0053] For each relationship parameter, according to the error corresponding to the relationship function, calculate the gradient of the cost function with respect to the relationship parameter;

[0054] The relationship parameters are updated according to a preset learning rate and the gradient.

[0055] In a third aspect, the present application further provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps described in the first aspect when executing the computer program.

[0056] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program implements the steps described in the first aspect above when executed by a processor.

[0057] In a fifth aspect, the present application further provides a computer program product, comprising a computer program, which implements the steps described in the first aspect when executed by a processor.

[0058] The resource allocation method, apparatus, computer device, computer-readable storage medium and computer program product described above obtain, for each application platform, historical delivery resources, historical exposure, historical click volume and historical installation quality parameters corresponding to each application of the application platform within a preset historical time period; according to a preset relationship function, the historical delivery resources, the historical exposure, the historical click volume and the historical installation quality parameters, a relationship parameter group corresponding to the application platform is fitted; the relationship parameter group includes a first relationship parameter between delivery resources and exposure, a second relationship parameter between exposure and click volume, and a third relationship parameter between click volume and installation quality parameters; according to the relationship function and the relationship parameter group, a target relationship function between delivery resources and installation quality parameters corresponding to the application platform is determined; according to the target relationship function corresponding to each of the application platforms and the preset total amount of resources, resource allocation data corresponding to the maximum installation quality parameter is determined, and the resource allocation data is used to perform resource allocation processing for the target application. In this way, a separate model is built for each application platform, and through the preset relationship function and the historical delivery resources, historical exposure, historical click volume and historical installation quality parameters corresponding to each application of the application platform within the preset historical time period, the first relationship parameter between the delivery resources and the exposure, the second relationship parameter between the exposure and the click volume, and the third relationship parameter between the click volume and the installation quality parameter are fitted respectively, so as to determine the target relationship function between the delivery resources and the installation quality parameters corresponding to the application platform, and according to the target relationship function corresponding to each application platform and the preset total amount of resources, the resource allocation data corresponding to the maximum installation quality parameter is determined, and used for resource allocation processing of the target application, taking into account the actual situation that the resource delivery standards of each application platform are different from the user quality conditions, under the given total resource constraints, the optimal resources are allocated to each application platform to maximize the installation quality of the application, and the application delivery resources can be reasonably allocated. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0060] FIG1 is a schematic diagram of a flow chart of a resource allocation method in an embodiment;

[0061] FIG2 is a flow chart showing the steps of fitting a relationship parameter group corresponding to an application platform according to a preset relationship function, historical delivery resources, historical exposure, historical click volume, and historical installation quality parameters in one embodiment;

[0062] FIG3 is a schematic diagram of a first relationship curve between the delivery resources and exposure of platform A in one embodiment;

[0063] FIG4 is a schematic diagram of a second relationship curve between exposure and click volume of platform A in one embodiment;

[0064] 5 is a schematic diagram of a first sub-relationship curve between the number of clicks and the number of installed devices, a second sub-relationship curve between the number of clicks and the number of logins, and a third sub-relationship curve between the number of clicks and the number of retained users on platform A in one embodiment;

[0065] FIG6 is a flow chart of the steps of determining the target relationship function between the deployment resources corresponding to the application platform and the installation quality parameters according to the relationship function and the relationship parameter group in one embodiment;

[0066] 7 is a flowchart of steps for determining a target parameter-free function between the delivery resources corresponding to the application platform and the installation quality parameters according to the first relationship function, the second relationship function and the third relationship function in one embodiment;

[0067] FIG8 is a flowchart of the steps further included in the method after determining the resource allocation data corresponding to the maximum installed quality parameter according to the target relationship function corresponding to each application platform and the preset total amount of resources in one embodiment;

[0068] FIG9 is a flow chart of the steps of updating the relationship parameters corresponding to the relationship function using a preset parameter updating algorithm according to the error corresponding to the relationship function in one embodiment;

[0069] FIG10 is a structural block diagram of a resource allocation device in one embodiment;

[0070] FIG. 11 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0071] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0072] In one embodiment, as shown in FIG1 , a resource allocation method is provided. This embodiment uses the method applied to a terminal as an example for illustration. It is understandable that the method can also be applied to a server, and can also be applied to a system including a terminal and a server, and is implemented through the interaction between the terminal and the server. Among them, the terminal can be, but is not limited to, various personal computers, laptops, smart phones, tablet computers, Internet of Things devices and portable wearable devices. The Internet of Things devices can be smart speakers, smart TVs, smart air conditioners, smart car-mounted devices, projection devices, etc. Portable wearable devices can be smart watches, smart bracelets, head-mounted devices, etc. The head-mounted devices can be virtual reality (VR) devices, augmented reality (AR) devices, smart glasses, etc. The server can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides cloud computing services. In this embodiment, the method includes the following steps:

[0073] Step 101, for each application platform, obtain the historical delivery resources, historical exposure, historical click volume and historical installation quality parameters corresponding to each application of the application platform within a preset historical time period.

[0074] In the embodiment of the present application, the application platform, also known as the application platform to be allocated, can be an application platform for the target application to be uploaded, or can be an application platform for the resources corresponding to the target application to be allocated. The application platform can also be an application platform whose data volume of historical delivery resources, historical exposure, historical click volume and historical installation quality parameters are all greater than the corresponding data volume threshold. In this way, filtering out application platforms with insufficient data can improve the confidence of application platform modeling and improve the accuracy of application resource allocation. The target application is the application of the application platform to be uploaded. The historical time period can be a recent period of time, for example, the past 3 months. The historical delivery resources are the delivery resources within the historical time period. The delivery resources are the resources that third-party developers deliver to the application on the application platform. The resources can be currency. The delivery resources can be used to exchange for the traffic of the application on the application platform. The historical exposure is the exposure within the historical time period. The exposure is the number of times the application is displayed to the user on that day. The historical click volume is the click volume within the historical time period. The click volume is the number of times the user clicks on the display information of the application on that day. The historical installation quality parameter is the installation quality parameter within the historical time period. The installation quality parameter is used to indicate the installation quality of the application, and is also used to indicate the promotion effect of the application. The installation quality parameter can be one or more parameters, including one or more of the following: installation volume, login volume, and retention volume. Installation volume refers to the number of times users download and install the app on the day. Login volume refers to the number of times users register and log in to the app on the day. Retention volume refers to the number of users who are currently logged in and remain active in the next 7 days.

[0075] In one example, at every preset first time interval, for each application platform, the terminal obtains the historical delivery resources, historical exposure, historical click volume, and historical installation quality parameters corresponding to each application of the application platform in a preset historical time period. The length of the first time interval may be less than the length of the historical time period, for example, the historical time period is the past 3 months, and the first time interval is 1 month.

[0076] Step 102, according to the preset relationship function, historical delivery resources, historical exposure, historical click volume and historical installation quality parameters, a relationship parameter group corresponding to the application platform is fitted.

[0077] The relationship parameter group includes a first relationship parameter between the delivery resources and the exposure, a second relationship parameter between the exposure and the click volume, and a third relationship parameter between the click volume and the installation quality parameter.

[0078] In an embodiment of the present application, a relationship function is used to represent the relationship between at least two resource delivery parameters of an application. Resource delivery parameters may include delivery resources, exposure, click volume, and installation quality parameters. The relationship function may be one or more. The relationship parameter group includes multiple relationship parameters, and the first relationship parameter, the second relationship parameter, and the third relationship parameter are all relationship parameters. The relationship parameter is a parameter in the relationship function, which is used to represent the relationship between at least two resource delivery parameters of an application.

[0079] The terminal obtains the first relationship parameter between the delivery resources and the exposure corresponding to the application platform by fitting according to the preset relationship function, the historical delivery resources and the historical exposure. The terminal obtains the second relationship parameter between the exposure and the click volume corresponding to the application platform by fitting according to the preset relationship function, the historical exposure and the historical click volume. The terminal obtains the third relationship parameter between the click volume and the installation quality parameter corresponding to the application platform by fitting according to the preset relationship function, the historical click volume and the historical installation quality parameter.

[0080] Step 103: Determine the target relationship function between the deployment resources and the installation quality parameters corresponding to the application platform according to the relationship function and the relationship parameter group.

[0081] In the embodiment of the present application, the terminal determines the target parameter-free function between the delivery resources corresponding to the application platform and the installation quality parameters according to the relationship function. Then, the terminal inputs the relationship parameter group into the target parameter-free function to obtain the target relationship function between the delivery resources corresponding to the application platform and the installation quality parameters.

[0082] Step 104, determining resource allocation data corresponding to the maximum installed quality parameter according to the target relationship function corresponding to each application platform and the preset total amount of resources, the resource allocation data is used to perform resource allocation processing for the target application.

[0083] In the embodiment of the present application, the terminal uses the preset total amount of resources as a constraint condition, and determines the resource allocation data corresponding to the maximum installed quality parameter under the constraint condition according to the target relationship function corresponding to each application platform. Then, the terminal performs resource allocation processing for the target application according to the resource allocation data. Among them, the total amount of resources is the total amount of resources to be deployed to each application platform.

[0084] In one embodiment, step 104 may be expressed as:

[0085] (1)

[0086] (2)

[0087] (3)

[0088] (4)

[0089] in, , and They represent the installed capacity, login volume, and retention volume of the p-th application platform under the r-th type of delivery resources; Indicates the rth type of delivery resource under the pth application platform; Indicates whether to select this delivery resource, if selected, it is 1, otherwise, it is 0; Indicates the total amount of resources.

[0090] In one embodiment, the terminal determines the resource allocation data corresponding to the maximum installed quality parameter according to the target relationship function corresponding to each application platform and the preset total amount of resources, which can be solved by the ortools.linear_solver package of Python3.

[0091] In the above resource allocation method, a separate model is built for each application platform, and the first relationship parameter between the delivery resources and the exposure, the second relationship parameter between the exposure and the click, and the third relationship parameter between the click and the installation quality parameter are fitted respectively through the preset relationship function and the historical delivery resources, historical exposure, historical click and historical installation quality parameters corresponding to each application of the application platform in the preset historical time period, so as to determine the target relationship function between the delivery resources and the installation quality parameters corresponding to the application platform, and according to the target relationship function corresponding to each application platform and the preset total amount of resources, the resource allocation data corresponding to the maximum installation quality parameter is determined, and it is used to perform resource allocation processing for the target application, considering the actual situation that the resource delivery standard of each application platform is different from the user quality situation, and under the given total amount of resources, the optimal resources are allocated to each application platform to maximize the installation quality of the application, so as to reasonably allocate application delivery resources. In addition, this method converts the resource allocation problem into the optimization problem of the installation quality of the application, which can ensure the realization of reasonable allocation of application delivery resources.

[0092] In one embodiment, the terminal sorts the allocated resources corresponding to each application platform in the resource allocation data corresponding to the largest installation quality parameter in order from large to small, and obtains a resource allocation sequence. Then, the terminal uses the application platforms with a preset proportion in the resource allocation sequence as target delivery platforms. Then, the terminal reallocates the total amount of resources to each target delivery platform according to the proportion of the allocated resources corresponding to each target delivery platform, obtains the updated allocated resources corresponding to each target delivery platform, and obtains the updated resource allocation data. Among them, the proportion can be 60%. In this way, no resource allocation is performed for application platforms that have a small impact on the installation quality of the application, which can reduce the complexity of application resource allocation, improve the efficiency of application resource allocation, and is more in line with actual conditions, which can improve the rationality of application resource allocation.

[0093] In an exemplary embodiment, as shown in FIG2 , according to a preset relationship function, historical delivery resources, historical exposure, historical click volume, and historical installation quality parameters, a specific process of fitting a relationship parameter group corresponding to an application platform includes the following steps:

[0094] Step 201, according to a preset first function model, historical delivery resources, and historical exposure, a first relationship curve between delivery resources and exposure is fitted, and according to the first relationship curve, a first relationship parameter corresponding to the application platform is determined.

[0095] In an embodiment of the present application, the terminal fits a first relationship curve between the delivery resources and the exposure according to a preset first function model, historical delivery resources, and historical exposure. Then, the terminal determines the first relationship parameter in the first relationship curve and obtains the first relationship parameter corresponding to the application platform. Among them, the first function model can be a logarithmic function model, or a monotonically increasing logarithmic function model. In this way, the use of a monotonically increasing logarithmic function model to represent the relationship between the delivery resources and the exposure can accurately represent the actual situation that when the allocated resources reach a certain level, the exposure does not have much room for growth, and can improve the accuracy of the relationship between the delivery resources and the exposure, thereby improving the rationality of the application resource allocation. The first relationship parameter is a parameter of the first relationship curve. The first relationship parameter can be multiple parameters. For example, when the first function model is a logarithmic function model, the first relationship parameter can include a first logarithmic parameter and a second logarithmic parameter.

[0096] In one embodiment, the first relationship curve can be expressed as:

[0097] (5)

[0098] in, is the exposure, To allocate resources, is the first logarithmic parameter, is the second logarithmic parameter.

[0099] Step 202, according to a preset second function model, historical exposure, and historical click volume, fit a second relationship curve between exposure and click volume, and determine a second relationship parameter corresponding to the application platform according to the second relationship curve.

[0100] In an embodiment of the present application, the terminal fits a second relationship curve between exposure and click volume according to a preset second function model, historical exposure, and historical click volume. Then, the terminal determines a second relationship parameter in the second relationship curve to obtain a second relationship parameter corresponding to the application platform. Among them, the second function model can be a linear regression function model. The second relationship parameter is a parameter of the second relationship curve. The second relationship parameter can be multiple parameters. For example, when the second function model is a linear regression function model, the second relationship parameter may include a second weight parameter and a second bias parameter.

[0101] In one embodiment, the second relationship curve can be expressed as:

[0102] (6)

[0103] in, For clicks, is the second weight parameter, is the second bias parameter.

[0104] Step 203, fitting a third relationship curve between the click volume and the installation quality parameter according to a preset third function model, historical click volume, and historical installation quality parameter, and determining a third relationship parameter corresponding to the application platform according to the third relationship curve.

[0105] In an embodiment of the present application, the terminal fits a third relationship curve between the click volume and the installation quality parameter according to a preset third function model, historical click volume, and historical installation quality parameter. Then, the terminal determines the third relationship parameter in the third relationship curve to obtain the third relationship parameter corresponding to the application platform. Among them, the third function model can be a linear regression function model. The third relationship parameter is a parameter of the third relationship curve. The third relationship parameter can be multiple parameters. The third relationship curve can include at least one sub-relationship curve.

[0106] In one example, the historical installed quality parameters include historical installed capacity, historical login volume and historical retention volume, the third relationship parameter includes a third weight parameter, a third bias parameter, a fourth weight parameter, a fourth bias parameter, a fifth weight parameter and a fifth bias parameter, and the third relationship curve includes a first sub-relationship curve, a second sub-relationship curve and a third sub-relationship curve. The terminal fits the first sub-relationship curve between the click volume and the installed capacity according to the preset third function model, the historical click volume and the historical installed capacity. Then, the terminal determines the third weight parameter and the third bias parameter in the first sub-relationship curve. The terminal fits the second sub-relationship curve between the click volume and the login volume according to the preset third function model, the historical click volume and the historical login volume. Then, the terminal determines the fourth weight parameter and the fourth bias parameter in the second sub-relationship curve. The terminal fits the third sub-relationship curve between the click volume and the retention volume according to the preset third function model, the historical click volume and the historical retention volume. Then, the terminal determines the fifth weight parameter and the fifth bias parameter in the third sub-relationship curve. For example, the first sub-relationship curve can be expressed as:

[0107] (7)

[0108] The second sub-relationship curve can be expressed as:

[0109] (8)

[0110] The third sub-relationship curve can be expressed as:

[0111] (9)

[0112] in, is the installed capacity, is the number of logins, For the retention amount, is the third weight parameter, is the third bias parameter, is the fourth weight parameter, is the fourth bias parameter, is the fifth weight parameter, is the fifth bias parameter.

[0113] In one embodiment, the first relationship curve between the delivery resources and exposure of platform A obtained by the terminal fitting is shown in Figure 3. The second relationship curve between the exposure and click volume of platform A obtained by the terminal fitting is shown in Figure 4. The first sub-relationship curve between the click volume and the installed volume of platform A, the second sub-relationship curve between the click volume and the login volume, and the third sub-relationship curve between the click volume and the retention volume obtained by the terminal fitting are shown in Figure 5.

[0114] In the above resource allocation method, when determining the relationship between the burst resources and the installation quality parameters of the application platform, the first relationship curve between the delivery resources and the exposure, the second relationship curve between the exposure and the click volume, and the third relationship curve between the click volume and the installation quality parameters are fitted respectively, and a step-by-step approach is adopted to model the application platform, which simplifies and clarifies the complex modeling problem, can reduce the error in the parameter fitting process, improve the accuracy of the relationship parameter group, and then improve the accuracy of the target relationship function between the delivery resources and the installation quality parameters, thereby further improving the rationality of the application delivery resource allocation.

[0115] In an exemplary embodiment, as shown in FIG6 , the relationship function includes a first relationship function between the delivery resources and the exposure, a second relationship function between the exposure and the click volume, and a third relationship function between the click volume and the installation quality parameter. According to the relationship function and the relationship parameter group, the specific process of determining the target relationship function between the delivery resources corresponding to the application platform and the installation quality parameter includes the following steps:

[0116] Step 601, determining a target parameter-free function between the delivery resources corresponding to the application platform and the installation quality parameters according to the first relationship function, the second relationship function and the third relationship function.

[0117] In an embodiment of the present application, the terminal determines a fourth relationship function between the delivery resources corresponding to the application platform and the number of clicks based on the first relationship function and the second relationship function. The terminal determines a target parameter-free function between the delivery resources corresponding to the application platform and the installation quality parameters based on the fourth relationship function and the third relationship function. Among them, the target parameter-free function is a parameter-free function between the delivery resources and the installation quality parameters. The parameters in the parameter-free function are known, and the values ​​of the parameters are unknown.

[0118] Step 602, input the relationship parameter group into the target parameter-free function to obtain the target relationship function between the deployment resources and the installation quality parameters corresponding to the application platform.

[0119] In the embodiment of the present application, the parameters in the target relationship function and the values ​​of the parameters are all known.

[0120] In the above resource allocation method, the target parameter-free function between the delivery resources and the installation quality parameters corresponding to the application platform is determined according to the first relationship function between the delivery resources and the exposure, the second relationship function between the exposure and the click volume, and the third relationship function between the click volume and the installation quality parameters; the relationship parameter group is input into the target parameter-free function to obtain the target relationship function between the delivery resources and the installation quality parameters corresponding to the application platform. In this way, the relationship between the burst resources and the installation quality parameters of the application platform is determined by deriving the first relationship function between the delivery resources and the exposure, the second relationship function between the exposure and the click volume, and the third relationship function between the click volume and the installation quality parameters, and the application platform is modeled in a progressive manner, which simplifies and clarifies the complex modeling problems, can reduce the error in the modeling process of the relationship between the burst resources and the installation quality parameters of the application platform, improve the accuracy of the target relationship function between the delivery resources and the installation quality parameters, and further improve the rationality of the application delivery resource allocation.

[0121] In an exemplary embodiment, as shown in FIG7 , the third relationship function includes a first sub-function between the number of clicks and the number of installations, a second sub-function between the number of clicks and the number of logins, and a third sub-function between the number of clicks and the number of retentions. The specific process of determining the target parameter-free function between the delivery resources corresponding to the application platform and the installation quality parameters according to the first relationship function, the second relationship function, and the third relationship function includes the following steps:

[0122] Step 701, determining a first parameter-free function between investment resources and installed capacity according to the first relationship function, the second relationship function and the first sub-function.

[0123] In the embodiment of the present application, the terminal determines the fourth relationship function between the delivery resources and the click volume corresponding to the application platform according to the first relationship function and the second relationship function. The terminal determines the first parameter-free function between the delivery resources and the installed capacity according to the fourth relationship function and the first sub-function.

[0124] Step 702: Determine a second parameter-free function between the delivery resources and the login amount according to the first relationship function, the second relationship function and the second sub-function.

[0125] In the embodiment of the present application, the terminal determines the fourth relationship function between the delivery resources and the click volume corresponding to the application platform according to the first relationship function and the second relationship function. The terminal determines the second parameter-free function between the delivery resources and the login volume according to the fourth relationship function and the second sub-function.

[0126] Step 703: Determine a third parameter-free function between the delivered resources and the retained amount according to the first relationship function, the second relationship function and the third sub-function.

[0127] In the embodiment of the present application, the terminal determines the fourth relationship function between the delivery resources and the click volume corresponding to the application platform according to the first relationship function and the second relationship function. The terminal determines the third parameter-free function between the delivery resources and the retention volume according to the fourth relationship function and the third sub-function.

[0128] Step 704, adding the first parameter-free function, the second parameter-free function and the third parameter-free function to obtain a target parameter-free function between the deployment resources and the installation quality parameters corresponding to the application platform.

[0129] In one embodiment, the first relationship function is shown in formula (5), the second relationship function is shown in formula (6), the first sub-function is shown in formula (7), the second sub-function is shown in formula (8), and the third sub-function is shown in formula (9). The terminal determines the first parameter-free function between the deployed resources and the installed capacity based on the first relationship function, the second relationship function and the first sub-function, which can be expressed as:

[0130] (10)

[0131] The terminal determines a second parameterless function between the delivery resources and the login amount according to the first relationship function, the second relationship function and the second sub-function, which can be expressed as:

[0132] (11)

[0133] The terminal determines a third parameter-free function between the delivery resources and the retention amount according to the first relationship function, the second relationship function and the third sub-function, which can be expressed as:

[0134] (12)

[0135] Then, the terminal adds the first parameter-free function, the second parameter-free function and the third parameter-free function to obtain the target parameter-free function between the delivery resources and the installation quality parameters corresponding to the application platform, which can be expressed as:

[0136] (13)

[0137] In the above resource allocation method, when determining the target parameter-free function between the delivery resources and the installation quality parameters corresponding to the application platform, the first parameter-free function between the delivery resources and the installation volume is determined according to the first relationship function, the second relationship function and the first sub-function between the click volume and the installation volume, the second parameter-free function between the delivery resources and the login volume is determined according to the first relationship function, the second relationship function and the second sub-function between the click volume and the login volume, the third parameter-free function between the delivery resources and the retention volume is determined according to the first relationship function, the second relationship function and the third sub-function between the click volume and the retention volume, and then the first parameter-free function, the second parameter-free function and the third parameter-free function are added together to obtain. In this way, the installation quality parameters include the installation volume, the login volume and the retention volume, and the maximum sum of the installation volume, the login volume and the retention volume is used as the target of application resource allocation, which not only comprehensively considers the installation quality of the application, is more in line with the actual situation, can further improve the rationality of the application delivery resource allocation, but also simplifies the application resource allocation optimization problem, can improve the efficiency of the application delivery resource allocation, and reduce the resource consumption of the application delivery resource allocation.

[0138] In an exemplary embodiment, as shown in FIG8 , after determining the resource allocation data corresponding to the maximum installed quality parameter according to the target relationship function corresponding to each application platform and the preset total amount of resources, the method further includes the following steps:

[0139] Step 801, when the preset resource allocation update conditions are met, for each application platform, the delivery resources, exposure, click volume and installation quality parameters of the target application on the application platform within a preset time period are obtained.

[0140] In an embodiment of the present application, after performing resource allocation processing of the target application, the terminal determines whether the preset resource allocation update condition is met. When the preset resource allocation update condition is met, for each application platform, the terminal obtains the delivery resources, exposure, click volume and installation quality parameters of the target application on the application platform within a preset time period. Among them, the resource allocation update condition can be a time condition. For example, if the time for performing resource allocation processing of the target application reaches a preset second time interval, the terminal determines that the preset resource allocation update condition is met. For example, after performing resource allocation processing of the target application, the terminal determines that the preset resource allocation update condition is met at every preset third time interval. The duration of the second time interval may be greater than the third time interval, and the third time interval may be 1 day. The resource allocation update condition may also be a non-time condition, such as an error condition. The duration of the preset time period may be less than the duration of the historical time period.

[0141] Step 802, determining the exposure prediction value, click prediction value and installation quality parameter prediction value according to the relationship function, relationship parameter group, delivery resources, exposure and click volume.

[0142] In the embodiment of the present application, the terminal inputs the relationship parameter corresponding to the first relationship function (i.e., the first relationship parameter) into the first relationship function to obtain the first relationship function determined by the parameter between the delivery resource corresponding to the application platform and the exposure. Then, the terminal calculates the exposure prediction value corresponding to the delivery resource based on the first relationship function determined by the parameter and the delivery resource.

[0143] The terminal inputs the relationship parameter corresponding to the second relationship function (i.e., the second relationship parameter) into the second relationship function to obtain the second relationship function determined by the parameter between the exposure and the click volume corresponding to the application platform. Then, the terminal calculates the click volume prediction value corresponding to the exposure volume according to the second relationship function determined by the parameter and the exposure volume.

[0144] The terminal inputs the relationship parameter (i.e., the third relationship parameter) corresponding to the third relationship function into the third relationship function, and obtains the third relationship function determined by the parameters between the click volume and the installed quality parameter corresponding to the application platform. Then, the terminal calculates the predicted value of the installed quality parameter corresponding to the click volume based on the third relationship function determined by the parameters and the click volume. It can be understood that the specific process of the terminal calculating the predicted value of the installed volume corresponding to the click volume, the specific process of the predicted value of the login volume corresponding to the click volume, and the specific process of calculating the predicted value of the retained volume corresponding to the click volume are similar to the specific process of the terminal calculating the predicted value of the installed volume corresponding to the click volume.

[0145] Step 803, calculating the errors corresponding to the relationship functions according to the preset cost function, delivery resources, exposure, clicks, installation quality parameters, exposure prediction value, click prediction value and installation quality parameter prediction value.

[0146] In the implementation of this application, the terminal calculates the error corresponding to the first relationship function according to the preset cost function, the delivery resources, the exposure and the exposure prediction value. The terminal calculates the error corresponding to the second relationship function according to the preset cost function, the exposure, the click volume and the click prediction value. The terminal calculates the error corresponding to the third relationship function according to the preset cost function, the click volume, the installation quality parameter and the installation quality parameter prediction value. Among them, the cost function can be the mean square error (MSE).

[0147] Step 804, for each relationship function, according to the error corresponding to the relationship function, a preset parameter update algorithm is used to update the relationship parameters corresponding to the relationship function, and return to the step of determining the target relationship function between the deployment resources corresponding to the application platform and the installation quality parameters according to the relationship function and the relationship parameter group.

[0148] In an embodiment of the present application, the parameter updating algorithm may be a gradient descent method.

[0149] In one example, for each relationship function, the terminal uses a preset parameter update algorithm to update the relationship parameters corresponding to the relationship function according to the error corresponding to the relationship function. The terminal updates the total amount of resources and returns to determine the target relationship function step between the resources and the installation quality parameters corresponding to the application platform according to the relationship function and the relationship parameter group. For example, the terminal uses the total amount of remaining resources as the updated total amount of resources. For example, the terminal uses the total amount of resources released this time as the updated total amount of resources. Among them, the total amount of resources corresponding to the two resource allocation data can be different or the same.

[0150] In the above resource allocation method, when the preset resource allocation update conditions are met, for each application platform, the delivery resources, exposure, click volume and installation quality parameters of the target application on the application platform within the preset time period are obtained; according to the relationship function, relationship parameter group, delivery resources, exposure and click volume, the exposure prediction value, click volume prediction value and installation quality parameter prediction value are determined; according to the preset cost function, delivery resources, exposure, click volume, installation quality parameter, exposure prediction value, click volume prediction value and installation quality parameter prediction value, the error corresponding to each relationship function is calculated; for each relationship function, according to the error corresponding to the relationship function, a preset parameter update algorithm is used to update the relationship parameters corresponding to the relationship function, and the step of determining the target relationship function between the delivery resources and installation quality parameters corresponding to the application platform according to the relationship function and the relationship parameter group is returned. In this way, after the application's delivery resources are actually delivered according to the previously determined resource delivery data, when the preset resource allocation update conditions are met, the actual values ​​of various parameters of the target application are collected, and the errors between the predicted values ​​and the actual values ​​of various parameters of the current models of each application platform are calculated. The relationship parameters previously obtained by fitting historical data are corrected in another way, and the original model of the application platform is corrected. Resources are reallocated to be more in line with the actual situation. Timely adjustment of the application platform model can improve the real-time allocation of application delivery resources, and further improve the rationality and accuracy of application delivery resource allocation.

[0151] In an exemplary embodiment, as shown in FIG9 , according to the error corresponding to the relationship function, a preset parameter updating algorithm is used to update the relationship parameter corresponding to the relationship function, and the specific process includes the following steps:

[0152] Step 901 : For each relationship parameter, the gradient of the cost function with respect to the relationship parameter is calculated according to the error corresponding to the relationship function.

[0153] Step 902: Update the relationship parameters according to a preset learning rate and gradient.

[0154] In the above resource allocation method, for each relationship parameter, a gradient descent method is used to update the relationship parameter respectively, which is more in line with the actual situation and can further improve the rationality and accuracy of application delivery resource allocation.

[0155] In one embodiment, the relationship function is a first relationship function, and the first relationship parameter includes a first logarithmic parameter and a second logarithmic parameter. The terminal calculates the exposure prediction value corresponding to the delivery resource according to the first relationship function and the delivery resource determined by the parameters, which can be expressed as:

[0156] (14)

[0157] Then, the terminal uses the mean square error to define the cost function. Then, the terminal calculates the error corresponding to the first relationship function according to the preset cost function, delivery resources, exposure and exposure prediction value, which can be expressed as:

[0158] (15)

[0159] Then, the terminal calculates the gradient of the cost function with respect to the first logarithmic parameter and the gradient of the cost function with respect to the second logarithmic parameter according to the error corresponding to the first relationship function, which can be expressed as:

[0160] (16)

[0161] (17)

[0162] Then, the terminal updates the first logarithmic parameter according to the gradient of the preset learning rate and cost function to the first logarithmic parameter, and updates the second logarithmic parameter according to the gradient of the preset learning rate and cost function to the second logarithmic parameter, which can be expressed as:

[0163] (18)

[0164] (19)

[0165] in, is the resources delivered to the i-th sample; is the predicted value of exposure corresponding to the delivery resource of the i-th sample; is the exposure corresponding to the delivery resource of the i-th sample (i.e., the actual value of the exposure); m is the number of samples; is the error corresponding to the first relationship function; is the gradient of the cost function with respect to the first logarithmic parameter; is the gradient of the cost function with respect to the second logarithmic parameter; is the learning rate, controlling the step length of parameter update, which may be 0.01. It is understandable that the specific process of updating the relationship parameters of other relationship functions of nonlinear models is similar to the specific process of this embodiment.

[0166] In one embodiment, the relationship function is a second relationship function, and the second relationship parameter includes a second weight parameter and a second bias parameter. The terminal calculates the click volume prediction value corresponding to the delivery resource according to the second relationship function and the exposure determined by the parameters, which can be expressed as:

[0167] (20)

[0168] Then, the terminal uses the mean square error to define the cost function. Then, the terminal calculates the error corresponding to the second relationship function according to the preset cost function, exposure, click volume and click volume prediction value, which can be expressed as:

[0169] (twenty one)

[0170] Then, the terminal calculates the gradient of the cost function to the second weight parameter and the gradient of the cost function to the second bias parameter according to the error corresponding to the second relationship function, which can be expressed as:

[0171] (twenty two)

[0172] (twenty three)

[0173] Then, the terminal updates the second logarithmic parameter according to the preset learning rate and the gradient of the cost function to the second logarithmic parameter, and updates the second logarithmic parameter according to the preset learning rate and the gradient of the cost function to the second logarithmic parameter, which can be expressed as:

[0174] (twenty four)

[0175] (25)

[0176] in, is the predicted value of clicks corresponding to the exposure of the i-th sample; is the corresponding click volume of the exposure of the i-th sample (i.e. the actual value of the click volume); is the error corresponding to the second relationship function; is the gradient of the cost function with respect to the second weight parameter; is the gradient of the cost function with respect to the second bias parameter; is the learning rate, controlling the step size of parameter update, which may be 0.01. It is understandable that the specific process of updating the relationship parameters of other relationship functions (eg, the third relationship function, the first sub-function, the second sub-function and the third sub-function) of the linear model is similar to the specific process of this embodiment.

[0177] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0178] Based on the same inventive concept, the embodiment of the present application also provides a resource allocation device for implementing the resource allocation method involved above. The implementation solution provided by the device to solve the problem is similar to the implementation solution recorded in the above method, so the specific limitations in one or more resource allocation device embodiments provided below can refer to the limitations on the resource allocation method above, and will not be repeated here.

[0179] In an exemplary embodiment, as shown in FIG10 , a resource allocation device 1000 is provided, comprising: a first acquisition module 1010, a fitting module 1020, a first determination module 1030 and a second determination module 1040, wherein:

[0180] The first acquisition module 1010 is used to acquire, for each application platform, historical delivery resources, historical exposure, historical click volume, and historical installation quality parameters corresponding to each application of the application platform within a preset historical time period;

[0181] The fitting module 1020 is used to fit the relationship parameter group corresponding to the application platform according to the preset relationship function, the historical delivery resources, the historical exposure, the historical click volume and the historical installation quality parameter; the relationship parameter group includes a first relationship parameter between delivery resources and exposure, a second relationship parameter between exposure and click volume, and a third relationship parameter between click volume and installation quality parameter;

[0182] A first determination module 1030 is used to determine a target relationship function between the delivery resources corresponding to the application platform and the installation quality parameters according to the relationship function and the relationship parameter group;

[0183] The second determination module 1040 is used to determine the resource allocation data corresponding to the maximum installed quality parameter according to the target relationship function corresponding to each of the application platforms and the preset total amount of resources, and the resource allocation data is used to perform resource allocation processing for the target application.

[0184] Optionally, the fitting module 1020 is specifically configured to:

[0185] Fitting a first relationship curve between delivery resources and exposure amounts according to a preset first function model, the historical delivery resources, and the historical exposure amounts, and determining a first relationship parameter corresponding to the application platform according to the first relationship curve;

[0186] Fitting a second relationship curve between exposure and click volume according to a preset second function model, the historical exposure volume, and the historical click volume, and determining a second relationship parameter corresponding to the application platform according to the second relationship curve;

[0187] According to the preset third function model, the historical click volume, and the historical installation quality parameters, a third relationship curve between the click volume and the installation quality parameters is fitted, and according to the third relationship curve, a third relationship parameter corresponding to the application platform is determined.

[0188] Optionally, the relationship function includes a first relationship function between delivery resources and exposure, a second relationship function between exposure and click volume, and a third relationship function between click volume and installation quality parameter. The first determination module 1030 is specifically configured to:

[0189] Determine a target parameter-free function between the delivery resources corresponding to the application platform and the installation quality parameter according to the first relationship function, the second relationship function and the third relationship function;

[0190] The relationship parameter group is input into the target parameter-free function to obtain the target relationship function between the deployment resources corresponding to the application platform and the installation quality parameters.

[0191] Optionally, the third relationship function includes a first sub-function between the number of clicks and the number of installed devices, a second sub-function between the number of clicks and the number of logins, and a third sub-function between the number of clicks and the number of retained users. The first determination module 1030 is specifically configured to:

[0192] Determine a first parameter-free function between the deployed resources and the installed capacity according to the first relationship function, the second relationship function and the first sub-function;

[0193] Determine a second parameter-free function between delivery resources and login quantity according to the first relationship function, the second relationship function and the second sub-function;

[0194] Determine a third parameter-free function between delivery resources and retention amount according to the first relationship function, the second relationship function and the third sub-function;

[0195] The first parameter-free function, the second parameter-free function and the third parameter-free function are added together to obtain a target parameter-free function between the deployment resources corresponding to the application platform and the installation quality parameters.

[0196] Optionally, after determining the resource allocation data corresponding to the maximum installed quality parameter according to the target relationship function corresponding to each of the application platforms and the preset total amount of resources, the resource allocation device 1000 further includes:

[0197] The second acquisition module is used to acquire, for each application platform, the delivery resources, exposure, click volume and installation quality parameters of the target application on the application platform within a preset time period when the preset resource allocation update condition is met;

[0198] A third determination module is used to determine the exposure prediction value, the click prediction value and the installation quality parameter prediction value according to the relationship function, the relationship parameter group, the delivery resources, the exposure and the click volume;

[0199] A calculation module, used for calculating the error corresponding to each of the relationship functions according to a preset cost function, the delivery resources, the exposure, the click volume, the installation quality parameter, the exposure prediction value, the click volume prediction value and the installation quality parameter prediction value;

[0200] The updating module is used to update the relationship parameters corresponding to each relationship function according to the error corresponding to the relationship function using a preset parameter updating algorithm, and return to the step of determining the target relationship function between the deployment resources corresponding to the application platform and the installation quality parameters according to the relationship function and the relationship parameter group.

[0201] Optionally, the update module is specifically used to:

[0202] For each relationship parameter, according to the error corresponding to the relationship function, calculate the gradient of the cost function with respect to the relationship parameter;

[0203] The relationship parameters are updated according to a preset learning rate and the gradient.

[0204] Each module in the resource allocation device can be implemented in whole or in part by software, hardware, or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute operations corresponding to each module.

[0205] In an exemplary embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be shown in FIG11. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit, and an input device. The processor, the memory, and the input / output interface are connected via a system bus, and the communication interface, the display unit, and the input device are connected to the system bus via the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner may be implemented through WIFI, a mobile cellular network, near field communication (NFC) or other technologies. When the computer program is executed by the processor, a resource allocation method is implemented. The display unit of the computer device is used to form a visually visible picture, which may be a display screen, a projection device, or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad set on the computer device shell, or an external keyboard, touchpad or mouse.

[0206] Those skilled in the art will understand that the structure shown in FIG. 11 is merely a block diagram of a partial structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different arrangement of components.

[0207] In an exemplary embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above-mentioned method embodiments when executing the computer program.

[0208] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0209] In one embodiment, a computer program product is provided, including a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0210] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0211] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., but are not limited to this.

[0212] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0213] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A resource allocation method, characterized in that: The method comprises: For each application platform, obtain the historical delivery resources, historical exposure, historical click volume and historical installation quality parameters corresponding to each application on the application platform within a preset historical time period; According to the preset relationship function, the historical delivery resources, the historical exposure, the historical click volume and the historical installation quality parameter, a relationship parameter group corresponding to the application platform is fitted; the relationship parameter group includes a first relationship parameter between delivery resources and exposure, a second relationship parameter between exposure and click volume, and a third relationship parameter between click volume and installation quality parameter; Determine, according to the relationship function and the relationship parameter group, a target relationship function between the delivery resources corresponding to the application platform and the installation quality parameters; According to the target relationship function corresponding to each of the application platforms and the preset total amount of resources, resource allocation data corresponding to the maximum installed quality parameter is determined, and the resource allocation data is used to perform resource allocation processing for the target application.

2. The method according to claim 1, characterized in that The fitting of the relationship parameter group corresponding to the application platform according to the preset relationship function, the historical delivery resources, the historical exposure, the historical click volume and the historical installation quality parameter includes: Fitting a first relationship curve between delivery resources and exposure amounts according to a preset first function model, the historical delivery resources, and the historical exposure amounts, and determining a first relationship parameter corresponding to the application platform according to the first relationship curve; Fitting a second relationship curve between exposure and click volume according to a preset second function model, the historical exposure volume, and the historical click volume, and determining a second relationship parameter corresponding to the application platform according to the second relationship curve; According to the preset third function model, the historical click volume, and the historical installation quality parameter, a third relationship curve between the click volume and the installation quality parameter is fitted, and according to the third relationship curve, a third relationship parameter corresponding to the application platform is determined.

3. The method according to claim 1, characterized in that The relationship function includes a first relationship function between delivery resources and exposure, a second relationship function between exposure and click volume, and a third relationship function between click volume and installation quality parameters. Determining the target relationship function between delivery resources corresponding to the application platform and installation quality parameters according to the relationship function and the relationship parameter group includes: Determine a target parameter-free function between the delivery resources corresponding to the application platform and the installation quality parameter according to the first relationship function, the second relationship function and the third relationship function; The relationship parameter group is input into the target parameter-free function to obtain the target relationship function between the deployment resources corresponding to the application platform and the installation quality parameters.

4. The method according to claim 3, characterized in that The third relationship function includes a first sub-function between click volume and installed volume, a second sub-function between click volume and login volume, and a third sub-function between click volume and retention volume. The target parameter-free function between the delivery resources corresponding to the application platform and the installed volume quality parameters is determined according to the first relationship function, the second relationship function, and the third relationship function, including: Determine a first parameter-free function between the deployed resources and the installed capacity according to the first relationship function, the second relationship function and the first sub-function; Determine a second parameter-free function between delivery resources and login quantity according to the first relationship function, the second relationship function and the second sub-function; Determine a third parameter-free function between delivery resources and retention amount according to the first relationship function, the second relationship function and the third sub-function; The first parameter-free function, the second parameter-free function and the third parameter-free function are added together to obtain a target parameter-free function between the deployment resources corresponding to the application platform and the installation quality parameters.

5. The method according to claim 1, characterized in that After determining the resource allocation data corresponding to the maximum installed quality parameter according to the target relationship function corresponding to each of the application platforms and the preset total amount of resources, the method further includes: When the preset resource allocation update conditions are met, for each application platform, the delivery resources, exposure, click volume and installation quality parameters of the target application on the application platform within a preset time period are obtained; Determine an exposure prediction value, a click prediction value, and an installation quality parameter prediction value according to the relationship function, the relationship parameter group, the delivery resources, the exposure, and the click volume; Calculate the error corresponding to each of the relationship functions according to a preset cost function, the delivery resources, the exposure, the click volume, the installation quality parameter, the exposure prediction value, the click volume prediction value and the installation quality parameter prediction value; For each relationship function, according to the error corresponding to the relationship function, a preset parameter updating algorithm is used to update the relationship parameters corresponding to the relationship function, and the step of determining the target relationship function between the deployment resources corresponding to the application platform and the installation quality parameters based on the relationship function and the relationship parameter group is returned.

6. The method according to claim 5, characterized in that The updating of the relationship parameters corresponding to the relationship function by using a preset parameter updating algorithm according to the error corresponding to the relationship function includes: For each relationship parameter, according to the error corresponding to the relationship function, calculate the gradient of the cost function with respect to the relationship parameter; The relationship parameters are updated according to a preset learning rate and the gradient.

7. A resource allocation device, characterized in that: The device comprises: The first acquisition module is used to acquire, for each application platform, historical delivery resources, historical exposure, historical click volume, and historical installation quality parameters corresponding to each application on the application platform within a preset historical time period; A fitting module, used for fitting a relationship parameter group corresponding to the application platform according to a preset relationship function, the historical delivery resources, the historical exposure, the historical click volume and the historical installation quality parameter; the relationship parameter group includes a first relationship parameter between delivery resources and exposure, a second relationship parameter between exposure and click volume, and a third relationship parameter between click volume and installation quality parameter; A first determination module, configured to determine a target relationship function between the delivery resources corresponding to the application platform and the installation quality parameters according to the relationship function and the relationship parameter group; The second determination module is used to determine the resource allocation data corresponding to the maximum installed quality parameter according to the target relationship function corresponding to each of the application platforms and the preset total amount of resources, and the resource allocation data is used to perform resource allocation processing on the target application.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images