Computing power matching method and system based on artificial intelligence

Abstract

The present invention discloses a computing power matching method and system based on artificial intelligence, the method comprising: obtaining multiple computing power task requests sent by multiple users and user parameters of each of the users; determining multiple candidate devices in a preset device library according to the task parameters corresponding to each of the computing power task requests; determining the computing power device characteristics corresponding to each of the users according to a neural network algorithm based on the task parameters and the user parameters; determining the computing power device corresponding to each of the users from the multiple candidate devices based on a dynamic programming algorithm according to the computing power device characteristics. It can be seen that the present invention can more accurately and efficiently predict the computing power device characteristics corresponding to the user, so as to improve the recommended matching degree of the computing power device, and improve the overall computing power device utilization and user experience.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a computing power matching method and system based on artificial intelligence. Background Art

[0002] With the improvement and popularization of algorithm technologies such as neural networks, more and more users have begun to use online computing resources to perform some customized algorithm training tasks or algorithm execution tasks. Correspondingly, more and more platforms have begun to integrate online computing resources to provide users with corresponding computing services, which puts higher requirements on how computing platforms can effectively match users' task requirements. When providing computing devices to users, most of the existing technologies only determine the computing devices through direct designation by the user, or through simple matching rules between tasks and computing devices, and do not fully combine task parameters and user parameters to match users with appropriate computing devices. Therefore, the utilization rate of computing devices is not high, and it is impossible to provide users with a better computing experience. It can be seen that the existing technology has defects that need to be solved urgently. Summary of the invention

[0003] The technical problem to be solved by the present invention is to provide an artificial intelligence-based computing power matching method and system, which can more accurately and efficiently predict the computing power device characteristics corresponding to the user, so as to improve the recommended matching degree of the computing power device, improve the overall computing power device utilization rate and user experience.

[0004] In order to solve the above technical problems, the first aspect of the present invention discloses a computing power matching method based on artificial intelligence, the method comprising:

[0005] Obtaining multiple computing task requests sent by multiple users and user parameters of each of the users;

[0006] According to the task parameters corresponding to each of the computing task requests, a plurality of candidate devices are determined in a preset device library;

[0007] Based on the task parameters and the user parameters, determining the computing power device characteristics corresponding to each of the users according to a neural network algorithm;

[0008] According to the characteristics of the computing power device, a computing power device corresponding to each of the users is determined from the multiple candidate devices based on a dynamic programming algorithm.

[0009] As an optional implementation, in the first aspect of the present invention, the user parameters include at least one of user physiological characteristics, user historical computing tasks, user device parameters and user web page operations.

[0010] As an optional embodiment, in the first aspect of the present invention, the task parameters include at least one of task data volume, task computing volume, task type, task data type, task purpose, task application environment and task programming language.

[0011] As an optional implementation, in the first aspect of the present invention, the determining of multiple candidate devices in a preset device library according to the task parameters corresponding to each computing task request includes:

[0012] Determine the device performance requirements corresponding to the task parameters corresponding to each of the computing power task requests according to the correspondence between the preset task parameters and the device performance requirements;

[0013] For each standby device in the preset device library, calculate the average value of the matching degree between the standby device and the device performance requirements corresponding to all the computing power task requests to obtain the matching degree parameter of the standby device;

[0014] All the backup devices whose matching degree parameters are greater than a preset first parameter threshold are determined as candidate devices.

[0015] As an optional implementation, in the first aspect of the present invention, determining the computing power device characteristics corresponding to each user based on the task parameters and the user parameters according to a neural network algorithm includes:

[0016] For each of the users, inputting the user parameter of the user into a trained first device preference prediction neural network to obtain the first device preference parameter of the user;

[0017] Inputting the task parameters of the user into a trained second device preference prediction neural network to obtain the second device preference parameters of the user;

[0018] Calculate the intersection of the first device preference parameter and the second device preference parameter to obtain a computing power device feature set corresponding to the user;

[0019] For any two of the users, the computing power device feature sets of the two users are modified according to the user parameters and the task parameters of the two users to obtain the computing power device features corresponding to the two users.

[0020] As an optional embodiment, in the first aspect of the present invention, the first device preference prediction neural network is a CNN neural network, which is trained by a training data set including multiple training user parameters and corresponding device preference parameter annotations; the second device preference prediction neural network is a CNN neural network, which is trained by a training data set including multiple training task parameters and corresponding device preference parameter annotations.

[0021] As an optional implementation, in the first aspect of the present invention, the computing power device feature set of the two users is modified according to the user parameters and the task parameters of the two users to obtain computing power device features corresponding to the two users, including:

[0022] For any device feature, the number of occurrences of the device feature in the computing power device feature set of the two users is calculated; the device feature is device performance, device hardware information, device model, device location or device provider information;

[0023] Calculating a first parameter similarity between the user parameters of the two users;

[0024] Calculating a second parameter similarity between the task parameters of the two users;

[0025] Calculating the reciprocal of the weighted average of the first parameter similarity and the second parameter similarity to obtain a difference parameter between the two users;

[0026] Calculating a number threshold proportional to the difference parameter;

[0027] Determine the device feature whose occurrence times are lower than the times threshold as a rejection feature;

[0028] The excluded features in the computing power equipment feature set of the two users are deleted to obtain the computing power equipment features corresponding to the two users.

[0029] As an optional implementation, in the first aspect of the present invention, determining the computing power device corresponding to each of the users from the plurality of candidate devices based on a dynamic programming algorithm according to the computing power device characteristics includes:

[0030] For each of the users, calculate the average of the similarities between the computing device feature of the user and the device feature of each of the candidate devices to obtain a feature universality parameter corresponding to the user;

[0031] calculating a correction weight inversely proportional to the feature universality parameter;

[0032] Calculate the product of the similarity between the computing power device feature of the user and the device feature of each candidate device and the modified weight to obtain the device matching degree corresponding to each candidate device;

[0033] The objective function is set to maximize the device matching degree of the computing power device corresponding to each user;

[0034] The restriction condition is set as that the number of users recommended for each computing device is less than a preset number threshold;

[0035] Based on a dynamic programming algorithm, according to the objective function and the constraint conditions, a computing device corresponding to each of the users is determined from the multiple candidate devices.

[0036] A second aspect of an embodiment of the present invention discloses a computing power matching system based on artificial intelligence, the system comprising:

[0037] An acquisition module, used to acquire multiple computing task requests sent by multiple users and user parameters of each of the users;

[0038] A first determination module is used to determine a plurality of candidate devices in a preset device library according to the task parameters corresponding to each computing task request;

[0039] A second determination module is used to determine the computing power device characteristics corresponding to each of the users based on the task parameters and the user parameters according to a neural network algorithm;

[0040] The third determination module is used to determine the computing power device corresponding to each of the users from the multiple candidate devices based on a dynamic programming algorithm according to the characteristics of the computing power device.

[0041] As an optional implementation, in the second aspect of the present invention, the user parameters include at least one of user physiological characteristics, user historical computing tasks, user device parameters and user web page operations.

[0042] As an optional implementation, in the second aspect of the present invention, the task parameters include at least one of task data volume, task computing volume, task type, task data type, task purpose, task application environment and task programming language.

[0043] As an optional implementation, in the second aspect of the present invention, the first determination module determines a specific manner of selecting multiple candidate devices in a preset device library according to the task parameters corresponding to each computing power task request, including:

[0044] Determine the device performance requirements corresponding to the task parameters corresponding to each of the computing power task requests according to the correspondence between the preset task parameters and the device performance requirements;

[0045] For each standby device in the preset device library, calculate the average value of the matching degree between the standby device and the device performance requirements corresponding to all the computing power task requests to obtain the matching degree parameter of the standby device;

[0046] All the backup devices whose matching degree parameters are greater than a preset first parameter threshold are determined as candidate devices.

[0047] As an optional implementation, in the second aspect of the present invention, the second determination module determines the specific manner of the computing power device characteristics corresponding to each of the users based on the task parameters and the user parameters according to the neural network algorithm, including:

[0048] For each of the users, inputting the user parameter of the user into a trained first device preference prediction neural network to obtain the first device preference parameter of the user;

[0049] Inputting the task parameters of the user into a trained second device preference prediction neural network to obtain the second device preference parameters of the user;

[0050] Calculate the intersection of the first device preference parameter and the second device preference parameter to obtain a computing power device feature set corresponding to the user;

[0051] For any two of the users, the computing power device feature sets of the two users are modified according to the user parameters and the task parameters of the two users to obtain the computing power device features corresponding to the two users.

[0052] As an optional embodiment, in the second aspect of the present invention, the first device preference prediction neural network is a CNN neural network, which is trained by a training data set including multiple training user parameters and corresponding device preference parameter annotations; the second device preference prediction neural network is a CNN neural network, which is trained by a training data set including multiple training task parameters and corresponding device preference parameter annotations.

[0053] As an optional implementation, in the second aspect of the present invention, the second determination module modifies the computing power device feature set of the two users according to the user parameters and the task parameters of the two users to obtain the specific manner of the computing power device features corresponding to the two users, including:

[0054] For any device feature, the number of occurrences of the device feature in the computing power device feature set of the two users is calculated; the device feature is device performance, device hardware information, device model, device location or device provider information;

[0055] Calculating a first parameter similarity between the user parameters of the two users;

[0056] Calculating a second parameter similarity between the task parameters of the two users;

[0057] Calculating the reciprocal of the weighted average of the first parameter similarity and the second parameter similarity to obtain a difference parameter between the two users;

[0058] Calculating a number threshold proportional to the difference parameter;

[0059] Determine the device feature whose occurrence times are lower than the times threshold as a rejection feature;

[0060] The excluded features in the computing power equipment feature set of the two users are deleted to obtain the computing power equipment features corresponding to the two users.

[0061] As an optional implementation, in the second aspect of the present invention, the third determination module determines the specific manner of the computing power device corresponding to each of the users from the multiple candidate devices based on a dynamic programming algorithm according to the computing power device characteristics, including:

[0062] For each of the users, calculate the average of the similarities between the computing device feature of the user and the device feature of each of the candidate devices to obtain a feature universality parameter corresponding to the user;

[0063] calculating a correction weight inversely proportional to the feature universality parameter;

[0064] Calculate the product of the similarity between the computing power device feature of the user and the device feature of each candidate device and the modified weight to obtain the device matching degree corresponding to each candidate device;

[0065] The objective function is set to maximize the device matching degree of the computing power device corresponding to each user;

[0066] The restriction condition is set as that the number of users recommended for each computing device is less than a preset number threshold;

[0067] Based on a dynamic programming algorithm, according to the objective function and the constraint conditions, a computing device corresponding to each of the users is determined from the multiple candidate devices.

[0068] The third aspect of the present invention discloses another computing power matching system based on artificial intelligence, the system comprising:

[0069] A memory storing executable program code;

[0070] a processor coupled to the memory;

[0071] The processor calls the executable program code stored in the memory to execute part or all of the steps in the artificial intelligence-based computing power matching method disclosed in the first aspect of the present invention.

[0072] The fourth aspect of the present invention discloses a computer storage medium, which stores computer instructions. When the computer instructions are called, they are used to execute some or all of the steps in the artificial intelligence-based computing power matching method disclosed in the first aspect of the present invention.

[0073] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0074] The present invention can determine candidate devices based on computing power task requests sent by multiple users, and then determine the computing power device characteristics corresponding to each user based on task parameters and user parameters based on a neural network algorithm, so as to determine the computing power device corresponding to each user, thereby being able to more accurately and efficiently predict the computing power device characteristics corresponding to the user, thereby improving the recommended matching degree of the computing power device, and improving the overall computing power device utilization and user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0075] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0076] Figure 1 It is a flowchart of a computing power matching method based on artificial intelligence disclosed in an embodiment of the present invention.

[0077] Figure 2 It is a structural diagram of a computing power matching system based on artificial intelligence disclosed in an embodiment of the present invention.

[0078] Figure 3 It is a structural diagram of another artificial intelligence-based computing power matching system disclosed in an embodiment of the present invention. DETAILED DESCRIPTION

[0079] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0080] The terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, device, product or equipment that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units that are not listed, or may optionally include other steps or units that are inherent to these processes, methods, products or equipment.

[0081] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present invention. The appearance of the phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0082] The present invention discloses a computing power matching method and system based on artificial intelligence, which can determine candidate devices based on computing power task requests sent by multiple users, and then determine the computing power device characteristics corresponding to each user according to task parameters and user parameters based on a neural network algorithm, thereby determining the computing power device corresponding to each user, so as to more accurately and efficiently predict the computing power device characteristics corresponding to the user, so as to improve the recommended matching degree of the computing power device, improve the overall computing power device utilization rate and user experience. The following are detailed descriptions.

[0083] Embodiment 1

[0084] See also Figure 1 , Figure 1 : is a flow chart of a computing power matching method based on artificial intelligence disclosed in an embodiment of the present invention. Figure 1 The described AI-based computing power matching method can be applied to a data processing system / data processing device / data processing server (wherein the server includes a local processing server or a cloud processing server). Figure 1 As shown, the computing power matching method based on artificial intelligence may include the following operations:

[0085] 101. Obtain multiple computing task requests sent by multiple users and user parameters of each user.

[0086] 102. According to the task parameters corresponding to each computing task request, multiple candidate devices are determined in a preset device library.

[0087] 103. Based on the task parameters and user parameters, the computing power device characteristics corresponding to each user are determined according to the neural network algorithm.

[0088] 104. According to the characteristics of the computing power device, determine the computing power device corresponding to each user from multiple candidate devices based on a dynamic programming algorithm.

[0089] It can be seen that the above-mentioned embodiment of the invention can determine the candidate devices based on the computing power task requests sent by multiple users, and then determine the computing power device characteristics corresponding to each user based on the task parameters and user parameters based on the neural network algorithm, so as to determine the computing power device corresponding to each user, thereby being able to more accurately and efficiently predict the computing power device characteristics corresponding to the user, so as to improve the recommended matching degree of the computing power device, improve the overall computing power device utilization and user experience.

[0090] As an optional embodiment, in the above steps, the user parameters include at least one of user physiological characteristics, user historical computing tasks, user device parameters and user web page operations.

[0091] It can be seen that through the above optional embodiments, the content of user parameters is clarified, which can more accurately characterize the user's characteristics, so as to facilitate the subsequent recommendation of computing power equipment, and assist in more accurate and efficient prediction of the computing power equipment characteristics corresponding to the user, so as to improve the recommended matching degree of computing power equipment, improve the overall computing power equipment utilization and user experience.

[0092] As an optional embodiment, in the above steps, the task parameters include at least one of task data volume, task computing volume, task type, task data type, task purpose, task application environment and task programming language.

[0093] It can be seen that through the above optional embodiments, the content of the task parameters is clarified, which can more accurately characterize the task characteristics of the computing power task, so as to facilitate the subsequent recommendation of computing power equipment, and assist in more accurate and efficient prediction of the computing power equipment characteristics corresponding to the user, so as to improve the recommended matching degree of the computing power equipment, improve the overall computing power equipment utilization and user experience.

[0094] As an optional embodiment, in the above steps, according to the task parameters corresponding to each computing task request, multiple candidate devices are determined in a preset device library, including:

[0095] Determine the device performance requirements corresponding to the task parameters corresponding to each computing task request based on the correspondence between the preset task parameters and the device performance requirements;

[0096] For each standby device in the preset device library, calculate the average value of the matching degree between the standby device and the device performance requirements corresponding to all computing task requests to obtain the matching degree parameter of the standby device;

[0097] All backup devices whose matching degree parameters are greater than a preset first parameter threshold are determined as candidate devices.

[0098] It can be seen that through the above-mentioned optional embodiments, the device performance requirements corresponding to multiple task requests can be determined based on the correspondence between task parameters and device performance requirements, and then multiple candidate devices that can meet the basic task requirements can be determined based on the calculation and screening of the matching degree between the backup devices and the performance requirements, so as to facilitate the subsequent recommendation of further computing power devices, and assist in achieving more accurate and efficient prediction of the computing power device characteristics corresponding to the user, so as to improve the recommended matching degree of the computing power equipment, and improve the overall computing power equipment utilization and user experience.

[0099] As an optional embodiment, in the above steps, based on the task parameters and the user parameters, according to the neural network algorithm, determining the computing power device characteristics corresponding to each user includes:

[0100] For each user, input the user parameter of the user into the trained first device preference prediction neural network to obtain the first device preference parameter of the user;

[0101] Inputting the task parameters of the user into the trained second device preference prediction neural network to obtain the second device preference parameters of the user;

[0102] Calculate the intersection of the first device preference parameter and the second device preference parameter to obtain a computing power device feature set corresponding to the user;

[0103] For any two users, the computing power device feature sets of the two users are modified according to the user parameters and task parameters of the two users to obtain the computing power device features corresponding to the two users.

[0104] It can be seen that through the above-mentioned optional embodiments, after the two device preference prediction neural networks predict the device preferences based on user parameters and task parameters respectively, the computing power device characteristics corresponding to each user can be accurately determined based on intersection calculation and correction rules, so as to facilitate the subsequent recommendation of further computing power devices, and assist in achieving more accurate and efficient prediction of the computing power device characteristics corresponding to the user, so as to improve the recommended matching degree of the computing power equipment, improve the overall computing power equipment utilization and user experience.

[0105] As an optional embodiment, in the above steps, the first device preference prediction neural network is a CNN neural network, which is trained by a training data set including multiple training user parameters and corresponding device preference parameter annotations; the second device preference prediction neural network is a CNN neural network, which is trained by a training data set including multiple training task parameters and corresponding device preference parameter annotations.

[0106] It can be seen that through the above optional embodiments, the details and training contents of the two device preference prediction neural networks are defined to more accurately predict device preferences, assist in more accurately and efficiently predicting the computing power device characteristics corresponding to the user, so as to improve the recommended matching degree of the computing power device, improve the overall computing power device utilization and user experience.

[0107] As an optional embodiment, in the above steps, according to the user parameters and task parameters of the two users, the computing power device feature sets of the two users are modified to obtain the computing power device features corresponding to the two users, including:

[0108] For any device feature, calculate the number of occurrences of the device feature in the computing power device feature sets of the two users; optionally, the device feature is device performance, device hardware information, device model, device location or device provider information;

[0109] Calculating a first parameter similarity between the user parameters of the two users;

[0110] Calculating a second parameter similarity between the task parameters of the two users;

[0111] Calculate the reciprocal of the weighted average of the first parameter similarity and the second parameter similarity to obtain a difference parameter between the two users;

[0112] Calculate a number threshold proportional to the difference parameter;

[0113] Determine the device features whose occurrence times are lower than the number threshold as the rejection features;

[0114] The excluded features in the computing power equipment feature set of the two users are deleted to obtain the computing power equipment features corresponding to the two users.

[0115] It can be seen that through the above-mentioned optional embodiments, the number threshold can be determined by calculating the difference between users, so as to eliminate low-probability features with extremely low occurrence times in the user's computing power device feature set, so as to obtain more accurate user device feature preferences, so as to facilitate the subsequent recommendation of further computing power devices, and assist in achieving more accurate and efficient prediction of the computing power device characteristics corresponding to the user, so as to improve the recommended matching degree of the computing power equipment, improve the overall computing power equipment utilization and user experience.

[0116] As an optional embodiment, in the above steps, according to the characteristics of the computing power device, determining the computing power device corresponding to each user from multiple candidate devices based on a dynamic programming algorithm includes:

[0117] For each user, calculate the average of the similarities between the computing power device features of the user and the device features of each candidate device to obtain the feature universality parameter corresponding to the user;

[0118] Calculate the correction weight inversely proportional to the feature universality parameter;

[0119] Calculate the product of the similarity between the computing power device feature of the user and the device feature of each candidate device and the modified weight to obtain the device matching degree corresponding to each candidate device;

[0120] The objective function is set to maximize the device matching degree of the computing power device corresponding to each user;

[0121] The restriction condition is set such that the number of users recommended for each computing device is less than the preset number threshold;

[0122] Based on the dynamic programming algorithm, according to the objective function and constraints, the computing power device corresponding to each user is determined from multiple candidate devices.

[0123] It can be seen that through the above-mentioned optional embodiments, the user's correction weight can be determined by calculating the universal parameters of the general similarity of the user's computing power device characteristics, so as to more accurately correct the matching degree between each device and the user's preferences, and achieve more accurate computing power device recommendations based on the dynamic programming algorithm, thereby improving the recommended matching degree of the computing power device, and improving the overall computing power device utilization and user experience.

[0124] Embodiment 2

[0125] See also Figure 2 , Figure 2 : is a schematic diagram of a computing power matching system based on artificial intelligence disclosed in an embodiment of the present invention. Figure 2 The described AI-based computing power matching system can be applied to a data processing system / data processing device / data processing server (wherein the server includes a local processing server or a cloud processing server). Figure 2 As shown, the AI-based computing power matching system may include:

[0126] The acquisition module 201 is used to acquire multiple computing task requests sent by multiple users and user parameters of each user.

[0127] The first determination module 202 is used to determine a plurality of candidate devices in a preset device library according to the task parameters corresponding to each computing task request.

[0128] The second determination module 203 is used to determine the computing power device characteristics corresponding to each user based on the task parameters and the user parameters according to the neural network algorithm.

[0129] The third determination module 204 is used to determine the computing power device corresponding to each user from multiple candidate devices based on a dynamic programming algorithm according to the characteristics of the computing power device.

[0130] It can be seen that the above-mentioned embodiment of the invention can determine the candidate devices based on the computing power task requests sent by multiple users, and then determine the computing power device characteristics corresponding to each user based on the task parameters and user parameters based on the neural network algorithm, so as to determine the computing power device corresponding to each user, thereby being able to more accurately and efficiently predict the computing power device characteristics corresponding to the user, so as to improve the recommended matching degree of the computing power device, improve the overall computing power device utilization and user experience.

[0131] As an optional embodiment, the user parameters include at least one of user physiological characteristics, user historical computing tasks, user device parameters and user web page operations.

[0132] It can be seen that through the above optional embodiments, the content of user parameters is clarified, which can more accurately characterize the user's characteristics, so as to facilitate the subsequent recommendation of computing power equipment, and assist in more accurate and efficient prediction of the computing power equipment characteristics corresponding to the user, so as to improve the recommended matching degree of computing power equipment, improve the overall computing power equipment utilization and user experience.

[0133] As an optional embodiment, the task parameters include at least one of task data volume, task computing volume, task type, task data type, task purpose, task application environment and task programming language.

[0134] It can be seen that through the above optional embodiments, the content of the task parameters is clarified, which can more accurately characterize the task characteristics of the computing power task, so as to facilitate the subsequent recommendation of computing power equipment, and assist in more accurate and efficient prediction of the computing power equipment characteristics corresponding to the user, so as to improve the recommended matching degree of the computing power equipment, improve the overall computing power equipment utilization and user experience.

[0135] As an optional embodiment, the first determination module determines a specific manner of selecting multiple candidate devices from a preset device library according to the task parameters corresponding to each computing task request, including:

[0136] Determine the device performance requirements corresponding to the task parameters corresponding to each computing task request based on the correspondence between the preset task parameters and the device performance requirements;

[0137] For each standby device in the preset device library, calculate the average value of the matching degree between the standby device and the device performance requirements corresponding to all computing task requests to obtain the matching degree parameter of the standby device;

[0138] All backup devices whose matching degree parameters are greater than a preset first parameter threshold are determined as candidate devices.

[0139] It can be seen that through the above-mentioned optional embodiments, the device performance requirements corresponding to multiple task requests can be determined based on the correspondence between task parameters and device performance requirements, and then multiple candidate devices that can meet the basic task requirements can be determined based on the calculation and screening of the matching degree between the backup devices and the performance requirements, so as to facilitate the subsequent recommendation of further computing power devices, and assist in achieving more accurate and efficient prediction of the computing power device characteristics corresponding to the user, so as to improve the recommended matching degree of the computing power equipment, and improve the overall computing power equipment utilization and user experience.

[0140] As an optional embodiment, the second determination module determines the specific manner of the computing power device characteristics corresponding to each user based on the task parameters and the user parameters according to the neural network algorithm, including:

[0141] For each user, input the user parameter of the user into the trained first device preference prediction neural network to obtain the first device preference parameter of the user;

[0142] Inputting the task parameters of the user into the trained second device preference prediction neural network to obtain the second device preference parameters of the user;

[0143] Calculate the intersection of the first device preference parameter and the second device preference parameter to obtain a computing power device feature set corresponding to the user;

[0144] For any two users, the computing power device feature sets of the two users are modified according to the user parameters and task parameters of the two users to obtain the computing power device features corresponding to the two users.

[0145] It can be seen that through the above-mentioned optional embodiments, after the two device preference prediction neural networks predict the device preferences based on user parameters and task parameters respectively, the computing power device characteristics corresponding to each user can be accurately determined based on intersection calculation and correction rules, so as to facilitate the subsequent recommendation of further computing power devices, and assist in achieving more accurate and efficient prediction of the computing power device characteristics corresponding to the user, so as to improve the recommended matching degree of the computing power equipment, improve the overall computing power equipment utilization and user experience.

[0146] As an optional embodiment, the first device preference prediction neural network is a CNN neural network, which is trained by a training data set including multiple training user parameters and corresponding device preference parameter annotations; the second device preference prediction neural network is a CNN neural network, which is trained by a training data set including multiple training task parameters and corresponding device preference parameter annotations.

[0147] It can be seen that through the above optional embodiments, the details and training contents of the two device preference prediction neural networks are defined to more accurately predict device preferences, assist in more accurately and efficiently predicting the computing power device characteristics corresponding to the user, so as to improve the recommended matching degree of the computing power device, improve the overall computing power device utilization and user experience.

[0148] As an optional embodiment, the second determination module modifies the computing power device feature set of the two users according to the user parameters and task parameters of the two users to obtain the specific manner of the computing power device features corresponding to the two users, including:

[0149] For any device feature, calculate the number of occurrences of the device feature in the computing power device feature sets of the two users; optionally, the device feature is device performance, device hardware information, device model, device location or device provider information;

[0150] Calculating a first parameter similarity between the user parameters of the two users;

[0151] Calculating a second parameter similarity between the task parameters of the two users;

[0152] Calculate the reciprocal of the weighted average of the first parameter similarity and the second parameter similarity to obtain a difference parameter between the two users;

[0153] Calculate a number threshold proportional to the difference parameter;

[0154] Determine the device features whose occurrence times are lower than the number threshold as the rejection features;

[0155] The excluded features in the computing power equipment feature set of the two users are deleted to obtain the computing power equipment features corresponding to the two users.

[0156] It can be seen that through the above-mentioned optional embodiments, the number threshold can be determined by calculating the difference between users, so as to eliminate low-probability features with extremely low occurrence times in the user's computing power device feature set, so as to obtain more accurate user device feature preferences, so as to facilitate the subsequent recommendation of further computing power devices, and assist in achieving more accurate and efficient prediction of the computing power device characteristics corresponding to the user, so as to improve the recommended matching degree of the computing power equipment, improve the overall computing power equipment utilization and user experience.

[0157] As an optional embodiment, the third determination module determines the specific manner of the computing power device corresponding to each user from multiple candidate devices based on a dynamic programming algorithm according to the computing power device characteristics, including:

[0158] For each user, calculate the average of the similarities between the computing power device features of the user and the device features of each candidate device to obtain the feature universality parameter corresponding to the user;

[0159] Calculate the correction weight inversely proportional to the feature universality parameter;

[0160] Calculate the product of the similarity between the computing power device feature of the user and the device feature of each candidate device and the modified weight to obtain the device matching degree corresponding to each candidate device;

[0161] The objective function is set to maximize the device matching degree of the computing power device corresponding to each user;

[0162] The restriction condition is set such that the number of users recommended for each computing device is less than the preset number threshold;

[0163] Based on the dynamic programming algorithm, according to the objective function and constraints, the computing power device corresponding to each user is determined from multiple candidate devices.

[0164] It can be seen that through the above-mentioned optional embodiments, the user's correction weight can be determined by calculating the universal parameters of the general similarity of the user's computing power device characteristics, so as to more accurately correct the matching degree between each device and the user's preferences, and achieve more accurate computing power device recommendations based on the dynamic programming algorithm, thereby improving the recommended matching degree of the computing power device, and improving the overall computing power device utilization and user experience.

[0165] Embodiment 3

[0166] See also Figure 3 , Figure 3 It is another computing power matching system based on artificial intelligence disclosed in the embodiment of the present invention. Figure 3 The described AI-based computing power matching system is applied to a data processing system / data processing device / data processing server (wherein the server includes a local processing server or a cloud processing server). Figure 3 As shown, the AI-based computing power matching system may include:

[0167] A memory 301 storing executable program codes;

[0168] a processor 302 coupled to the memory 301;

[0169] The processor 302 calls the executable program code stored in the memory 301 to execute the steps of the artificial intelligence-based computing power matching method described in the first embodiment.

[0170] Embodiment 4

[0171] An embodiment of the present invention discloses a computer-readable storage medium storing a computer program for electronic data exchange, wherein the computer program enables a computer to execute the steps of the artificial intelligence-based computing power matching method described in Embodiment 1.

[0172] Embodiment 5

[0173] An embodiment of the present invention discloses a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to enable a computer to execute the steps of the artificial intelligence-based computing power matching method described in Example 1.

[0174] The above describes specific embodiments of the present specification, and other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the accompanying drawings do not necessarily have to be performed in the specific order or sequential order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0175] The systems, devices, modules or units described in the above embodiments may be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, the computer may be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

[0176] For the convenience of description, the above device is described in various units according to their functions. Of course, when implementing this specification, the functions of each unit can be implemented in the same or multiple software and / or hardware.

[0177] Those skilled in the art will appreciate that the embodiments of this specification may be provided as methods, systems, or computer program products. Therefore, the embodiments of this specification may be in the form of complete hardware embodiments, complete software embodiments, or embodiments in combination with software and hardware. Moreover, the embodiments of this specification may be in the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0178] This specification is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of this specification. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0179] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0180] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0181] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0182] The memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0183] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0184] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0185] This specification may be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. This specification may also be practiced in distributed computing environments where tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules may be located in local and remote computer storage media, including storage devices.

[0186] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0187] Finally, it should be noted that the artificial intelligence-based computing power matching method and system disclosed in the embodiments of the present invention only disclose the preferred embodiments of the present invention, which are only used to illustrate the technical solution of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments can still be modified, or some of the technical features therein can be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A computing power matching method based on artificial intelligence, characterized in that: The method comprises: Obtaining multiple computing task requests sent by multiple users and user parameters of each of the users; According to the task parameters corresponding to each of the computing task requests, a plurality of candidate devices are determined in a preset device library; Based on the task parameters and the user parameters, the computing power device characteristics corresponding to each user are determined according to a neural network algorithm, including: For each of the users, inputting the user parameter of the user into a trained first device preference prediction neural network to obtain the first device preference parameter of the user; Inputting the task parameters of the user into the trained second device preference prediction neural network to obtain the second device preference parameters of the user; the first device preference prediction neural network is a CNN neural network, which is trained by a training data set including a plurality of training user parameters and corresponding device preference parameter annotations; the second device preference prediction neural network is a CNN neural network, which is trained by a training data set including a plurality of training task parameters and corresponding device preference parameter annotations; Calculate the intersection of the first device preference parameter and the second device preference parameter to obtain a computing power device feature set corresponding to the user; For any two of the users, for any device feature, calculate the number of occurrences of the device feature in the computing power device feature set of the two users; the device feature is device performance, device hardware information, device model, device location or device provider information; Calculating a first parameter similarity between the user parameters of the two users; Calculating a second parameter similarity between the task parameters of the two users; Calculating the reciprocal of the weighted average of the first parameter similarity and the second parameter similarity to obtain a difference parameter between the two users; Calculating a number threshold proportional to the difference parameter; Determine the device feature whose occurrence times are lower than the times threshold as a rejection feature; Deleting the removed features from the computing power equipment feature set of the two users to obtain computing power equipment features corresponding to the two users; According to the computing power device characteristics, determining the computing power device corresponding to each of the users from the multiple candidate devices based on a dynamic programming algorithm includes: For each of the users, calculate the average of the similarities between the computing device feature of the user and the device feature of each of the candidate devices to obtain a feature universality parameter corresponding to the user; calculating a correction weight inversely proportional to the feature universality parameter; Calculate the product of the similarity between the computing power device feature of the user and the device feature of each candidate device and the modified weight to obtain the device matching degree corresponding to each candidate device; The objective function is set to maximize the device matching degree of the computing power device corresponding to each user; The restriction condition is set as that the number of users recommended for each computing device is less than a preset number threshold; Based on a dynamic programming algorithm, according to the objective function and the constraint conditions, a computing device corresponding to each of the users is determined from the multiple candidate devices.

2. The computing power matching method based on artificial intelligence according to claim 1, characterized in that: The user parameters include at least one of user physiological characteristics, user historical computing tasks, user device parameters and user web page operations.

3. The computing power matching method based on artificial intelligence according to claim 2 is characterized in that: The task parameters include at least one of task data volume, task computing volume, task type, task data type, task purpose, task application environment and task programming language.

4. The computing power matching method based on artificial intelligence according to claim 3 is characterized in that: According to the task parameters corresponding to each computing task request, a plurality of candidate devices are determined in a preset device library, including: Determine the device performance requirements corresponding to the task parameters corresponding to each of the computing power task requests according to the correspondence between the preset task parameters and the device performance requirements; For each standby device in the preset device library, calculate the average value of the matching degree between the standby device and the device performance requirements corresponding to all the computing power task requests to obtain the matching degree parameter of the standby device; All the backup devices whose matching degree parameters are greater than a preset first parameter threshold are determined as candidate devices.

5. A computing power matching system based on artificial intelligence, characterized in that: The system comprises: An acquisition module, used to acquire multiple computing task requests sent by multiple users and user parameters of each of the users; A first determination module is used to determine a plurality of candidate devices in a preset device library according to the task parameters corresponding to each computing task request; The second determination module is used to determine the computing power device characteristics corresponding to each of the users based on the task parameters and the user parameters according to a neural network algorithm, including: For each of the users, inputting the user parameter of the user into a trained first device preference prediction neural network to obtain the first device preference parameter of the user; Inputting the task parameters of the user into the trained second device preference prediction neural network to obtain the second device preference parameters of the user; the first device preference prediction neural network is a CNN neural network, which is trained by a training data set including a plurality of training user parameters and corresponding device preference parameter annotations; the second device preference prediction neural network is a CNN neural network, which is trained by a training data set including a plurality of training task parameters and corresponding device preference parameter annotations; Calculate the intersection of the first device preference parameter and the second device preference parameter to obtain a computing power device feature set corresponding to the user; For any two of the users, for any device feature, calculate the number of occurrences of the device feature in the computing power device feature set of the two users; the device feature is device performance, device hardware information, device model, device location or device provider information; Calculating a first parameter similarity between the user parameters of the two users; Calculating a second parameter similarity between the task parameters of the two users; Calculating the reciprocal of the weighted average of the first parameter similarity and the second parameter similarity to obtain a difference parameter between the two users; Calculating a number threshold proportional to the difference parameter; Determine the device feature whose occurrence times are lower than the times threshold as a rejection feature; Deleting the removed features from the computing power equipment feature set of the two users to obtain computing power equipment features corresponding to the two users; The third determination module is used to determine the computing power device corresponding to each of the users from the plurality of candidate devices based on a dynamic programming algorithm according to the computing power device characteristics, including: For each of the users, calculate the average of the similarities between the computing device feature of the user and the device feature of each of the candidate devices to obtain a feature universality parameter corresponding to the user; calculating a correction weight inversely proportional to the feature universality parameter; Calculate the product of the similarity between the computing power device feature of the user and the device feature of each candidate device and the modified weight to obtain the device matching degree corresponding to each candidate device; The objective function is set to maximize the device matching degree of the computing power device corresponding to each user; The restriction condition is set as that the number of users recommended for each computing device is less than a preset number threshold; Based on a dynamic programming algorithm, according to the objective function and the constraint conditions, a computing device corresponding to each of the users is determined from the multiple candidate devices.

6. A computing power matching system based on artificial intelligence, characterized in that: The system comprises: A memory storing executable program code; a processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the artificial intelligence-based computing power matching method as described in any one of claims 1-4.

Images