Resource allocation method and apparatus, electronic device, and computer readable medium

Abstract

Embodiments of the present disclosure disclose resource allocation methods, devices, electronic equipment and computer readable media. A specific embodiment of the method comprises: in response to receiving a storage space adjustment instruction, obtaining a first institutional user data set and a second institutional user data set; determining the first user overlap information corresponding to each target institutional identifier group based on the first institutional user data set; updating the first institutional user data group set based on the first user overlap information to obtain an updated first institutional user data group set; generating an updated second institutional user data group set based on the second institutional user data set; generating an institutional hierarchical information set; detecting each institutional storage space usage information to obtain an institutional allocable storage space information set; generating an institutional expansion information set and an institutional contraction information set; and controlling a database server to adjust the storage space capacity of each institution. The embodiment can reduce the excessive occupation and waste of storage resources.

Description

TECHNICAL FIELD

[0001] Embodiments of the present disclosure relate to the technical field of computer, and specifically to a resource allocation method and device, an electronic device and a computer readable medium. BACKGROUND

[0002] Resource allocation plays an important role in balancing the database storage space of each institution. At present, when performing resource allocation, the commonly used way is that the database server divides the storage space for each institution by random allocation or average allocation to meet the user data storage needs of the institution, and when the storage space of an institution is insufficient, the database server will redivide the storage space to expand the storage space of the institution with insufficient storage space to complete resource allocation.

[0003] However, in practice, it is found that when the above way is used for resource allocation, the following technical problems often exist:

[0004] Due to the differences between the user groups of each institution, the storage space capacity required by each institution also has great differences. When the storage space capacity required by an institution is small and the unused storage space is large, it is easy to cause waste of storage resources; when the storage space required by an institution is large and the storage space capacity is insufficient, if the above redivision way is used for storage space expansion, it will lead to excessive occupation of storage resources.

[0005] The above information disclosed in the background section is only intended to enhance the understanding of the background of the present disclosure concept, and therefore, it can contain information that does not form the prior art known to those of ordinary skill in the art. SUMMARY

[0006] The summary section is provided to introduce the concepts briefly in a simplified form, which will be described in detail in the specific embodiments section. The summary section is not intended to identify key or essential features of the claimed technology nor is it intended to be used to limit the scope of the claimed technology.

[0007] Some embodiments of the present disclosure propose a resource allocation method and device, an electronic device and a computer readable medium to solve one or more of the technical problems mentioned in the background section.

[0008] In a first aspect, some embodiments of the present disclosure provide a resource allocation method, comprising: in response to confirming that the database storage space for storing the user data of the institution is insufficient, sending a database storage space adjustment request to a server; in response to receiving the storage space adjustment instruction sent by the server, obtaining a first institution user data set and a second institution user data set, wherein each first institution user data in the first institution user data set is log data when the institution calls a preset business product with user information as a query primary key, and each second institution user data in the second institution user data set is user behavior label data collected by the institution; based on the first institution user data set, determining the first user overlap information corresponding to each target institution identification group in the target institution identification group set, wherein each target institution identification group in the target institution identification group set is composed of two different preset institution identification groups; based on the obtained first user overlap information, updating the first institution user data group set to obtain an updated first institution user data group set; based on the second institution user data set, generating an updated second institution user data group set; based on the updated first institution user data group set and the updated second institution user data group set, generating an institution hierarchical information set; detecting each institution storage space usage information in the pre-obtained institution storage space usage information set to obtain an institution allocatable storage space information set; based on the institution hierarchical information set and the institution allocatable storage space information set, generating an institution expansion information set and an institution contraction information set; based on the institution expansion information set and the institution contraction information set, controlling the associated database server to adjust the storage space capacity of each institution.

[0009] In a second aspect, some embodiments of the present disclosure provide a resource allocation apparatus, the apparatus comprising: a sending unit configured to send, in response to confirming that a database storage space for storing institution user data is insufficient, a database storage space adjustment request to a server; an obtaining unit configured to obtain, in response to receiving a storage space adjustment instruction sent by the server, a first institution user data set and a second institution user data set, wherein each first institution user data in the first institution user data set is log data when an institution calls user information as a query primary key to invoke a preset business product, and each second institution user data in the second institution user data set is user behavior label data collected by the institution; a determining unit configured to determine, based on the first institution user data set, first user overlap information corresponding to each target institution identifier group in a target institution identifier group set, wherein each target institution identifier group in the target institution identifier group set is composed of two different preset institution identifier groups in a preset institution identifier group; an updating processing unit configured to update the first institution user data group set based on the obtained first user overlap information, to obtain an updated first institution user data group set; a first generating unit configured to generate an updated second institution user data group set based on the second institution user data set; a second generating unit configured to generate an institution hierarchical information set based on the updated first institution user data group set and the updated second institution user data group set; a detection processing unit configured to detect each institution storage space usage information in a pre-obtained institution storage space usage information set, to obtain an institution allocable storage space information set; the second generating unit is configured to generate an institution expansion information set and an institution contraction information set based on the institution hierarchical information set and the institution allocable storage space information set; and a control unit configured to control a database server associated to adjust the storage space capacity of each institution based on the institution expansion information set and the institution contraction information set.

[0010] In a third aspect, some embodiments of the present disclosure provide an electronic device, comprising: one or more processors; and a storage device having one or more programs stored thereon, wherein the one or more programs, when executed by the one or more processors, cause the one or more processors to implement the method described in any implementation manner of the first aspect.

[0011] In a fourth aspect, some embodiments of the present disclosure provide a computer readable medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the method described in any implementation manner of the first aspect.

[0012] The above various embodiments of the present disclosure have the following beneficial effects: through the resource allocation method of some embodiments of the present disclosure, the excessive occupation and waste of storage resources can be reduced. Specifically, the reason for the waste or excessive occupation of storage resources is that due to the differences between the user groups of various institutions, the storage space capacities required by various institutions also have great differences. When the required storage space of an institution is small and there is a lot of unused storage space, it is easy to cause waste of storage resources. When the required storage space of an institution is large and the storage space capacity is insufficient, if the above-mentioned re-division method is used for storage space expansion, it will lead to excessive occupation of storage resources. Based on this, the resource allocation method of some embodiments of the present disclosure first, in response to confirming that the database storage space for storing institution user data is insufficient, sends a database storage space adjustment request to the server. Secondly, in response to receiving the storage space adjustment instruction sent by the server, the first institution user data set and the second institution user data set are obtained. Each first institution user data in the first institution user data set is the log data when the institution calls the preset business product with user information as the query primary key, and each second institution user data in the second institution user data set is the user behavior label data collected by the institution. Thus, two different source types of sample data sets corresponding to each institution can be obtained. Then, based on the first institution user data set, the first user overlap information corresponding to each target institution identification group in the target institution identification group set is determined, wherein each target institution identification group in the target institution identification group set is composed of two different preset institution identification groups in the preset institution identification group. Thus, the information of the user group overlap between different institutions corresponding to the product call log record can be determined. Then, based on the obtained first user overlap information, the first institution user data group set is updated to obtain an updated first institution user data group set. Thus, according to the user overlap sample data between different institutions, the original sample data of each institution can be supplemented based on the original sample data of each institution to make up for the lack of original sample data. Next, based on the second institution user data set, an updated second institution user data group set is generated. Thus, the original sample data of the institution corresponding to the user behavior can be supplemented to make up for the lack of original sample data. Then, based on the updated first institution user data group set and the updated second institution user data group set, an institution classification information set is generated. Thus, the level information of each institution can be obtained. Then, each institution storage space usage information in the pre-obtained institution storage space usage information set is detected to obtain an institution allocatable storage space information set. Thus, the allocatable state of the current storage space capacity of each institution can be obtained.Further, based on the above-mentioned institution hierarchical information set and the above-mentioned institution allocable storage space information set, an institution expansion information set and an institution contraction information set are generated. Thus, the resource information of each institution to be expanded and the resource information of each institution to be contracted can be obtained. Finally, based on the above-mentioned institution expansion information set and the above-mentioned institution contraction information set, the associated database server is controlled to adjust the storage space capacity of each institution. Therefore, the resource allocation method of some embodiments of the present disclosure can determine the level of an institution according to the customer information of the institution when the storage space of the institution is insufficient, and on this basis, determine the institution expansion information set and the institution contraction information set according to the institution hierarchical information and the current resource allocable state of the institution, so as to contract the institution with less storage space capacity and more unused storage space, and use the contracted capacity to expand the storage space of the institution which needs to use more storage space and has insufficient storage space capacity. Thus, the excessive occupation and waste of storage resources can be reduced. BRIEF DESCRIPTION OF DRAWINGS

[0013] The above and other features, advantages and aspects of embodiments of the present disclosure will become more apparent by describing in detail some embodiments thereof with reference to the attached drawings. In the drawings:

[0014] Figure 1 is a flowchart of some embodiments of the resource allocation method according to the present disclosure;

[0015] Figure 2 is a structural schematic diagram of some embodiments of the resource allocation apparatus according to the present disclosure;

[0016] Figure 3 is a structural schematic diagram of an electronic device suitable for implementing some embodiments of the present disclosure. DETAILED DESCRIPTION

[0017] Embodiments of the present disclosure will be described below in more detail with reference to the accompanying drawings. Although some embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure can be implemented in various forms, and should not be interpreted as being limited to the embodiments set forth herein. On the contrary, these embodiments are provided to make the present disclosure more thorough and complete. It should be understood that the drawings and embodiments of the present disclosure are only for exemplary purposes, and are not intended to limit the scope of protection of the present disclosure.

[0018] It should also be noted that, for the convenience of description, only the parts related to the invention are shown in the drawings. The embodiments in the present disclosure and the features in the embodiments can be combined with each other without conflict.

[0019] It should be noted that the terms "first", "second", and the like mentioned in the present disclosure are only used to distinguish different devices, modules or units, and are not intended to limit the order or interdependence of the functions performed by these devices, modules or units.

[0020] It should be noted that the terms "one", "multiple" mentioned in the present disclosure are illustrative and not restrictive, and those skilled in the art should understand that unless otherwise explicitly indicated in the context, it should be understood as "one or more".

[0021] The names of the messages or information exchanged between the devices in the embodiments of the present disclosure are only for illustrative purposes, and are not intended to limit the scope of the messages or information.

[0022] The collection, storage, use, etc. of log data and user behavior label data involved in the present disclosure, before performing the corresponding operation, the relevant organization or individual shall fulfill the obligations including carrying out personal information security impact assessment, fulfilling the notification obligation to the personal information subject, obtaining the prior authorization consent of the personal information subject, etc.

[0023] The present disclosure will be described in detail below with reference to the accompanying drawings and in conjunction with embodiments.

[0024] Figure 1 Flow 100 showing some embodiments of the resource allocation method according to the present disclosure. The resource allocation method comprises the following steps:

[0025] Step 101, in response to confirming that the database storage space for storing the user data of the institution is insufficient, sending a database storage space adjustment request to the server.

[0026] In some embodiments, the subject (e.g. computing device) of the resource allocation method can send a database storage space adjustment request to the server in response to confirming that the database storage space for storing the user data of the institution is insufficient. Wherein the above-mentioned database storage space adjustment request can be a network request for confirming whether to adjust the database storage space.

[0027] Step 102, in response to receiving the storage space adjustment instruction sent by the server, obtaining the first institution user data set and the second institution user data set.

[0028] In some embodiments, the execution subject can obtain the first institution user data set and the second institution user data set from the database through wired connection or wireless connection in response to receiving the storage space adjustment instruction sent by the server. The storage space adjustment instruction can be an instruction for adjusting the storage space of the database. Each first institution user data in the first institution user data set can be log data when the institution calls a preset business product with user information as a query primary key. Each second institution user data in the second institution user data set can be user behavior label data collected by the institution. The institution can be an institution in a preset industry. The preset industry can be a pre-set industry. For example, the preset industry can be the financial industry, and the institution can be a bank. The preset business product in each preset business product can be a pre-set business product called by the institution. For example, the business product can be, but is not limited to, one of the following: credit value evaluation product, identity verification product. The user information can include user identification and user mobile phone number. The user identification can be a unique identification of the user. For example, the user identification can be the user's ID card number. Each first institution user data in the first institution user data set can include, but is not limited to, user identification, user mobile phone number, product identification, product calling time, calling institution identification. The product identification can be a unique identification of the business product. The product calling time can be the time of calling the business product. The calling institution identification can be a unique identification of the calling institution. The calling institution can be the institution that calls the business product. In addition, the user behavior label data can include, but is not limited to, user identification, user mobile phone number, behavior label, associated institution identification, and backtracking time. The behavior label can represent whether the user repays on time. When the behavior label is 1, it means that the user fails to repay on time; when the behavior label is 0, it means that the user repays on time. The associated institution identification can be a unique identification of the associated institution. The associated institution can be the institution that provides the user behavior label data. The backtracking time can be a time range for evaluating the user's overdue state. For example, the backtracking time can be the past 3 months.

[0029] It should be noted that the wireless connection mode can include, but is not limited to, 3G / 4G connection, WiFi connection, Bluetooth connection, WiMAX connection, Zigbee connection, UWB (ultra wideband) connection, and other now known or future developed wireless connection modes.

[0030] In step 103, based on the first institution user data set, determine the first user overlap information corresponding to each target institution identification group in the target institution identification group set.

[0031] In some embodiments, the execution subject can determine the first user overlap information corresponding to each target institution identification group in the target institution identification group set based on the first institution user data set in various ways. Each target institution identification group in the target institution identification group set can be composed of two different preset institution identification groups in the preset institution identification group. The preset institution identification in the preset institution identification group can be the identification of a preset institution. The first user overlap information can be information about the degree of overlap of users between the corresponding two calling institutions.

[0032] In some optional implementations of some embodiments, the execution subject can determine the first user overlap information corresponding to each target institution identification group in the target institution identification group set based on the first institution user data set by the following steps:

[0033] Firstly, each first institution user data in the first institution user data set is classified to obtain a first institution user data group set. Each first institution user data group in the first institution user data group set can correspond to a preset institution identification in the preset institution identification group one by one. Each first institution user data group in the first institution user data group set can be a collection of various first institution user data corresponding to the same calling institution. Each first institution user data in the first institution user data set can be classified according to the calling institution identification included in the first institution user data to obtain the first institution user data group set.

[0034] Secondly, based on the first institution user data group set, the total number of users and the first institution user identification group set are generated. The total number of users can be the number of all users corresponding to the first institution user data group set. Each first institution user identification group in the first institution user identification group set can correspond to a preset institution identification in the preset institution identification group one by one. The first institution user identification group in the first institution user identification group set can be a collection of various user identifications corresponding to the same calling institution. Firstly, each user identification included in the first institution user data group set and not repeated is determined as a total user identification set. Then, the number of each total user identification in the total user identification set is determined as the total number of users. After that, for each first institution user data group in the first institution user data group set, each user identification included in the first institution user data group is determined as a first institution user identification group.

[0035] Thirdly, for each preset institution identification in the preset institution identification group, the following steps are performed:

[0036] A first sub-step is to select a first institution user identification group matching the preset institution identification from the first institution user identification group set.

[0037] A second sub-step is to determine the number of each first institution user identification in the selected first institution user identification group as a first institution user number.

[0038] A third sub-step is to determine the ratio of the first institution user number to the total user number as an institution user proportion.

[0039] A fourth step is to, in response to determining that each institution user proportion corresponding to the preset institution identification group meets a preset user proportion condition, perform two-by-two grouping processing on the preset institution identification in the preset institution identification group to obtain a target institution identification group set. The preset user proportion condition can be that each institution user proportion in the institution user proportion is less than a corresponding target institution user proportion. Each institution user proportion corresponds to a unique target institution user proportion. The target institution user proportion can be the ratio of the number of users of an institution to the total number of users in the market.

[0040] A fifth step is to, for each target institution identification group in the target institution identification group set, perform matching processing on the two first institution user identification groups corresponding to the target institution identification group to obtain first user overlap information. For each target institution identification group in the target institution identification group set, the two first institution user identification groups corresponding to the target institution identification group can be matched to obtain first user overlap information by the following steps:

[0041] A first sub-step is to determine the intersection between the two first institution user identification groups corresponding to the target institution identification group as an overlapping first institution user identification group.

[0042] A second sub-step is to determine the number of each overlapping first institution user identification in the overlapping first institution user identification group as a first user overlap number.

[0043] A third sub-step is to determine the target institution identification group, the overlapping first institution user identification group, and the first user overlap number as the first user overlap information.

[0044] Step 104 is to update the first institution user data group set based on the obtained first user overlap information to obtain an updated first institution user data group set.

[0045] In some embodiments, the execution subject can update the first institutional user data group set based on the obtained first user coincidence information, and obtain an updated first institutional user data group set. The updated first institutional user data group in the updated first institutional user data group set can correspond to the first institutional user data group in the first institutional user data group set. The updated first institutional user data group in the updated first institutional user data group set can be the first institutional user data group after adding the institutional user data.

[0046] In some optional implementations of some embodiments, each of the obtained first user coincidence information can include a first user coincidence number. The execution subject can update the first institutional user data group set based on the obtained first user coincidence information, and obtain an updated first institutional user data group set by the following steps:

[0047] First, for each of the obtained first user coincidence information, the following steps are performed:

[0048] First, determine the first preset institutional identifier and the second preset institutional identifier corresponding to the first user coincidence information. First, one of the target institutional identifier group corresponding to the first user coincidence information can be determined as the first preset institutional identifier. Then, another preset institutional identifier in the target institutional identifier group corresponding to the first user coincidence information that is different from the first preset institutional identifier is determined as the second preset institutional identifier.

[0049] Second, determine the first sampling ratio between the first user coincidence number included in the first user coincidence information and the number of institutional users corresponding to the first preset institutional identifier.

[0050] Third, determine the second sampling ratio between the first user coincidence number included in the first user coincidence information and the number of institutional users corresponding to the second preset institutional identifier.

[0051] Fourth sub-step, based on the first sampling ratio, the first preset mechanism identification corresponding first mechanism user data group is processed to obtain the first sampling mechanism user data group, and the first sampling mechanism user data group and the second preset mechanism identification are determined as the first to be supplemented mechanism user data. Wherein, the first sampling mechanism user data in the first sampling mechanism user data group can be randomly sampled first mechanism user data. First, the product of the first sampling ratio and the first mechanism user number corresponding to the first preset mechanism identification is determined as the first value. Then, the first mechanism user data of the first value is randomly sampled from the first mechanism user data group corresponding to the first preset mechanism identification, and each obtained to-be-sampled first mechanism user data is determined as the first sampling mechanism user data, and the first sampling mechanism user data group is obtained.

[0052] Fifth sub-step, based on the second sampling ratio, the second preset mechanism identification corresponding first mechanism user data group is processed to obtain the second sampling mechanism user data group, and the second sampling mechanism user data group and the first preset mechanism identification are determined as the second to be supplemented mechanism user data. Wherein, the second sampling mechanism user data in the second sampling mechanism user data group can be randomly sampled first mechanism user data. First, the product of the second sampling ratio and the first mechanism user number corresponding to the second preset mechanism identification is determined as the second value. Then, the first mechanism user data of the second value is randomly sampled from the first mechanism user data group corresponding to the second preset mechanism identification, and each obtained first mechanism user data is determined as the second sampling mechanism user data, and the second sampling mechanism user data group is obtained.

[0053] Second step, based on the obtained each first to be supplemented mechanism user data and each second to be supplemented mechanism user data, the first mechanism user data group set is updated to obtain the updated first mechanism user data group set. For each first mechanism user data group in the first mechanism user data group set, the following steps are executed:

[0054] First sub-step, the mechanism identification corresponding to the first mechanism user data group is determined as the to-be-supplemented data mechanism identification.

[0055] Second sub-step, the first to be supplemented mechanism user data including the second preset mechanism identification same as the to-be-supplemented data mechanism identification is selected from each first to be supplemented mechanism user data as target first to be supplemented mechanism user data, and the target first to be supplemented mechanism user data set is obtained.

[0056] A third sub-step is to select, from the second to-be-supplemented institution user data, second to-be-supplemented institution user data including a first preset institution identifier identical to the to-be-supplemented data institution identifier as target second to-be-supplemented institution user data, to obtain a target second to-be-supplemented institution user data set.

[0057] A fourth sub-step is to aggregate the target first to-be-supplemented institution user data set, the target second to-be-supplemented institution user data set, and the first institution user data group, to obtain an updated first institution user data group. First, each call institution identifier in the target first to-be-supplemented institution user data set and each call institution identifier in the target second to-be-supplemented institution user data set are replaced by the to-be-supplemented data institution identifier, and each target second to-be-supplemented institution user data after replacement and each target second to-be-supplemented institution user data are determined as updated institution user data, to obtain an updated institution user data set. Then, each updated institution user data and each first institution user data are determined as updated first institution user data, to obtain an updated first institution user data group.

[0058] It should be noted that for each to-be-classified institution, the execution subject can generate an updated first institution user data group by adding new data to the first institution user data group corresponding to the institution, so as to increase the data sample size of the institution, and facilitate more comprehensive and accurate classification and evaluation of the institution in the future.

[0059] Step 105 is to generate an updated second institution user data group set based on the second institution user data set.

[0060] In some embodiments, the execution subject can generate an updated second institution user data group set based on the second institution user data set in various ways. The updated second institution user data group in the updated second institution user data group set can correspond one-to-one to the second institution user data group in the second institution user data group set. The updated second institution user data group in the updated second institution user data group set can be a second institution user data group after adding new data.

[0061] In some optional implementations of some embodiments, the execution subject can generate an updated second institution user data group set based on the second institution user data set by the following steps:

[0062] In a first step, each second-institution user data in the second-institution user data set is classified to obtain a second-institution user data group set. Each second-institution user data group in the second-institution user data group set can correspond to a preset-institution identifier in the preset-institution identifier group. Each second-institution user data group in the second-institution user data group set can include second-institution user data corresponding to the same associated institution. The second-institution user data in the second-institution user data group set can be classified according to the associated institution identifier included in the second-institution user data.

[0063] In a second step, a second-institution user identifier group set is generated based on the second-institution user data group set. Each second-institution user identifier group in the second-institution user identifier group set can correspond to a preset-institution identifier in the preset-institution identifier group. Each second-institution user identifier group in the second-institution user identifier group set can include identifiers of users corresponding to the same associated institution. For each second-institution user data group in the second-institution user data group set, each user identifier included in the second-institution user data group can be determined as a second-institution user identifier group.

[0064] In a third step, second user overlap information corresponding to each target-institution identifier group in the target-institution identifier group set is determined. The second user overlap information can be information about the degree of overlap of users between two associated institutions. For each target-institution identifier group in the target-institution identifier group set, the second user overlap information can be generated by performing the following steps:

[0065] In a first sub-step, the intersection between two second-institution user identifier groups corresponding to the target-institution identifier group is determined as an overlapping second-institution user identifier group.

[0066] In a second sub-step, the number of overlapping second-institution user identifiers in the overlapping second-institution user identifier group is determined as a second user overlap number.

[0067] In a third sub-step, the target-institution identifier group, the overlapping second-institution user identifier group, and the second user overlap number are determined as the second user overlap information.

[0068] In a fourth step, third to-be-supplemented institution user data and fourth to-be-supplemented institution user data corresponding to each second user overlap information in the obtained second user overlap information are determined. The following steps can be performed:

[0069] In a first sub-step, for each preset-institution identifier in the preset-institution identifier group, the following steps are performed:

[0070] Sub-step one, selecting a second-institution user identification group matching the preset institution identification from the second-institution user identification group set.

[0071] Sub-step two, determining the number of each second-institution user identification in the selected second-institution user identification group as the second-institution user number.

[0072] Second sub-step, for each second user coincidence information in the second user coincidence information, performing the following steps:

[0073] Sub-step one, determining the third preset institution identification and the fourth preset institution identification corresponding to the second user coincidence information. First, one preset institution identification in the target institution identification group corresponding to the second user coincidence information can be determined as the third preset institution identification. Then, another preset institution identification in the target institution identification group corresponding to the second user coincidence information, which is different from the third preset institution identification, is determined as the fourth preset institution identification.

[0074] Sub-step two, determining the ratio between the second user coincidence number included in the second user coincidence information and the second-institution user number corresponding to the third preset institution identification as the third sampling ratio.

[0075] Sub-step three, determining the ratio between the second user coincidence number included in the second user coincidence information and the second-institution user number corresponding to the fourth preset institution identification as the fourth sampling ratio.

[0076] Sub-step four, based on the third sampling ratio, performing data sampling processing on the second-institution user data group corresponding to the third preset institution identification to obtain a third sampling institution user data group, and determining the third sampling institution user data group and the fourth preset institution identification as the third to-be-supplemented institution user data. The third sampling institution user data in the third sampling institution user data group can be randomly sampled second-institution user data. First, the product of the third sampling ratio and the second-institution user number corresponding to the third preset institution identification is determined as a third numerical value. Then, third numerical value second-institution user data is randomly sampled from the second-institution user data group corresponding to the third preset institution identification, and each second-institution user data obtained is determined as a third sampling institution user data to obtain a third sampling institution user data group.

[0077] Sub-step five, based on the fourth sampling ratio, the second institutional user data set corresponding to the fourth preset institutional identifier is processed to obtain a fourth sampled institutional user data set, and the fourth sampled institutional user data set and the third preset institutional identifier are determined as fourth to-be-supplemented institutional user data. The third sampled institutional user data in the third sampled institutional user data set can be randomly sampled second institutional user data. First, the product of the fourth sampling ratio and the number of second institutional users corresponding to the fourth preset institutional identifier is determined as a fourth value. Then, fourth value second institutional user data is randomly sampled from the second institutional user data set corresponding to the fourth preset institutional identifier, and each second institutional user data obtained is determined as fourth sampled institutional user data to obtain the fourth sampled institutional user data set.

[0078] Step five, based on the obtained third to-be-supplemented institutional user data and fourth to-be-supplemented institutional user data, the second institutional user data set is updated to obtain an updated second institutional user data set. The generation of the updated second institutional user data set can refer to the generation of the updated first institutional user data set in step 104, which will not be repeated here.

[0079] It should be noted that, depending on the second institutional user data set, for each to-be-classified institution, the execution subject can also generate an updated second institutional user data set by adding new data to the second institutional user data set corresponding to the institution, so that the number of data samples corresponding to the institution increases, so that the institution can be more comprehensively and accurately classified and evaluated in the future.

[0080] Step 106, based on the updated first institutional user data set and the updated second institutional user data set, an institutional classification information set is generated.

[0081] In some embodiments, the execution subject can generate an institutional classification information set based on the updated first institutional user data set and the updated second institutional user data set in various ways. The institutional classification information can be information of a value level obtained by classifying the institution. The value level can represent the risk level of the credit value of the institution. For example, the value level can be one of the following: first level, second level, third level, and fourth level. The first level can represent a lower risk. The second level can represent a moderate risk. The third level can represent a higher risk. The fourth level can represent a very high risk. In addition, the value level decreases from the first level to the fourth level in order from high to low.

[0082] In some optional implementations of some embodiments, the execution subject can generate the institution classification information set based on the updated first institution user data set and the updated second institution user data set by the following steps:

[0083] Firstly, determine each updated first institution user data in the updated first institution user data set that is not repeated as the first to-be-evaluated user data set.

[0084] Secondly, determine each updated second institution user data in the updated second institution user data set that is not repeated as the second to-be-evaluated user data set.

[0085] Thirdly, input each first to-be-evaluated user data in the first to-be-evaluated user data set into the pre-trained first user value scoring model in sequence to obtain the first user score information set. Each first user score information in the first user score information set can correspond to the first to-be-evaluated user data in the first to-be-evaluated user data set one by one. The first user score information in the first user score information set can be the information of the score representing the value level of the user corresponding to the first to-be-evaluated user data. The first user value scoring model can be an integrated model based on logistic regression with the first to-be-evaluated user data as the input and the first user score information as the output.

[0086] Fourthly, input each second to-be-evaluated user data in the second to-be-evaluated user data set into the pre-trained second user value scoring model in sequence to obtain the second user score information set. Each second user score information in the second user score information set can correspond to the second to-be-evaluated user data in the second to-be-evaluated user data set one by one. The second user score information in the second user score information set can be the information of the score representing the value level of the user corresponding to the second to-be-evaluated user data. The second user value scoring model can be an integrated model based on logistic regression with the second to-be-evaluated user data as the input and the second user score information as the output. The second user value scoring model can have the same training method as the first user value scoring model.

[0087] Fifthly, generate the institution classification information set based on the first user score information set and the second user score information set. The execution subject can generate the institution classification information set based on the first user score information set and the second user score information set in various ways.

[0088] It needs to be explained that in the process of solving the technical problem one proposed in the background section by adopting the technical solutions, the following technical problem two is often accompanied: how to determine the demand difference of each institution for storage space capacity to reduce resource waste or shortage in storage resource allocation. In view of the above technical problem two, the general solution is generally: according to the user scores corresponding to each institution, determine the demand difference level between institutions through a single index such as maximum value or median value, and allocate resources according to the different demand differences of institutions. However, the above conventional solution still has the following problems: since the single index can only determine the demand difference of different institutions for storage space capacity from one dimension, it is difficult to comprehensively identify the demand difference between different institutions, so that when allocating storage resources to each institution, it is easy to cause waste or shortage of allocated storage resources of institutions. Therefore, in the face of the above technical problem two, combined with the technical advantages of the scheme research and development team in the field of computer technology, the present disclosure decides to adopt the following scheme.

[0089] In some optional implementations of some embodiments, the above execution subject can generate the institution classification information set based on the above first user score information set and the above second user score information set by the following steps:

[0090] Step one, for each preset institution identifier in the above preset institution identifier group, the following steps are executed:

[0091] Substep one, select the first user score information matched with the preset institution identifier from the first user score information set as the target first user score information, and obtain the target first user score information group. Wherein, the first user score information matched with the preset institution identifier can be: the update first institution user data corresponding to the first user score information includes the same calling institution identifier as the preset institution identifier.

[0092] Substep two, select the second user score information matched with the preset institution identifier from the second user score information set as the target second user score information, and obtain the target second user score information group. Wherein, the second user score information matched with the preset institution identifier can be: the update second institution user data corresponding to the second user score information includes the same association institution identifier as the preset institution identifier.

[0093] Step three, based on the target first user score information set and the target second user score information set, generate first user score distribution information and second user score distribution information. The first user score distribution information can be information of various indicators representing the distribution of each user score corresponding to the target first user score information set. The various indicators can include but are not limited to maximum value, minimum value, median, first quartile, and third quartile. The second user score distribution information can be information of various indicators representing the distribution of each user score corresponding to the target second user score information set. A preset indicator generation interface can be called to generate the first user score distribution information and the second user score distribution information based on the target first user score information set and the target second user score information set. The preset indicator generation interface can encapsulate an indicator generation function. The indicator generation function can generate values of various indicators corresponding to an input data set according to the input data set.

[0094] Step four, input the first user score distribution information into a preset institution hierarchical decision tree to obtain first institution level information. The institution hierarchical decision tree can be a decision tree taking user score distribution information as input and taking institution level information as output. The user score distribution information can be information of various indicators representing the distribution of each user score. The institution level information can be information of the value level of an institution. The institution level information can be a value level identifier. The value level identifier can be a unique identifier of the value level. The first institution level information can be institution level information corresponding to the first user score distribution information.

[0095] It should be noted that an institution with a high value level can usually provide better products and services to users, attract more users, and thus generate more user data that needs to be stored in the storage space allocated by the database server.

[0096] Step five, input the second user score distribution information into the institution hierarchical decision tree to obtain second institution level information. The second institution level information can be institution level information corresponding to the second user score distribution information.

[0097] Step six, determine the first institution level information, the second institution level information, and the preset institution identifier as institution hierarchical information.

[0098] The above mechanism classification information generation step and its related content, as one of the invention points of the embodiments of the present disclosure, solves the above technical problem two "how to determine the demand difference of each institution for storage space capacity to reduce the waste or shortage of storage resources allocation". However, the conventional solution has the problem of easy waste or shortage of institutional storage resources, and the reason for this problem is that a single indicator can only determine the demand difference of different institutions for storage space capacity from one dimension, and it is difficult to comprehensively identify the demand difference between different institutions, so that when allocating storage resources to each institution, it is easy to cause waste or shortage of allocated storage resources of the institution. If the above problem is solved, the effect of reducing the waste or shortage of allocated storage resources of the institution can be achieved. In order to achieve this effect, first, for each institution, two user score distribution information can be determined according to the corresponding target first user score information set and the target second user score information set, and input into the mechanism classification decision tree to obtain two mechanism classification information. In this way, the demand difference of each institution for storage space capacity can be more accurately determined from different data set angles and comprehensive index angles. Then, according to the demand difference level between different institutions and the current storage space usage information of each institution, it is determined whether to expand or shrink the capacity of each institution. In this way, the resource adjustment demand of each institution can be obtained more accurately. Finally, according to the more accurate resource demand, the storage space capacity of each institution is adjusted. In this way, the waste or shortage of allocated storage resources of the institution can be reduced. Thus, the balance of storage resource allocation is improved.

[0099] Continuing, in the process of using the technical solution to solve the above technical problem one and technical problem two, the following technical problem three often accompanies: how to determine the demand of institutions for storage space capacity and reduce storage resource waste. In view of the above technical problem three, the conventional solution is generally: generating user score information representing storage resource occupation demand according to user data through complex machine learning method, and then allocating storage resources according to user score information. However, the above conventional solution still has the following problems: because the complex machine learning method such as deep neural network model structure is complex and has long running time, when it is necessary to adjust the storage resources in real time, it is difficult to generate the user score information of the institution in time for timely allocation of storage resources, so that it is difficult to reduce the waste of storage resources in time. Therefore, in the face of the above technical problem three, combined with the technical advantages of the scheme R&D team in the field of computer technology, the present disclosure decides to use the following scheme.

[0100] Optionally, before inputting each first to be evaluated user data in the above first to be evaluated user data set into the pre-trained first user value scoring model to obtain the first user score information set, the above execution subject can further perform the following steps:

[0101] In the first step, a sample institution user data set is obtained. Each sample user data in the sample user data set can be first institution user data used as a sample. Each sample user data in the sample user data set corresponds to a sample label. The sample label can be 0 or 1. 0 can represent on-time repayment. 1 represents failure to repay on time. The sample institution user data set can be obtained from a database.

[0102] In the second step, each sample user data in the sample user data set is preprocessed to obtain a preprocessed sample user data set. The preprocessed sample user data set can be a sample user data set after encoding of text type variables. For each sample user data in the sample user data set, a preset data encoding interface can be called to preprocess the sample user data to obtain preprocessed sample user data. The preset data encoding interface can encapsulate a data encoding function. The data encoding function can perform type detection and encoding on input data including each variable. The type detection can be detection of whether a variable is a text type variable. The encoding can be one-hot encoding of a text type variable with a data category between 2 and 8, and label encoding of a text type variable with a data category greater than 10.

[0103] In the third step, the preprocessed sample user data set is subjected to indiscriminate feature filtering to obtain a filtered sample user data set. The filtered sample user data set can be a preprocessed sample user data set from which a feature variable with a variance less than a preset variance value is filtered out. The preset variance value can be a lower limit value of the variance. For example, the preset variance value can be 0.010. The preprocessed sample user data set can be subjected to indiscriminate feature filtering by a variance filtering method to obtain the filtered sample user data set.

[0104] In the fourth step, the filtered sample user data set is subjected to feature selection to obtain a target sample user data set. The target sample user data set can be composed of selected feature variables from the filtered sample user data set. The filtered sample user data set can be subjected to feature selection by an XGBoost (Extreme Gradient Boosting) model to obtain the target sample user data set.

[0105] In the fifth step, the target sample user data set is classified to obtain a first sample user data set and a second sample user data set. The first sample user data in the first sample user data set can be target sample user data with a corresponding sample label of 0. The second sample user data in the second sample user data set can be target sample user data with a corresponding sample label of 1. The target sample user data set can be classified according to the sample label corresponding to the target sample user data to obtain the first sample user data set and the second sample user data set.

[0106] In the sixth step, a sample quantity ratio is generated based on the first sample user data set and the second sample user data set. The sample quantity ratio can be the ratio of the number of first sample user data to the number of second sample user data. First, the number of each first sample user data in the first sample user data set is determined as a first sample quantity. Then, the number of each second sample user data in the second sample user data set is determined as a second sample quantity. Finally, the ratio of the first sample quantity to the second sample quantity is determined as the sample quantity ratio.

[0107] In the seventh step, in response to determining that the sample quantity ratio is less than a preset sample quantity ratio threshold, the target sample user data set is updated to obtain an updated target sample user data set. The preset sample quantity ratio threshold can be a lower limit value of the sample quantity ratio. The updated target sample user data set can be the target sample user data set after adding newly generated target sample user data. First, a synthetic sample user data set can be generated based on the second sample user data set by using a SMOTE (Synthetic Minority Over-sampling Technique) sampling method. Then, each synthetic sample user data in the synthetic sample user data set is added to the target sample user data set as target sample user data to obtain the updated target sample user data set.

[0108] In the eighth step, the updated target sample user data set is used as a training sample set, and a binary tree set is constructed based on the training sample set. Each leaf node in the binary tree set corresponds to at least one training sample. Each binary tree in the binary tree set can be constructed by feature splitting based on the structural split gain for the training sample set. A binary tree can be generated by feature splitting based on the structural split gain for the training sample set. When the binary tree reaches the maximum depth or has no structural split gain, the next tree is added, and the fitting residual of the previous tree is used for training.

[0109] In the ninth step, for each leaf node corresponding to the binary tree set, a logistic regression model is trained based on at least one training sample corresponding to the leaf node.

[0110] In the tenth step, based on the obtained logistic regression models, a prediction is made for each training sample in the training sample set to obtain a training sample prediction value set. Each training sample prediction value in the training sample prediction value set corresponds to a training sample in the training sample set. The training sample prediction value in the training sample prediction value set can be an average of each prediction result of the corresponding training sample. The prediction result in each prediction result can be a result of a prediction made by a logistic regression model corresponding to a leaf node to which the training sample belongs. The prediction result in each prediction result corresponds to a binary tree in the binary tree set. For each training sample in the training sample set, each binary tree can be traversed to find a leaf node to which the training sample belongs, and a prediction result can be obtained by using the logistic regression model corresponding to the leaf node. The prediction result of each binary tree can be averaged to determine the training sample prediction value.

[0111] In the eleventh step, based on the training sample prediction value set, a first user value scoring model is obtained by using a preset logistic regression iterative algorithm. The logistic regression iterative algorithm can be an Adaboost (Adaptive Boosting) iterative algorithm based on a logistic regression model.

[0112] As an example, first, the training sample prediction value set can be used as a target training sample set, and the weights of all target training samples can be initialized. Then, in each iteration, the current weights are used to train the logistic regression model, the model error and the model weight are calculated, the weights of all target training samples are updated, the next iteration is started, and finally, after multiple iterations, the logistic regression models in each iteration are combined by using the model weights corresponding to the logistic regression models in each iteration to obtain the first user value scoring model. By updating the weights of all target training samples in each iteration, the weights of the target training samples with classification errors can be increased, and the weights of the target training samples with correct classification can be decreased.

[0113] The first user value scoring model training step and related content thereof are an inventive point of an embodiment of the present disclosure, and solve the third technical problem of how to determine the storage space capacity occupation demand of an organization and reduce storage resource waste in a timely manner. The conventional solution has the problem of being difficult to reduce storage resource waste in a timely manner, and the reason for this problem is that the complex machine learning method, such as the deep neural network model, has a complex structure and a long running time, so that when it is necessary to adjust the storage resources in real time, it is difficult to generate user score information of the organization in a timely manner for timely allocation of storage resources, thereby causing difficulty in reducing storage resource waste in a timely manner. If the above problem is solved by the present solution, the effect of reducing storage resource waste in a timely manner can be achieved. In order to achieve this effect, first, the sample user data set is preprocessed and feature screened. In this way, the target sample user data set including various feature variables that have a greater impact on the prediction result after coding can be obtained. Second, it is detected whether the target sample user data set is unbalanced, and the unbalanced data is supplemented to obtain an updated target sample user data set. In this way, the updated target sample user data set can be distributed evenly, thereby facilitating the improvement of the accuracy and reliability of the model. Then, the gain of the structure is used for tree splitting, and a plurality of binary tree trees are formed by fitting the residual. Next, each leaf node of each tree is trained using logistic regression, and thus each logistic regression model can be obtained. Then, according to each logistic regression model, the prediction value result of each training sample is determined. After that, according to each prediction value result, the logistic regression base model is iteratively trained using the Adaboost algorithm to obtain the first user value scoring model. In this way, the first user value scoring model with a relatively simple structure can be obtained, thereby shortening the running time of the model, generating user score information in a timely manner, and generating organization classification information representing storage capacity occupation demand in a timely manner. Finally, the user score information of each organization can be used to adjust the storage resources of each organization in a timely manner. In this way, the waste of storage resources can be reduced in a timely manner. Moreover, the first user value scoring model integrates a plurality of logistic regression base models, thereby making the model more interpretable. Furthermore, the model uses an ensemble learning method during training, thereby making the model more accurate.

[0114] In practice, when the amount of original sample data corresponding to the current organization is insufficient, the amount of original sample data corresponding to the current organization can be supplemented by using the user overlap information between organizations and combining sample data of other organizations with similar characteristics, thereby making the amount of sample data corresponding to the current organization sufficient without consuming a long time for data accumulation. Moreover, two different source types of sample data sets are also used to generate organization classification information, thereby obtaining classification information of different data dimensions for each organization.

[0115] In step 107, each piece of the pre-acquired institution storage space usage information is detected to obtain a set of institution allocable storage space information.

[0116] In some embodiments, the execution subject can detect each piece of the pre-acquired institution storage space usage information in various ways to obtain a set of institution allocable storage space information. The institution storage space usage information in the set of institution storage space usage information can be information of the allocated storage space of the corresponding institution. The institution allocable storage space information in the set of institution allocable storage space information can be information of the allocation state of the corresponding institution when participating in the storage space capacity allocation. The allocation state can be one of the following: expansion state and contraction state. The expansion state can represent expansion of the storage space. The contraction state can represent contraction of the storage space.

[0117] In some optional implementations of some embodiments, each piece of the institution storage space usage information in the set of institution storage space usage information can include an institution identifier, an allocated storage capacity, and a used storage capacity. The allocated storage capacity can be the capacity of the storage space pre-allocated to the institution. The used storage capacity can be the storage capacity used when storing the user data of the institution. The execution subject can detect each piece of the pre-acquired institution storage space usage information in the following steps to obtain a set of institution allocable storage space information:

[0118] For each piece of the institution storage space usage information in the set of institution storage space usage information, the following steps are performed:

[0119] First, the ratio of the used storage capacity to the allocated storage capacity in the institution storage space usage information is determined as the used proportion.

[0120] Second, in response to determining that the used proportion is less than a first preset used space proportion, historical storage space usage data sequence corresponding to the institution storage space usage information is obtained. The first preset used space proportion can be a pre-set lower limit value of the used proportion. For example, the first preset used space proportion can be 0.5. The historical storage space usage data sequence can correspond to the same institution as the institution storage space usage information. The historical storage space usage data sequence can be a sequence of historical storage space usage data arranged in chronological order. The historical storage space usage data can include, but is not limited to, storage date and used capacity.

[0121] In the third step, the historical storage space usage data sequence is input into a preset storage space usage prediction model to obtain a storage space usage prediction data sequence. The storage space usage prediction model can be a time series prediction model with the historical storage space usage data sequence as input and the storage space usage prediction data sequence as output. For example, the storage space usage prediction model can be, but is not limited to, one of the following: an autoregressive moving average model, a long short-term memory network. The storage space usage prediction data sequence can be a sequence of daily storage space usage in a future time period. The future time period can be one month. The storage space usage prediction data in the storage space usage prediction data sequence can include a storage date and an estimated usage capacity. The estimated usage capacity can be the predicted daily storage space usage.

[0122] In the fourth step, a target usage capacity is generated based on the used storage capacity included in the institutional storage space usage information and the storage space usage prediction data sequence. The target usage capacity can be the total usage of the storage space after one month. The sum of the used storage capacity included in the institutional storage space usage information and each estimated usage capacity included in the storage space usage prediction data sequence can be determined as the target usage capacity.

[0123] In the fifth step, a difference between the allocated storage capacity included in the institutional storage space usage information and the target usage capacity is determined as the allocable space capacity.

[0124] In the sixth step, in response to determining that the allocable space capacity is greater than a preset new capacity threshold, an institutional identifier included in the institutional storage space usage information, the allocable space capacity, and a preset capacity reduction identifier are determined as institutional allocable storage space information. The preset new capacity threshold can be a lower limit value of the new storage capacity of each level of institution. For example, the preset new capacity threshold can be 5G. The preset capacity reduction identifier can represent that the storage space of the corresponding institution can be reduced.

[0125] Optionally, the execution subject can also determine an institutional identifier included in the institutional storage space usage information and a preset capacity expansion identifier as the institutional allocable storage space information in response to determining that the used ratio is greater than a second preset used space ratio. The second preset used space ratio can be a preset upper limit value of the used ratio. For example, the second preset used space ratio can be 0.95. The preset capacity expansion identifier can represent that the storage space of the corresponding institution needs to be expanded.

[0126] In step 108, institutional capacity expansion information and institutional capacity reduction information are generated based on the set of institutional classification information and the set of institutional allocable storage space information.

[0127] In some embodiments, the execution subject described above can generate an institution expansion information set and an institution contraction information set based on the institution hierarchical information set and the institution allocable storage space information set in various ways. The institution expansion information in the institution expansion information set can be information about the storage capacity to be expanded by the corresponding institution. The institution contraction information in the institution contraction information set can be information about the storage capacity to be contracted by the corresponding institution. The following steps can be performed in particular:

[0128] First, for each institution hierarchical information in the institution hierarchical information set, the following steps are performed:

[0129] In the first sub-step, in response to determining that the value level corresponding to the first institution level information included in the institution hierarchical information is higher than the value level corresponding to the second institution level information included in the institution hierarchical information, the second institution level information included in the institution hierarchical information is determined as the target institution level information.

[0130] In the second sub-step, in response to determining that the value level corresponding to the first institution level information included in the institution hierarchical information is lower than or equal to the value level corresponding to the second institution level information included in the institution hierarchical information, the first institution level information included in the institution hierarchical information is determined as the target institution level information.

[0131] Second, each institution allocable storage space information in the institution allocable storage space information set is grouped to obtain a first institution allocable storage space information group and a second institution allocable storage space information group. The first institution allocable storage space information in the first institution allocable storage space information group can be institution allocable storage space information including a preset expansion identifier. The second institution allocable storage space information in the second institution allocable storage space information group can be institution allocable storage space information including a preset contraction identifier.

[0132] Third, according to the institution hierarchical information set, each first institution allocable storage space information in the first institution allocable storage space information group is arranged in descending order by a preset sorting algorithm to obtain a first institution allocable storage space information sequence. The first institution allocable storage space information sequence can be a set of each first institution allocable storage space information arranged in descending order according to the level of the corresponding institution. For example, the sorting algorithm can be, but is not limited to, one of the following: bubble sort, quicksort.

[0133] Fourthly, according to the sorting algorithm, the second-agency allocable storage space information in the second-agency allocable storage space information set is arranged in descending order according to the size of the allocable space capacity, and a second-agency allocable storage space information sequence is obtained as the capacity-reduced allocable storage space information sequence.

[0134] Fifthly, the pre-design counter is initialized to obtain a target counter value. The target counter value can be initially set as 1.

[0135] Sixthly, based on the first-agency allocable storage space information sequence, the target counter value and the capacity-reduced allocable storage space information sequence, the following steps of generating an agency capacity expansion information set and an agency capacity reduction information set are performed:

[0136] Firstly, the first-agency allocable storage space information corresponding to the same sequence number as the target counter value in the first-agency allocable storage space information sequence is determined as a target-agency allocable storage space information.

[0137] Secondly, from the obtained target-agency level information, a target-agency level information corresponding to the same agency identifier as the agency identifier included in the target-agency allocable storage space information is selected as a to-be-expanded agency level information.

[0138] Thirdly, from the capacity-reduced allocable storage space information sequence, each capacity-reduced allocable storage space information corresponding to a target-agency level information satisfying a preset level condition is selected in sequence to obtain at least one capacity-reduced allocable storage space information. The preset level condition can be that the value level corresponding to the capacity-reduced allocable storage space information is lower than the value level corresponding to the to-be-expanded agency level information.

[0139] Fourthly, from the preset hierarchical resource tentative allocation information set, a hierarchical resource tentative allocation information matching the to-be-expanded agency level information is selected. Each hierarchical resource tentative allocation information in the hierarchical resource tentative allocation information set includes a value level identifier and a tentative allocation resource amount. The tentative allocation resource amount can be the capacity of the storage resource allocated each time. The matching of the to-be-expanded agency level information can be that the value level identifier corresponding to the hierarchical resource tentative allocation information is the same as the value level identifier corresponding to the to-be-expanded agency level information.

[0140] In the fifth sub-step, a capacity expansion resource amount and a capacity reduction organization resource information set are generated through a preset resource matching interface according to the selected hierarchical resource tentative allocation information and the at least one capacity reduction allocation storage space information. The capacity expansion resource amount can be a storage space capacity to be used for capacity expansion. Each capacity reduction organization resource information in the capacity reduction organization resource information set can include a capacity reduction organization identifier and a capacity reduction resource amount. The capacity reduction organization identifier can be an identifier of an organization to be reduced in capacity. The capacity reduction resource amount can be a storage space capacity to be used for capacity reduction. The sum of the capacity reduction resource amounts included in the capacity reduction organization resource information set is greater than or equal to the capacity expansion resource amount. The resource matching interface can be encapsulated with a resource matching function. The resource matching function can determine the capacity expansion resource amount and the capacity reduction organization resource information set according to the input hierarchical resource tentative allocation information and the at least one capacity reduction allocation storage space information.

[0141] For example, when the tentative allocation resource amount included in the hierarchical resource tentative allocation information is 20 and the at least one capacity reduction allocation storage space information is {“organization 1”: 15, “organization 2”: 10, “organization 3”: 6, “organization 4”: 4}, the capacity expansion resource amount can be 21, and the capacity reduction organization resource information set can be {“organization 1”: 15, “organization 3”: 6}. In addition, when the tentative allocation resource amount is 40 and the at least one capacity reduction allocation storage space information remains unchanged, the capacity expansion resource amount can be 40, and the capacity reduction organization resource information set can be {“organization 1”: 15, “organization 2”: 10, “organization 3”: 6, “organization 4”: 4, “database server”: 5}, where {“database server”: 5} represents that a new storage space is allocated by the database server.

[0142] In the sixth sub-step, each capacity reduction allocation storage space information in the capacity reduction allocation storage space information sequence that matches the capacity reduction organization resource information set is deleted to obtain a capacity reduction allocation storage space information sequence after deletion. The matching of the capacity reduction organization resource information set can be that the organization identifier corresponding to the capacity reduction allocation storage space information is the same as the capacity reduction organization identifier included in any capacity reduction organization resource information.

[0143] In the seventh sub-step, the organization identifier included in the target organization allocatable storage space information and the capacity expansion resource amount are determined as organization capacity expansion information.

[0144] An eighth sub-step, in response to determining that the target count value is equal to or greater than the preset count value, determining the obtained each institution expansion information as the institution expansion information set, and determining each shrinkage institution resource information in the obtained each shrinkage institution resource information set as the institution shrinkage information set. The preset count value can be the number of each first institution allocable storage space information in the first institution allocable storage space information set. The target expansion institution identifier can be the institution identifier included in the target institution allocable storage space information corresponding to the shrinkage institution resource information.

[0145] Optionally, in response to determining that the target count value is less than the preset count value, updating the target count value to obtain an updated count value, and taking the updated count value as the target count value, and taking the deleted shrinkage allocation storage space information sequence as the shrinkage allocation storage space information sequence, and executing the institution expansion information set and the institution shrinkage information set generation step again. The target count value and 1 can be determined as the updated count value.

[0146] Step 109, based on the institution expansion information set and the institution shrinkage information set, controlling the associated database server to adjust the storage space capacity of each institution.

[0147] In some embodiments, the execution subject can control the associated database server to adjust the storage space capacity of each institution based on the institution expansion information set and the institution shrinkage information set. The database server can expand the storage space capacity of each expansion institution corresponding to the institution expansion information set according to the institution shrinkage information corresponding to the expansion institution. The database server can also shrink the storage space capacity of each shrinkage institution corresponding to the institution shrinkage information set according to the institution shrinkage information corresponding to the shrinkage institution.

[0148] It should be noted that the resource allocation method of the present disclosure can not only be used to balance the database storage space of each institution, but also be adaptively used to adjust the bandwidth resources of each institution access server and the financial limit applicable to each institution, so as to realize the balance of bandwidth resources and financial limit resources, and reduce the waste and excessive occupation of resources.

[0149] The above various embodiments of the present disclosure have the following beneficial effects: through the resource allocation method of some embodiments of the present disclosure, the excessive occupation and waste of storage resources can be reduced. Specifically, the reason for the waste or excessive occupation of storage resources is that due to the differences between the user groups of various institutions, the storage space capacities required by various institutions also have great differences. When the required storage space of an institution is small and there is a lot of unused storage space, it is easy to cause waste of storage resources. When the required storage space of an institution is large and the storage space capacity is insufficient, if the above-mentioned re-division method is used for storage space expansion, it will lead to excessive occupation of storage resources. Based on this, the resource allocation method of some embodiments of the present disclosure first, in response to confirming that the database storage space for storing institution user data is insufficient, sends a database storage space adjustment request to the server. Secondly, in response to receiving the storage space adjustment instruction sent by the server, the first institution user data set and the second institution user data set are obtained. Each first institution user data in the first institution user data set is the log data when the institution calls the preset business product with user information as the query primary key, and each second institution user data in the second institution user data set is the user behavior label data collected by the institution. Thus, two different source types of sample data sets corresponding to each institution can be obtained. Then, based on the first institution user data set, the first user overlap information corresponding to each target institution identification group in the target institution identification group set is determined, wherein each target institution identification group in the target institution identification group set is composed of two different preset institution identification groups in the preset institution identification group. Thus, the information of the user group overlap between different institutions corresponding to the product call log record can be determined. Then, based on the obtained first user overlap information, the first institution user data group set is updated to obtain an updated first institution user data group set. Thus, according to the user overlap sample data between different institutions, the original sample data of each institution can be supplemented based on the original sample data of each institution to make up for the lack of original sample data. Next, based on the second institution user data set, an updated second institution user data group set is generated. Thus, the original sample data of the institution corresponding to the user behavior can be supplemented to make up for the lack of original sample data. Then, based on the updated first institution user data group set and the updated second institution user data group set, an institution classification information set is generated. Thus, the level information of each institution can be obtained. Then, each institution storage space usage information in the pre-obtained institution storage space usage information set is detected to obtain an institution allocatable storage space information set. Thus, the allocatable state of the current storage space capacity of each institution can be obtained.Further, based on the above-mentioned institution classification information set and the above-mentioned institution allocable storage space information set, an institution expansion information set and an institution contraction information set are generated. In this way, the resource information of each institution to be expanded and the resource information of each institution to be contracted can be obtained. Finally, based on the above-mentioned institution expansion information set and the above-mentioned institution contraction information set, the associated database server is controlled to adjust the storage space capacity of each institution. Therefore, the resource allocation method of some embodiments of the present disclosure can, when the institution user data storage space is insufficient, determine the level of the institution according to the customer information of the institution, and on this basis, determine the institution expansion information set and the institution contraction information set according to the institution classification information and the current resource allocable state of the institution, so as to contract the institution with less storage space capacity and more unused storage space, and use the contracted capacity to expand the storage space of the institution which needs to use more storage space and has insufficient storage space capacity. Thus, the excessive occupation and waste of storage resources can be reduced.

[0150] Further reference Figure 2 , as an implementation of the method shown in the above-mentioned figures, the present disclosure provides some embodiments of a resource allocation device, which device embodiments correspond to those method embodiments shown in Figure 1 , and the resource allocation device 200 can be specifically applied to various electronic devices.

[0151] As Figure 2As shown, the resource allocation apparatus 200 of some embodiments includes a sending unit 201, an obtaining unit 202, a determining unit 203, an updating processing unit 204, a first generating unit 205, a second generating unit 206, a detecting processing unit 207, a third generating unit 208, and a control unit 209. The sending unit 201 is configured to send a database storage space adjustment request to a server in response to confirming that the database storage space for storing the agency user data is insufficient. The obtaining unit 202 is configured to obtain a first agency user data set and a second agency user data set in response to receiving the storage space adjustment instruction sent by the server. Each first agency user data in the first agency user data set is log data when the agency calls a preset business product with user information as a query primary key, and each second agency user data in the second agency user data set is user behavior label data collected by the agency. The determining unit 203 is configured to determine first user overlap information corresponding to each target agency identification group in a target agency identification group set based on the first agency user data set. Each target agency identification group in the target agency identification group set is composed of two different preset agency identification groups in a preset agency identification group set. The updating processing unit 204 is configured to perform updating processing on the first agency user data group set based on the obtained first user overlap information, to obtain an updated first agency user data group set. The first generating unit 205 is configured to generate an updated second agency user data group set based on the second agency user data set. The second generating unit 206 is configured to generate an agency hierarchical information set based on the updated first agency user data group set and the updated second agency user data group set. The detecting processing unit 207 is configured to perform detecting processing on each agency storage space usage information in a pre-obtained agency storage space usage information set, to obtain an agency allocable storage space information set. The third generating unit 208 is configured to generate an agency expansion information set and an agency contraction information set based on the agency hierarchical information set and the agency allocable storage space information set. The control unit 209 is configured to control the associated database server to perform storage space capacity adjustment on each agency based on the agency expansion information set and the agency contraction information set.

[0152] It can be understood that the units described in the resource allocation apparatus 200 correspond to the respective steps in the method described with reference to Figure 1 The operations, features, and advantages described above for the method also apply to the resource allocation apparatus 200 and the units included therein, and are not repeated here.

[0153] Further reference is made to Figure 3 which shows a structural schematic diagram of an electronic device 300 suitable for implementing some embodiments of the present disclosure. Figure 3The electronic device shown is merely one example, and should not limit the scope of functionality or use of embodiments of the disclosure.

[0154] As shown in Figure 3 The electronic device 300 can include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 301 that can perform various appropriate actions and processes according to programs stored in a read-only memory (ROM) 302 or loaded into a random access memory (RAM) 303 from a storage device 308. Various programs and data required for the operation of the electronic device 300 are also stored in the RAM 303. The processing device 301, the ROM 302, and the RAM 303 are connected to each other through a bus 304. An input / output (I / O) interface 305 is also connected to the bus 304.

[0155] In general, the following devices can be connected to the I / O interface 305: input devices 306 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; output devices 307 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; storage devices 308 including, for example, a magnetic tape, a hard disk, etc.; and communication devices 309. The communication devices 309 can allow the electronic device 300 to communicate wirelessly or wired with other devices to exchange data. Although Figure 3 The electronic device 300 is shown with various devices, but it should be understood that not all of the devices shown are required to be implemented or present. More or fewer devices can alternatively be implemented or present. Figure 3 Each block shown in the flowcharts can represent a device or multiple devices as needed.

[0156] In particular, processes described above with reference to the flowcharts can be implemented as a computer software program according to some embodiments of the disclosure. For example, some embodiments of the disclosure include a computer program product comprising a computer program carried on a computer readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In some such embodiments, the computer program can be downloaded and installed from a network through the communication devices 309, or installed from the storage devices 308, or installed from the ROM 302. When the computer program is executed by the processing device 301, the above-mentioned functions defined in the methods of some embodiments of the disclosure are performed.

[0157] Note that the computer-readable medium or media used to provide the computer program sequence to the computer system can be accompanied by information sufficient to load the program sequence into an internal register, or onto a storage device, or into the system memory, or to be executed by the computer system. As such, the computer-readable medium or media is persistant storage that can be read by a machine, such as a computer. The computer-readable medium or media may, for example, include a floppy disk, a ZIP® disk, an optical disk, an electrical connection employing electrical conductors, a telephone line, a coaxial cable, a fiber-optic cable, a CAT 5 cable, or a combination of these or other computer-readable media.

[0158] In some embodiments, the user terminals, servers can communicate using any known or future developed network protocols, such as HTTP (HyperText Transfer Protocol), and can be interconnected with digital data communication (e.g., communication networks) of any form or medium, including the Internet. Examples of communication networks include local area networks ("LANs"), wide area networks ("WANs"), internetworks (e.g., the Internet), and peer-to-peer networks (e.g., ad hoc peer-to-peer networks), as well as any currently known or future developed networks.

[0159] The computer readable medium can be included in the device; or exist separately and not be assembled into the electronic device. The computer readable medium carries one or more programs, which, when executed by the electronic device, cause the electronic device to: in response to confirming that the database storage space for storing the institution user data is insufficient, send a database storage space adjustment request to a server; in response to receiving a storage space adjustment instruction sent by the server, obtain a first institution user data set and a second institution user data set, wherein each first institution user data in the first institution user data set is log data when the institution calls a preset business product with user information as a query primary key, and each second institution user data in the second institution user data set is user behavior label data collected by the institution; based on the first institution user data set, determine first user overlap information corresponding to each target institution identifier group in a target institution identifier group set, wherein each target institution identifier group in the target institution identifier group set is composed of two different preset institution identifier groups; based on the obtained first user overlap information, update the first institution user data group set to obtain an updated first institution user data group set; based on the second institution user data set, generate an updated second institution user data group set; based on the updated first institution user data group set and the updated second institution user data group set, generate an institution hierarchical information set; detect each institution storage space usage information in a pre-obtained institution storage space usage information set to obtain an institution allocable storage space information set; based on the institution hierarchical information set and the institution allocable storage space information set, generate an institution expansion information set and an institution contraction information set; and based on the institution expansion information set and the institution contraction information set, control the associated database server to adjust the storage space capacity of each institution.

[0160] Computer program code for carrying out operations of some embodiments of the disclosure can be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C++ or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code can execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection can be made to an external computer (for example, through the Internet using an Internet Service Provider).

[0161] The flow and block diagrams in the drawings represent possible architectural, functional, and operational scenarios of systems, methods, and computer program products according to various embodiments of the present disclosure. In this regard, each block can represent a module, a segment, or a portion of code that comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that in some alternative implementations, the functions noted in the blocks can occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently or in the reverse order, depending on the functionality involved. It will also be noted that each block of the block diagrams and / or flowchart illustrations, and combinations thereof, can be implemented by a dedicated hardware-based system that performs the specified functions or operations, or combinations of hardware and software.

[0162] The units described in some embodiments of the present disclosure can be implemented by means of software, or can be implemented by hardware. The described units can also be arranged in a processor, for example, can be described as: a processor comprising a sending unit, an obtaining unit, a determining unit, an updating processing unit, a first generating unit, a second generating unit, a detection processing unit, a third generating unit, and a control unit. Among them, the names of these units do not constitute a limitation to the units themselves in some cases, for example, the sending unit can also be described as "a unit for sending a database storage space adjustment request to a server".

[0163] The functions described above in the present document can be performed, at least in part, by one or more hardware logic components. For example, and without limitation, example types of hardware logic components that can be used include Field-programmable Gate Arrays (FPGAs), Application-specific Integrated Circuits (ASICs), Application-specific Standard Products (ASSPs), System-on-a-chip systems (SOCs), Complex Programmable Logic Devices (CPLDs), etc.

[0164] The above description is merely some of the preferred embodiments of the present disclosure and a description of the principles of the technology used. Those skilled in the art should understand that the scope of the application involved in the embodiments of the present disclosure is not limited to the technical solutions formed by the specific combinations of the above technical features, and should also cover other technical solutions formed by any combinations of the above technical features or equivalent features without departing from the above inventive concept. For example, the above features are replaced with technical features with similar functions disclosed in the embodiments of the present disclosure (but not limited to) to form technical solutions.

Claims

1. A resource allocation method, comprising: In response to the confirmation that the database storage space used to store institutional user data is insufficient, a database storage space adjustment request is sent to the server; In response to receiving a storage space adjustment instruction sent by the server, a first institutional user dataset and a second institutional user dataset are obtained, wherein each first institutional user data in the first institutional user dataset is log data when the institution calls a preset business product using user information as the query primary key, and each second institutional user data in the second institutional user dataset is user behavior tag data collected by the institution. The first institution user data in the first institution user dataset is classified to obtain the first institution user data set. Based on the first institution user data set, the first user overlap information corresponding to each target institution identifier group in the target institution identifier set is determined. Each target institution identifier group in the target institution identifier set is composed of two different preset institution identifiers in the preset institution identifier group. Based on the obtained overlapping information of each first user, the first institution user data set is supplemented with data to obtain an updated first institution user data set. The second institution user data in the second institution user dataset is classified to obtain a second institution user data set. Based on the second institution user data set, the second user overlap information corresponding to each target institution identifier group in the target institution identifier group is determined. Based on the second user overlap information, the second institution user data set is supplemented to obtain an updated second institution user data set. Based on the updated first institutional user data set and the updated second institutional user data set, an institutional hierarchical information set is generated; The storage space usage information of each institution in the pre-acquired institution storage space usage information set is detected and processed to obtain the institution's allocable storage space information set. Based on the institution hierarchical information set and the institution's adjustable storage space information set, an institution expansion information set and an institution reduction information set are generated. Based on the expansion information set and the contraction information set of the institutions, the associated database server is controlled to adjust the storage space capacity of each institution.

2. The method according to claim 1, wherein, The step of determining the first user overlap information corresponding to each target institution identifier group in the target institution identifier group set based on the first institution user data set includes: The first institution user data in the first institution user dataset is classified to obtain a first institution user data set, wherein each first institution user data set corresponds to a preset institution identifier in a preset institution identifier group; Based on the first institutional user data set, a total number of users and a first institutional user identifier set are generated, wherein each first institutional user identifier set in the first institutional user identifier set corresponds to a preset institutional identifier in the preset institutional identifier set; For each preset organization identifier in the preset organization identifier group, perform the following steps: Select a first organization user identifier group that matches the preset organization identifier from the first organization user identifier group set; The number of each first organization user identifier in the selected first organization user identifier group is determined as the number of first organization users; The ratio of the first institutional user count to the total number of users is determined as the institutional user percentage. In response to determining that the proportion of users of each institution corresponding to the preset institution identifier group meets the preset user proportion condition, the preset institution identifiers in the preset institution identifier group are grouped in pairs to obtain the target institution identifier group set; For each target organization identifier group in the target organization identifier group set, the two first organization user identifier groups corresponding to the target organization identifier group are matched to obtain the first user overlap information.

3. The method according to claim 2, wherein, Each of the obtained first user overlap information includes the number of first user overlaps; and the step of supplementing the first institution user data set with data based on the obtained first user overlap information to obtain an updated first institution user data set includes: For each piece of first user overlap information obtained from the various first user overlap information, perform the following steps: Determine the first preset organization identifier and the second preset organization identifier corresponding to the first user overlap information; The ratio between the number of overlapping users in the first user overlap information and the number of institutional users corresponding to the first preset institutional identifier is determined as the first sampling ratio. The ratio between the number of overlapping first users included in the first user overlap information and the number of institutional users corresponding to the second preset institutional identifier is determined as the second sampling ratio. Based on the first sampling ratio, data sampling processing is performed on the first institution user data group corresponding to the first preset institution identifier to obtain the first sampled institution user data group, and the first sampled institution user data group and the second preset institution identifier are determined as the first institution user data to be supplemented. Based on the second sampling ratio, data sampling processing is performed on the second institution user data group corresponding to the second preset institution identifier to obtain the second sampled institution user data group, and the second sampled institution user data group and the first preset institution identifier are determined as the second institution user data to be supplemented. Based on the obtained first and second institutional user data to be supplemented, the first institutional user data set is updated to obtain an updated first institutional user data set.

4. The method according to claim 3, wherein, The step of generating and updating the second institution user data set based on the second institution user data set includes: The individual user data of the second institution in the second institution user dataset is classified to obtain the second institution user data set; Based on the second institution user data set, a second institution user identifier set is generated, wherein each second institution user identifier set in the second institution user identifier set corresponds to a preset institution identifier in the preset institution identifier set; Determine the second user overlap information corresponding to each target organization identifier group in the target organization identifier group set; Determine the third and fourth supplementary organization user data corresponding to each second user overlap information obtained; Based on the obtained user data of each third institution to be supplemented and user data of each fourth institution to be supplemented, the user data set of the second institution is updated to obtain the updated user data set of the second institution.

5. The method according to claim 1, wherein, The step of generating an institutional hierarchy information set based on the updated first institutional user data set and the updated second institutional user data set includes: Each unique updated first institution user data set in the updated first institution user data set is identified as the first user dataset to be evaluated; Each unique updated second-institution user data set in the updated second-institution user data set is identified as the second user dataset to be evaluated; Each first user data in the first user dataset to be evaluated is input into a pre-trained first user value scoring model to obtain a first user score information set, wherein each first user score information in the first user score information set corresponds to the first user data in the first user dataset to be evaluated. Each second user data in the second user dataset to be evaluated is input into a pre-trained second user value scoring model to obtain a second user score information set, wherein each second user score information in the second user score information set corresponds to a second user data in the second user dataset to be evaluated. Based on the first user score information set and the second user score information set, an institutional classification information set is generated.

6. The method according to claim 1, wherein, Each organization storage space usage information in the organization storage space usage information set includes the organization identifier, allocated storage capacity, and used storage capacity; The process of detecting and processing the storage space usage information of each organization in the pre-acquired organization storage space usage information set to obtain the organization's allocateable storage space information set includes: For each piece of institutional storage space usage information in the institutional storage space usage information set, perform the following steps: The ratio of used storage capacity to allocated storage capacity, which is included in the organization's storage space usage information, is determined as the percentage of used storage capacity. In response to determining that the percentage of used space is less than a first preset percentage of used space, the historical storage space usage data sequence corresponding to the institutional storage space usage information is obtained; The historical storage space usage data sequence is input into a preset storage space usage prediction model to obtain a storage space usage prediction data sequence. Based on the used storage capacity and the predicted storage space usage data sequence included in the institutional storage space usage information, a target usage capacity is generated; The difference between the allocated storage capacity and the target usage capacity included in the institutional storage space usage information is determined as the adjustable space capacity; In response to determining that the adjustable space capacity is greater than a preset new capacity threshold, the organization identifier, the adjustable space capacity, and the preset reduction identifier included in the organization storage space usage information are determined as organization adjustable storage space information.

7. The method according to claim 6, wherein, The method further includes: In response to determining that the percentage of used space is greater than the second preset percentage of used space, the organization identifier and the preset expansion identifier included in the organization storage space usage information are determined as the organization's adjustable storage space information.

8. A resource allocation device, comprising: The sending unit is configured to send a database storage space adjustment request to the server in response to a confirmation that the database storage space used to store institutional user data is insufficient. The acquisition unit is configured to acquire a first institutional user dataset and a second institutional user dataset in response to receiving a storage space adjustment instruction sent by the server. Each first institutional user dataset in the first institutional user dataset is log data when the institution calls a preset business product using user information as the query primary key. Each second institutional user dataset in the second institutional user dataset is user behavior tag data collected by the institution. The determining unit is configured to determine the first user overlap information corresponding to each target institution identifier group in the target institution identifier group set based on the first institution user data set, wherein each target institution identifier group in the target institution identifier group set is composed of two different preset institution identifiers in the preset institution identifier group; the first institution user data set is obtained by classifying each first institution user data in the first institution user dataset. The update processing unit is configured to supplement the first institution user data set with data based on the obtained overlap information of each first user, so as to obtain an updated first institution user data set. The first generation unit is configured to determine the second user overlap information corresponding to each target institution identifier group in the target institution identifier group based on the second institution user data set, and supplement the second institution user data set with data based on the second user overlap information to obtain an updated second institution user data set; the second institution user data set is obtained by classifying and processing each second institution user data in the second institution user dataset. The second generation unit is configured to generate an institutional hierarchy information set based on the updated first institutional user data set and the updated second institutional user data set; The detection and processing unit is configured to detect and process the storage space usage information of each organization in the pre-acquired organization storage space usage information set to obtain the organization's allocatable storage space information set. The third generation unit is configured to generate an organization expansion information set and an organization reduction information set based on the organization hierarchical information set and the organization adjustable storage space information set. The control unit is configured to control the associated database server to adjust the storage space capacity of each organization based on the organization expansion information set and the organization reduction information set.

9. An electronic device, comprising: One or more processors; Storage device, on which one or more programs are stored, When the one or more programs are executed by the one or more processors, the one or more processors implement the method as described in any one of claims 1-7.

10. A computer-readable medium having a computer program stored thereon, wherein, When the computer program is executed by a processor, it implements the method as described in any one of claims 1-7.

Images