CN110659133B - Resource allocation method and allocation device, storage medium and electronic equipment - Google Patents
Resource allocation method and allocation device, storage medium and electronic equipment Download PDFInfo
- Publication number
- CN110659133B CN110659133B CN201910822837.7A CN201910822837A CN110659133B CN 110659133 B CN110659133 B CN 110659133B CN 201910822837 A CN201910822837 A CN 201910822837A CN 110659133 B CN110659133 B CN 110659133B
- Authority
- CN
- China
- Prior art keywords
- target
- resource
- partition
- information
- target object
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000013468 resource allocation Methods 0.000 title claims abstract description 65
- 238000000034 method Methods 0.000 title claims abstract description 51
- 238000003860 storage Methods 0.000 title claims abstract description 10
- 238000005192 partition Methods 0.000 claims abstract description 183
- 238000012545 processing Methods 0.000 claims abstract description 68
- 238000011156 evaluation Methods 0.000 claims abstract description 59
- 230000006870 function Effects 0.000 claims description 75
- 238000004590 computer program Methods 0.000 claims description 5
- 238000004891 communication Methods 0.000 description 10
- 230000008676 import Effects 0.000 description 8
- 238000010586 diagram Methods 0.000 description 6
- 238000009826 distribution Methods 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 5
- 238000005457 optimization Methods 0.000 description 4
- 230000005236 sound signal Effects 0.000 description 4
- 230000002159 abnormal effect Effects 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 238000003672 processing method Methods 0.000 description 3
- 230000001960 triggered effect Effects 0.000 description 3
- 230000008859 change Effects 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 230000005856 abnormality Effects 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 230000000712 assembly Effects 0.000 description 1
- 238000000429 assembly Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000001012 protector Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
- G06F9/5027—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals
Landscapes
- Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a resource allocation method and device, a storage medium and electronic equipment. Comprising the following steps: receiving an input resource processing request carrying a selected keyword; searching target historical information matched with the selected keywords from the historical data; importing the target history information into a gradient promotion tree gbdt model to predict whether a target object in the target partition continues to use the target resource; counting the target quantity of the prediction results of the target objects in the target partition for continuing to use the target resources; evaluating the target quantity by using an approximation function of the absolute value loss function to obtain an evaluation parameter; optimizing gbdt the model according to the evaluation parameters; predicting the resource use condition of a target object in the target partition on a target resource by using the optimized gbdt model; and distributing the target resource to the target object of the target partition according to the resource use condition and the current information of the target object in the target partition. The embodiment of the invention can improve the accuracy of resource allocation.
Description
Technical Field
The present invention relates to the field of data processing technologies, and in particular, to a resource allocation method and apparatus, a storage medium, and an electronic device.
Background
In daily life, a certain resource is required to be allocated in many times, and the resource is allocated to a target object of a target partition according to a preset rule so as to meet the production and life requirements of people. For example, in the cloud computing scenario, each cloud server is a resource, a target partition is a specific cluster (cluster), a target object is a target physical server under the specific cluster, and each cloud server is allocated to the target physical server of the target partition for processing, so as to meet the requirement of data computing. However, how to effectively allocate resources and improve the accuracy of allocation of resources is a problem faced by the prior art.
Disclosure of Invention
In order to improve the accuracy of resource allocation, the invention provides a resource allocation method, a resource allocation device, a storage medium and electronic equipment.
The first aspect of the embodiment of the invention discloses a resource allocation method, which comprises the following steps:
receiving an input resource processing request, wherein the resource processing request carries a selection keyword;
Searching target historical information matched with the selected keywords from the historical data; the target history information at least comprises target objects in a target partition and target resources used by the target objects in the target partition;
Importing the target history information into a gradient promotion tree gbdt model to predict whether a target object in the target partition continues to use the target resource;
counting the target quantity of the predicted result of the target object in the target partition, which continuously uses the target resource;
evaluating the target quantity by using an approximation function of the absolute value loss function to obtain an evaluation parameter;
optimizing the gbdt model according to the evaluation parameters;
predicting the resource use condition of the target object in the target partition for the target resource by using the optimized gbdt model;
And distributing the target resource to the target object of the target partition according to the resource use condition and the current information of the target object in the target partition.
The second aspect of the embodiment of the invention discloses a resource allocation device, which comprises:
the receiving unit is used for receiving an input resource processing request, wherein the resource processing request carries a selection keyword;
a search unit for searching for target history information matching the selected keyword from the history data; the target history information at least comprises target objects in a target partition and target resources used by the target objects in the target partition;
the first prediction unit is used for importing the target history information into a gradient lifting tree gbdt model so as to predict whether a target object in the target partition continuously uses the target resource;
the statistics unit is used for counting the target quantity of the prediction results of the target resource which is continuously used by the target object in the target partition;
An evaluation unit for evaluating the target number by using an approximation function of the absolute value loss function to obtain an evaluation parameter;
An optimizing unit, configured to optimize the gbdt model according to the evaluation parameter;
A second prediction unit, configured to predict a resource usage of the target resource by the target object in the target partition using the optimized gbdt model;
and the allocation unit is used for allocating the target resource to the target object of the target partition according to the resource use condition and the current information of the target object in the target partition.
A third aspect of the embodiment of the present invention discloses an electronic device, including:
A processor;
and a memory, wherein the memory stores computer readable instructions, and when the computer readable instructions are executed by the processor, the method for allocating resources is realized.
A fourth aspect of the embodiment of the present invention discloses a computer-readable storage medium storing a computer program that causes a computer to execute a resource allocation method disclosed in the first aspect of the embodiment of the present invention.
The technical scheme provided by the embodiment of the invention can comprise the following beneficial effects:
The resource allocation method provided by the invention comprises the following steps: receiving an input resource processing request, wherein the resource processing request carries a selection keyword; searching target historical information matched with the selected keywords from the historical data, wherein the target historical information at least comprises target objects in the target partition and target resources used by the target objects in the target partition; importing the target history information into a gradient promotion tree gbdt model to predict whether a target object in a target partition continues to use a target resource; counting the target quantity of the prediction results of the target objects in the target partition for continuing to use the target resources; evaluating the target quantity by using an approximation function of the absolute value loss function to obtain an evaluation parameter; optimizing gbdt the model according to the evaluation parameters; predicting the resource use condition of a target object in the target partition on a target resource by using the optimized gbdt model; and distributing the target resources to the target objects of the target partition according to the use condition of the resources and the current information of the target objects in the target partition.
Under the method, because the target history information is obtained by matching the selected keywords, the target history information has higher credibility; in addition, by using the approximation function to replace the absolute value loss function to evaluate the target number and optimizing the gbdt model, the availability of the gbdt model can be increased; in addition, according to the use condition of the resources and the current information of the target objects in the target partition, the target resources are distributed to the target objects of the target partition, so that the target resources of the target objects distributed to the target partition more accord with the target objects in the target partition, and the distribution accuracy of the resources is improved.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention as claimed.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and together with the description, serve to explain the principles of the invention.
FIG. 1 is a schematic view of an apparatus according to an embodiment of the present invention;
FIG. 2 is a flow chart of a method for resource allocation according to an embodiment of the present invention;
FIG. 3 is a flow chart of another method of resource allocation disclosed in an embodiment of the present invention;
FIG. 4 is a flow chart of yet another method of resource allocation disclosed in an embodiment of the present invention;
Fig. 5 is a schematic structural diagram of a resource allocation device according to an embodiment of the present invention;
Fig. 6 is a schematic structural diagram of another resource allocation device according to an embodiment of the present invention;
fig. 7 is a schematic structural diagram of another resource allocation device according to an embodiment of the present invention.
Detailed Description
Reference will now be made in detail to exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, the same numbers in different drawings refer to the same or similar elements, unless otherwise indicated. The implementations described in the following exemplary examples do not represent all implementations consistent with the invention. Rather, they are merely examples of apparatus and methods consistent with aspects of the invention as detailed in the accompanying claims.
In an exemplary embodiment, the embodiment of the invention can be applied to a cloud computing scenario, wherein a target resource is a certain cloud server, a target partition is a certain specific cluster, a target object is a physical server under the specific cluster, and in order to meet the requirement of data computing, the cloud server can be allocated to the physical server under the specific cluster according to a preset rule. In another exemplary embodiment, the embodiment of the present invention may be applied in a business transaction scenario, where the target resource is a certain insurance policy, the target partition is a certain specific city, and the target object is an applicant in the specific city, and in order to satisfy the allocation of the number of the additional insurance, the insurance policy may be allocated to the applicant in the specific city according to a preset rule.
The implementation environment of the invention can be an electronic device, such as a smart phone, a tablet computer, a desktop computer.
Fig. 1 is a schematic structural view of an apparatus according to an embodiment of the present invention. The apparatus 100 may be the electronic device described above. As shown in fig. 1, the apparatus 100 may include one or more of the following components: a processing component 102, a memory 104, a power supply component 106, a multimedia component 108, an audio component 110, a sensor component 114, and a communication component 116.
The processing component 102 generally controls overall operation of the device 100, such as operations associated with display, telephone calls, data communications, camera operations, and recording operations, among others. The processing component 102 may include one or more processors 118 to execute instructions to perform all or part of the steps of the methods described below. Further, the processing component 102 can include one or more modules to facilitate interactions between the processing component 102 and other components. For example, the processing component 102 may include a multimedia module for facilitating interaction between the multimedia component 108 and the processing component 102.
The memory 104 is configured to store various types of data to support operations at the apparatus 100. Examples of such data include instructions for any application or method operating on the device 100. The Memory 104 may be implemented by any type of volatile or non-volatile Memory device or combination thereof, such as static random access Memory (Static Random Access Memory, SRAM), electrically erasable Programmable Read-Only Memory (ELECTRICALLY ERASABLE PROGRAMMABLE READ-Only Memory, EEPROM), erasable Programmable Read-Only Memory (Erasable Programmable Read Only Memory, EPROM), programmable Read-Only Memory (PROM), read-Only Memory (ROM), magnetic Memory, flash Memory, magnetic disk, or optical disk. Also stored in the memory 104 are one or more modules configured to be executed by the one or more processors 118 to perform all or part of the steps in the methods shown below.
The power supply assembly 106 provides power to the various components of the device 100. The power components 106 may include a power management system, one or more power sources, and other components associated with generating, managing, and distributing power for the device 100.
The multimedia component 108 includes a screen between the device 100 and the user that provides an output interface. In some embodiments, the screen may include a Liquid Crystal Display (LCD) and a touch panel. If the screen includes a touch panel, the screen may be implemented as a touch screen to receive input signals from a user. The touch panel includes one or more touch sensors to sense touches, swipes, and gestures on the touch panel. The touch sensor may sense not only the boundary of a touch or sliding action, but also the duration and pressure associated with the touch or sliding operation. The screen may also include an Organic LIGHT EMITTING DISPLAY (OLED for short).
The audio component 110 is configured to output and/or input audio signals. For example, the audio component 110 includes a Microphone (MIC) configured to receive external audio signals when the device 100 is in an operational mode, such as a call mode, a recording mode, and a voice recognition mode. The received audio signals may be further stored in the memory 104 or transmitted via the communication component 116. In some embodiments, the audio component 110 further comprises a speaker for outputting audio signals.
The sensor assembly 114 includes one or more sensors for providing status assessment of various aspects of the device 100. For example, the sensor assembly 114 may detect an on/off state of the device 100, a relative positioning of the assemblies, the sensor assembly 114 may also detect a change in position of the device 100 or a component of the device 100, and a change in temperature of the device 100. In some embodiments, the sensor assembly 114 may also include a magnetic sensor, a pressure sensor, or a temperature sensor.
The communication component 116 is configured to facilitate communication between the apparatus 100 and other devices in a wired or wireless manner. The device 100 may access a Wireless network based on a communication standard, such as WiFi (Wireless-Fidelity). In an embodiment of the present invention, the communication component 116 receives a broadcast signal or broadcast-related information from an external broadcast management system via a broadcast channel. In an embodiment of the present invention, the Communication component 116 further includes a Near Field Communication (NFC) module for facilitating short range communications. For example, the NFC module may be implemented based on radio frequency identification (Radio Frequency Identification, RFID) technology, ultra Wideband (UWB) technology, bluetooth technology, and other technologies.
In an exemplary embodiment, the apparatus 100 may be implemented by one or more Application SPECIFIC INTEGRATED Circuits (ASICs), digital signal processors, digital signal processing devices, programmable logic devices, field programmable gate arrays, controllers, microcontrollers, microprocessors, or other electronic components for performing the methods described below.
Referring to fig. 2, fig. 2 is a flow chart of a resource allocation method according to an embodiment of the invention. As shown in fig. 2, the method may include the steps of:
201. an input resource processing request is received.
In the embodiment of the invention, the resource processing request carries the selection keyword. In an embodiment of the present invention, the resource processing request is a message requesting processing of a resource. In the context of cloud computing, a resource processing request is a message requesting allocation to a cloud server. In the business transaction scenario, the resource processing request is a message that allocates an insurance policy. In the embodiment of the present invention, the selection keyword is a word for representing a condition of selecting the history data. In one embodiment, it may be selected by a user from a drop-down list comprising a plurality of candidate selection criteria, wherein the candidate selection criteria in the drop-down list is a union of keywords of history information in the history data, such as a city, B resource, or C time, etc., and embodiments of the present invention are not limited.
As an alternative embodiment, the resource allocation device may be provided with an image capturing module, and after the input resource processing request is received in step 201 and before the target history information matching the selection keyword is searched for in the history data in step 202, the following steps may be further performed:
starting a camera module;
Shooting to obtain a face image of a user using the resource allocation device;
identifying a face image to obtain identity information of the user;
Acquiring an authority list of the resource allocation device;
Judging whether the identity information has the authority or not according to the authority list;
If the authority is provided, executing step 202 to search the history data for the target history information matched with the selected keyword;
If the user does not have the authority, a prompt message is output to prompt the user that the user does not have the authority.
In this embodiment, the face image of the user using the resource allocation device is acquired, the face image is identified to acquire the identity information of the user, then whether the identity information has authority is judged according to the authority table of the resource allocation device, if so, step 202 is executed to search the history data for the target history information matched with the selected keyword; if the user does not have the right, outputting prompt information to prompt the user that the user does not have the right; the safety of the historical data can be ensured, and the historical data is prevented from being leaked.
202. And searching target historical information matched with the selected keywords from the historical data.
In the embodiment of the invention, the target history information at least comprises a target object in the target partition and a target resource used by the target object in the target partition. For example, in the cloud computing scenario, the target partition is a specific cluster D, the target object is a physical server under the specific cluster D, and the target resource is a cloud server E used by the physical server under the specific cluster D. Also for example, in a business transaction scenario, the target partition is a specific city F, the target object is an applicant in the specific city F, and the target resource is an insurance policy G applied by the applicant in the specific city F. The embodiment of the invention is assumed to be applied to a business handling scene, wherein the target object in the target partition at least comprises the age, sex and income of the applicant in the specific city F, and the target resource used by the target object in the target partition at least comprises the value of the insurance risk G and the age bracket of the user group aimed at by the insurance risk G.
In the embodiment of the invention, it can be understood that when the target resource is allocated, related data is often stored as historical data. Therefore, in the embodiment of the invention, the resource allocation device can search the history data for the target history information matched with the selection keyword.
203. The target history information is imported into the gradient lift tree gbdt model to predict whether the target object in the target partition continues to use the target resource.
In the embodiment of the present invention, gbdt models may be used to predict whether the target object in the target partition continues to use the target resource, and the whole prediction process may be implemented in Python, where Python is a dynamic, object-oriented scripting language, and the embodiment of the present invention is not limited. In the embodiment of the invention, gbdt models are utilized to predict whether the target object in the target partition continues to use the target resource. For example, in the context of cloud computing, it may be understood that the gbdt model is utilized to predict whether a physical server under a particular cluster D continues to use cloud server E. Also for example, in a business transaction scenario, it can be understood that gbdt models are utilized to predict whether an applicant in a particular city F will continue to underwent underwriting of insurance policy G.
204. And counting the target number of the predicted results of the target objects in the target partition which continue to use the target resources.
In the embodiment of the present invention, the resource allocation device may accumulate the predicted result of the target object predicted in step 203 to continue to use the target resource, so as to count the target number, and the whole statistical process may be implemented in Python. In the embodiment of the present invention, for example, in the context of cloud computing, the target number may be understood as the number of cloud servers E that are predicted by using gbdt models. Also for example, in a business transaction scenario, the target number may be understood as the number of added protectors predicted using the gbdt model.
205. The target number is evaluated using an approximation function of the absolute value loss function to obtain an evaluation parameter.
In the embodiment of the invention, the loss function can be used for evaluating the degree of difference between the predicted value and the actual value of the model, wherein the smaller the obtained evaluation parameter is, the smaller the degree of difference between the predicted value and the actual value is, and the better the performance of the model is further indicated.
In the embodiment of the present invention, it can be understood that the resource allocation device predicts whether the target object in the target partition continues to use the target resource by using the gbdt model, but since it cannot be determined whether the prediction result is accurate, it is necessary to evaluate the prediction result by using the absolute value loss function, and then adjust the gbdt model according to the evaluation result. However, due to the property that the absolute value loss function is not conductive, the gbdt model is not available, so in the embodiment of the invention, the resource allocation device can replace the absolute value loss function by using an approximate function of the absolute value loss function, evaluate the predicted target quantity, and the smaller the obtained evaluation parameter is, the more accurate the predicted result of the gbdt model is.
As an alternative embodiment, after the objective number is evaluated with the approximation function of the absolute value loss function in step 205 to obtain the evaluation parameter, and before step 206 optimizes gbdt the model according to the evaluation parameter, the following steps may be further performed:
judging whether the evaluation parameter is smaller than a preset threshold value or not;
If the target object is smaller than the preset threshold value, predicting the resource use condition of the target object in the target partition on the target resource by directly utilizing gbdt models;
If the evaluation parameter is greater than or equal to the preset threshold, step 206 is performed to optimize gbdt the model according to the evaluation parameter.
In this alternative embodiment, before optimizing gbdt the model according to the evaluation parameter, it is determined whether the evaluation parameter is smaller than a preset threshold, if so, it indicates that the prediction accuracy of the gbdt model is already satisfied, that is, the difference between the predicted value predicted by the gbdt model and the actual value is smaller than the preset difference, so the resource allocation device may directly predict the resource usage of the target object in the target partition for the target resource by using the gbdt model.
206. The gbdt model is optimized according to the evaluation parameters.
In the embodiment of the invention, the resource allocation device evaluates the target quantity by using the approximate function of the absolute value loss function to obtain the evaluation parameter, and then optimizes the gbdt model based on the evaluation parameter, so that the degree of difference between the predicted value predicted by the gbdt model and the true value meets the requirement, that is, the difference between the predicted value and the true value is smaller than the preset difference, wherein the preset difference can be set by a tester according to a large number of experimental results.
207. And predicting the resource use condition of the target object in the target partition on the target resource by using the optimized gbdt model.
In the embodiment of the invention, after the gbdt model is optimized according to the evaluation parameter, the resource allocation device can predict the resource use condition of the target object in the target partition for the target resource by using the optimized gbdt model, and the result obtained by prediction is accurate because the optimization is already performed. In the embodiment of the present invention, the resource usage condition may be the number of the predicted results of using the target resource by the target object in the target partition, which is not limited by the embodiment of the present invention.
208. And distributing the target resources to the target objects of the target partition according to the use condition of the resources and the current information of the target objects in the target partition.
In the embodiment of the invention, the resource allocation device may allocate the target resource to the target object of the target partition according to the use condition of the resource (such as the number of the predicted target object in the target partition using the target resource) and the current information of the target object in the target partition (such as the credit of the target object).
It can be seen that, when implementing the method described in fig. 2, since the target history information is obtained by matching the selected keywords, the target history information tends to have higher reliability; in addition, by using the approximation function to replace the absolute value loss function to evaluate the target number and optimizing the gbdt model, the availability of the gbdt model can be increased; in addition, according to the use condition of the resources and the current information of the target objects in the target partition, the target resources are distributed to the target objects of the target partition, so that the target resources of the target objects distributed to the target partition more accord with the target objects in the target partition, and the distribution accuracy of the resources is improved.
Referring to fig. 3, fig. 3 is a flow chart illustrating another resource allocation method according to an embodiment of the invention. As shown in fig. 3, the method may include the steps of:
301. an input resource processing request is received.
In the embodiment of the invention, the resource processing request carries the selection keyword.
302. And searching target historical information matched with the selected keywords from the historical data.
In the embodiment of the invention, the target history information at least comprises a target object in the target partition and a target resource used by the target object in the target partition.
303. Judging whether the data format of the target history information is a designated format or not; if not, go to step 304; if so, step 305 is performed.
In the embodiment of the invention, the designated format may be an ORC File format, wherein the ORC File is a File storage format of the hive database, so that the performance of reading, writing and processing data of the hive database can be improved. In the embodiment of the present invention, the resource allocation device may determine whether the data format of the target history information is the ORC File format, if so, directly execute step 305 to import the target history information with the data format being the specified format into the hive database, and if not, execute step 304 to convert the data format of the target history information into the ORC File format.
304. Converting the data format of the target history information into a specified format;
305. and importing the target history information with the data format being the specified format into a hive database.
In the embodiment of the present invention, step 303 to step 305 are implemented, by first determining whether the data format of the target history information is the ORC File format, if not, converting the data format of the target history information into the ORC File format, so as to increase the readability of the target history information imported into the hive database and improve the importing efficiency.
306. And in the hive database, performing missing value processing on the target historical information with the data format being the specified format to obtain first target information.
In the embodiment of the invention, the acquired target historical information may cause one or more data to be missing due to some reasons, such as errors in the data acquisition process or data not stored in the historical data, and in the use of the gbdt model, if we directly neglect the problem, an abnormality may be triggered, so that the gbdt model is not available. Therefore, before using gbdt model, the missing value processing may be performed on the target history information to obtain the first target information. In the embodiment of the present invention, the processing manner of the missing value may be deleting a row containing the missing value, or deleting a column containing the missing value, or deleting data containing the missing value according to a preset condition, for example, all rows with null values, and rows with the number of missing values greater than a specified threshold, etc., which are not limited in the embodiment of the present invention.
307. And performing outlier processing on the first target information to obtain second target information.
In the embodiment of the invention, the obtained target history information may have some unreasonable values for some reasons, for example, the age is-1, which is an unreasonable value, and in use of the gbdt model, if we directly neglect the problem, an anomaly may be triggered, so that the gbdt model is not available. Thus, prior to using the gbdt model, outlier processing may be performed on the first target information to obtain the second target information. In the embodiment of the present invention, the processing method of the outlier may be to correct the outlier by using the average value, or may be to treat the outlier as a missing value by using the processing method of the missing value, which is not limited in the embodiment of the present invention.
308. And discretizing the second target information to obtain third target information.
In the embodiment of the present invention, since the gbdt model is developed based on discrete data, the second target information may be discretized to obtain the third target information before the gbdt model is used. In the embodiment of the present invention, the discretization processing method may be an equidistant method or an equal frequency method, which is not limited in the embodiment of the present invention.
In the embodiment of the invention, the steps 304-308 are implemented, and the missing value processing, the abnormal value processing and the discretization processing are sequentially carried out on the target historical information, so that the normal operation of the gbdt model after the gbdt model is imported can be ensured, and the prediction result of the gbdt model is more stable.
309. Third target information is imported into the gradient lift tree gbdt model to predict whether target objects in the target partition continue to use the target resources.
310. And marking the predicted result of the target object in the target partition, which continues to use the target resource, with a first identifier.
311. And marking the predicted result of the target object in the target partition, which does not continue to use the target resource, with a second identifier.
In the embodiment of the invention, the first identifier and the second identifier are both of the types of identifiers readable by a computer.
In the embodiment of the present invention, since the simplest marks readable by a computer are "1" and "0", the resource allocation device may mark the prediction result that the target object in the target partition continues to use the target resource as "1", and the prediction result that the target object in the target partition does not continue to use the target resource as "0", which is not limited by the embodiment of the present invention.
312. And classifying all the prediction results according to the identification types.
313. The number of the predicted results marked with the first identifier is counted, and the number of the predicted results marked with the first identifier is used as the target number of the predicted results of the target resource which is continuously used by the target object in the target partition.
In the embodiment of the invention, the resource allocation device can distinguish the first mark '1' from the second mark '0', and then directly count the number of the first marks '1', namely the target number.
In the embodiment of the present invention, steps 310 to 313 are implemented, where the prediction result that the target object in the target partition continues to use the target resource and the prediction result that the target object in the target partition does not continue to use the target resource are marked with the first identifier "1" and the second identifier "0", respectively, so that the readability of the computer for reading the prediction result can be increased, and the working efficiency is improved.
314-317; Steps 314 to 317 are the same as steps 205 to 208 in the second embodiment, and are not described herein.
It can be seen that, compared with the implementation of the method described in fig. 2, by implementing the method described in fig. 3, the readability of the target history information imported into the hive database can be increased and the import efficiency can be improved by converting the data format of the target history information into a specified format. In addition, the missing value processing, the abnormal value processing and the discretization processing are sequentially carried out on the target historical information, so that the gbdt model can be ensured to normally operate after the gbdt model is imported, and the prediction result of the gbdt model is more stable. In addition, the prediction result is marked by the identification type readable by the computer, so that the readability of the computer for reading the prediction result can be improved, and the working efficiency is improved.
Referring to fig. 4, fig. 4 is a flowchart illustrating another resource allocation method according to an embodiment of the present invention. As shown in fig. 4, the method may include the steps of:
401-404; steps 401 to 404 are the same as steps 201 to 204 in the second embodiment, and are not described herein.
405. The historical number of target resources of the target partition in the historical data is obtained.
406. And substituting the target quantity and the historical quantity into an approximation function of the absolute value loss function for evaluation to obtain an evaluation parameter.
In the embodiment of the invention, the closer the predicted target number is to the corresponding historical number of the target resources of the target partition, namely the smaller the difference between the predicted target number and the corresponding historical number of the target resources of the target partition, the smaller the obtained evaluation parameter is, which indicates that the more accurate the predicted result of the gbdt model is.
In the embodiment of the present invention, the absolute value loss function may be as shown in formula (1):
where i denotes a partition number, n denotes a partition number, y i denotes a history number corresponding to a target resource of an i-th partition in the history data, and y' i denotes a target number of a predicted i-th partition.
In the embodiment of the present invention, it may be understood that, due to the property that the absolute value loss function is not conductive, the gbdt model is not available, so in the embodiment of the present invention, as an alternative implementation manner, the resource allocation device may replace the absolute value loss function with an approximate function of the absolute value loss function, and evaluate the predicted target number, where the approximate function of the absolute value loss function is shown in formula (2):
wherein i represents the partition number, n represents the number of partitions, y i represents the historical number corresponding to the target resource of the ith partition in the historical data, y' i represents the target number of the ith partition, and alpha represents the controllable precision.
By implementing the embodiment, the estimated target quantity is estimated by replacing the absolute value loss function with the approximate function, so that the usability of the gbdt model can be increased, and the problem that the gbdt model is not available due to the fact that the absolute value loss function is not conductive is solved.
In the embodiment of the invention, step 405-step 406 are implemented, and the estimated parameters are obtained by substituting the predicted target number and the historical number corresponding to the target resource of the target partition in the historical data into the approximate function of the absolute value loss function, so that the problem that gbdt models are not available due to the fact that the absolute value loss function is not conductive is solved, and the accuracy of the estimated parameters is improved.
407. The gbdt model is optimized according to the evaluation parameters.
408. And predicting the resource use condition of the target object in the target partition on the target resource by using the optimized gbdt model.
409. And determining the quantity to be allocated of the target resources according to the use condition of the resources.
410. And determining the weight value of each target object in the target partition according to the current information of the target objects in the target partition.
In the embodiment of the invention, the current information of the target object in the target partition can be the reliability of the target object in the target partition, namely, the higher the reliability is, the larger the weight value is; the current information of the target object in the target partition may also be a resource usage emergency coefficient of the target object in the target partition, that is, the larger the resource usage emergency coefficient is, the more urgent the target object is for the use of the resource, the larger the weight value is, which is not limited by the embodiment of the present invention.
411. And distributing the target resources with the quantity to be distributed to the target objects of the target partition according to the sequence of the weight values from large to small.
In the embodiment of the present invention, steps 409 to 411 are implemented to provide a method for allocating target resources to target objects of a target partition, which includes determining a to-be-allocated number of target resources according to a resource usage condition, and then determining a weight value of each target object in the target partition according to current information of the target objects in the target partition, where the weight value is greater, so that the to-be-allocated number of target resources is allocated to the target objects of the target partition, thereby further improving accuracy of resource allocation.
It can be seen that, compared with the implementation of the method described in fig. 2, the implementation of the method described in fig. 4, by substituting the predicted target number and the historical number corresponding to the target resource of the target partition in the historical data into the approximate function of the absolute value loss function, the evaluation parameter is obtained, which solves the problem that the gbdt model is not available due to the non-conduction of the absolute value loss function, and improves the accuracy of the evaluation parameter. In addition, the method for distributing the target resources to the target objects of the target partition is provided, and the accuracy of resource distribution is further improved.
Referring to fig. 5, fig. 5 is a schematic structural diagram of a resource allocation device according to an embodiment of the present invention. As shown in fig. 5, the resource allocation apparatus may include: a receiving unit 501, a searching unit 502, a first predicting unit 503, a statistics unit 504, an evaluation unit 505, an optimization unit 506, a second predicting unit 507, and an allocation unit 508, wherein,
A receiving unit 501, configured to receive an input resource processing request.
In the embodiment of the invention, the resource processing request carries the selection keyword.
And a search unit 502 for searching for target history information matching the selected keyword from the history data.
In the embodiment of the invention, the target history information at least comprises a target object in the target partition and a target resource used by the target object in the target partition.
As an alternative embodiment, the receiving unit 501 may be provided with an image capturing module, and after the receiving unit 501 receives the input resource processing request and before the searching unit 502 searches the history data for the target history information matching the selection keyword, the following steps may be further performed:
starting a camera module;
Shooting to obtain a face image of a user using the resource allocation device;
identifying a face image to obtain identity information of the user;
Acquiring an authority list of the resource allocation device;
Judging whether the identity information has the authority or not according to the authority list;
If the authority is provided, the triggering searching unit 502 searches the history data for target history information matched with the selected keywords;
If the user does not have the authority, a prompt message is output to prompt the user that the user does not have the authority.
In this embodiment, the face image of the user using the resource allocation device is acquired, the face image is identified to obtain the identity information of the user, then whether the identity information has authority is judged according to the authority table of the resource allocation device, and if so, the searching unit 502 is triggered to search the history data for the target history information matched with the selected keyword; if not, the receiving unit 501 outputs a prompt message to prompt the user that the authority is not available; the safety of the historical data can be ensured, and the historical data is prevented from being leaked.
The first prediction unit 503 is configured to import the target history information into the gradient lift tree gbdt model, so as to predict whether the target object in the target partition continues to use the target resource.
A statistics unit 504, configured to count the target number of the prediction results of the target object in the target partition for continuing to use the target resource.
An evaluation unit 505 for evaluating the target number by using an approximation function of the absolute value loss function to obtain an evaluation parameter.
An optimizing unit 506, configured to optimize gbdt the model according to the evaluation parameter.
As an alternative embodiment, after the evaluation unit 505 evaluates the target number with the approximation function of the absolute value loss function to obtain the evaluation parameter, and before the optimization unit 506 optimizes gbdt the model according to the evaluation parameter, the following steps may be further performed:
judging whether the evaluation parameter is smaller than a preset threshold value or not;
If the target object is smaller than the preset threshold value, predicting the resource use condition of the target object in the target partition on the target resource by directly utilizing gbdt models;
If the evaluation parameter is greater than or equal to the preset threshold, the triggering optimization unit 506 optimizes gbdt the steps of the model according to the evaluation parameter.
In this alternative embodiment, before optimizing gbdt the model according to the evaluation parameter, it is determined whether the evaluation parameter is smaller than a preset threshold, if so, it indicates that the prediction accuracy of the gbdt model is already satisfied, that is, the difference between the predicted value predicted by the gbdt model and the actual value is smaller than the preset difference, so that the gbdt model can be directly used to predict the resource usage of the target object in the target partition for the target resource.
And a second prediction unit 507, configured to predict a resource usage of the target resource by the target object in the target partition using the optimized gbdt model.
In the embodiment of the present invention, the resource usage condition may be the number of the predicted results of using the target resource by the target object in the target partition, which is not limited by the embodiment of the present invention.
And the allocation unit 508 is configured to allocate the target resource to the target object in the target partition according to the resource usage and the current information of the target object in the target partition.
In an embodiment of the present invention, the allocation unit 508 may allocate the target resource to the target object of the target partition according to the usage condition of the resource (such as the number of the predicted target object in the target partition using the target resource) and the current information of the target object in the target partition (such as the credit of the target object).
It can be seen that, implementing the resource allocation apparatus described in fig. 5, since the target history information is obtained by matching the selected keywords, the target history information tends to have higher reliability; in addition, by using the approximation function to replace the absolute value loss function to evaluate the target number and optimizing the gbdt model, the availability of the gbdt model can be increased; in addition, according to the use condition of the resources and the current information of the target objects in the target partition, the target resources are distributed to the target objects of the target partition, so that the target resources of the target objects distributed to the target partition more accord with the target objects in the target partition, and the distribution accuracy of the resources is improved.
Referring to fig. 6, fig. 6 is a schematic structural diagram of another resource allocation device according to an embodiment of the present invention. The resource allocation device shown in fig. 6 is optimized by the resource allocation device shown in fig. 5. In comparison with the resource allocation device shown in fig. 5, the resource allocation device shown in fig. 6 may further include: a first processing unit 509, a second processing unit 510, a third processing unit 511, a judging unit 512, a converting unit 513, an importing unit 514, a first marking unit 515 and a second marking unit 516, wherein,
The first processing unit 509 is configured to perform missing value processing on the target history information after the searching unit 502 searches the target history information matching the selection keyword from the history data and before the first predicting unit 503 imports the target history information into the gradient lift tree gbdt model to predict whether the target object in the target partition continues to use the target resource, so as to obtain the first target information.
The second processing unit 510 is configured to perform outlier processing on the first target information to obtain second target information.
The third processing unit 511 is configured to perform discretization processing on the second target information to obtain third target information.
The first prediction unit 503 is specifically configured to import third target information into the gradient lift tree gbdt model, so as to predict whether the target object in the target partition continues to use the target resource.
A judging unit 512 for judging whether the data format of the target history information is a specified format after the searching unit 502 searches the target history information matching the selection keyword from the history data and before the first processing unit 509 performs the missing value processing on the target history information to obtain the first target information.
A conversion unit 513 for converting the data format of the target history information into the specified format when the judgment unit 512 judges that the data format of the target history information is not the specified format.
An importing unit 514, configured to import, when the determining unit 512 determines that the data format of the target history information is the specified format, the target history information whose data format is the specified format into the hive database.
The importing unit 514 is further configured to import the target history information whose data format is the specified format into the hive database after the judging unit 512 judges that the data format of the target history information is not the specified format and the converting unit 513 converts the data format of the target history information into the specified format.
In the embodiment of the invention, the designated format may be an ORC File format, wherein the ORC File is a File storage format of the hive database, so that the performance of reading, writing and processing data of the hive database can be improved.
The first processing unit 509 is configured to perform missing value processing on the target history information to obtain first target information, where a manner of obtaining the first target information is specifically: and in the hive database, performing missing value processing on the target historical information with the data format being the specified format to obtain first target information.
A first marking unit 515, configured to mark the predicted result of the target object in the target partition continuing to use the target resource with the first identifier after the first predicting unit 503 imports the third target information into the gradient lift tree gbdt model to predict whether the target object in the target partition continues to use the target resource, and before the statistics unit 504 counts the target number of the predicted result of the target object in the target partition continuing to use the target resource.
And a second marking unit 516, configured to mark the predicted result that the target object in the target partition does not continue to use the target resource with a second identifier.
In the embodiment of the invention, the first identifier and the second identifier are both of the types of identifiers readable by a computer.
In the embodiment of the present invention, since the simplest marks readable by a computer are "1" and "0", the first marking unit 515 may mark the prediction result of the target object in the target partition that continues to use the target resource as "1", and the second marking unit 516 may mark the prediction result of the target object in the target partition that does not continue to use the target resource as "0", which is not limited by the embodiment of the present invention.
The statistics unit 504 includes:
a classification subunit 5041, configured to classify all prediction results according to the identification types;
A statistics subunit 5042, configured to count the number of predictors marked with the first identifier, and take the number of predictors marked with the first identifier as a target number of predictors that the target object in the target partition continues to use the target resource.
It can be seen that, compared with the resource allocation apparatus described in fig. 5, implementing the resource allocation apparatus described in fig. 6 can increase the readability of the target history information imported into the hive database by converting the data format of the target history information into a specified format, and improve the importing efficiency. In addition, the missing value processing, the abnormal value processing and the discretization processing are sequentially carried out on the target historical information, so that the gbdt model can be ensured to normally operate after the gbdt model is imported, and the prediction result of the gbdt model is more stable. In addition, the prediction result is marked by the identification type readable by the computer, so that the readability of the computer for reading the prediction result can be improved, and the working efficiency is improved.
Referring to fig. 7, fig. 7 is a schematic structural diagram of another resource allocation device according to an embodiment of the present invention. The resource allocation device shown in fig. 7 is optimized by the resource allocation device shown in fig. 5. Compared with the resource allocation device shown in fig. 5:
The distribution unit 508 includes:
a first determining subunit 5081, configured to determine, according to a resource usage situation, a to-be-allocated number of the target resource;
A second determining subunit 5082, configured to determine a weight value of each target object in the target partition according to the current information of the target object in the target partition;
The allocation subunit 5083 is configured to allocate the number of target resources to be allocated to the target objects of the target partition in the order from the large weight value to the small weight value.
In the embodiment of the invention, the current information of the target object in the target partition can be the reliability of the target object in the target partition, namely, the higher the reliability is, the larger the weight value is; the current information of the target object in the target partition may also be a resource usage emergency coefficient of the target object in the target partition, that is, the larger the resource usage emergency coefficient is, the more urgent the target object is for the use of the resource, the larger the weight value is, which is not limited by the embodiment of the present invention.
The evaluation unit 505 includes:
An obtaining subunit 5051, configured to obtain a historical number of target resources of the target partition in the historical data;
An evaluation subunit 5052 is configured to perform evaluation by substituting the target number and the historical number into the approximate function of the absolute value loss function, so as to obtain an evaluation parameter.
In the embodiment of the present invention, the absolute value loss function is shown in formula (1) in the fourth embodiment, where i represents the partition number, n represents the number of partitions, y i represents the historical number corresponding to the target resource of the i-th partition in the historical data, and y' i represents the target number of the i-th partition to be predicted. In the embodiment of the present invention, it may be understood that, due to the property that the absolute value loss function is not conductive, the gbdt model may not be available, so as to replace the absolute value loss function with an approximation function of the absolute value loss function, as shown in equation (2) in the fourth embodiment, where i represents the partition number, n represents the partition number, y i represents the historical number corresponding to the target resource of the i-th partition in the historical data, y' i represents the target number of the i-th partition to be predicted, and α represents the controllable accuracy, which is an alternative implementation manner. By implementing the embodiment, the estimated target quantity is estimated by replacing the absolute value loss function with the approximate function, so that the usability of the gbdt model can be increased, and the problem that the gbdt model is not available due to the fact that the absolute value loss function is not conductive is solved.
As can be seen, compared with the resource allocation device described in fig. 5, the implementation of the resource allocation device described in fig. 7 obtains the evaluation parameter by substituting the predicted target number and the history number corresponding to the target resource of the target partition in the history data into the approximate function of the absolute value loss function, which solves the problem that the gbdt model is not available due to the non-conduction of the absolute value loss function, and improves the accuracy of the evaluation parameter. In addition, the method for distributing the target resources to the target objects of the target partition is provided, and the accuracy of resource distribution is further improved.
The invention also provides an electronic device, comprising:
A processor;
A memory having stored thereon computer readable instructions which, when executed by a processor, implement a resource allocation method as previously described.
The electronic device may be the apparatus 100 shown in fig. 1.
In an exemplary embodiment, the invention also provides a computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements a resource allocation method as previously indicated.
It is to be understood that the invention is not limited to the precise arrangements and instrumentalities shown in the drawings, which have been described above, and that various modifications and changes may be effected without departing from the scope thereof. The scope of the invention is limited only by the appended claims.
Claims (7)
1. A method of resource allocation, the method comprising:
receiving an input resource processing request, wherein the resource processing request carries a selection keyword;
Searching target historical information matched with the selected keywords from the historical data; the target history information at least comprises target objects in a target partition and target resources used by the target objects in the target partition;
Importing the target history information into a gradient promotion tree gbdt model to predict whether a target object in the target partition continues to use the target resource;
counting the target quantity of the predicted result of the target object in the target partition, which continuously uses the target resource;
evaluating the target quantity by using an approximation function of the absolute value loss function to obtain an evaluation parameter;
optimizing the gbdt model according to the evaluation parameters;
predicting the resource use condition of the target object in the target partition for the target resource by using the optimized gbdt model;
distributing the target resource to the target object of the target partition according to the resource use condition and the current information of the target object in the target partition;
The method for allocating the target resource to the target object of the target partition according to the resource use condition and the current information of the target object in the target partition comprises the following steps: determining the quantity to be allocated of the target resource according to the resource use condition; determining a weight value of each target object in the target partition according to the current information of the target object in the target partition; according to the sequence of the weight values from large to small, distributing the target resources with the quantity to be distributed to the target objects of the target partition;
evaluating the target number with an approximation function of the absolute value loss function to obtain an evaluation parameter, comprising: acquiring the historical quantity of the target resources of the target partition in the historical data; substituting the target quantity and the historical quantity into an approximation function of an absolute value loss function for evaluation so as to obtain an evaluation parameter;
The approximation function of the absolute value loss function is:
where i denotes a partition number, n denotes a partition number, y i denotes the history number of the i-th partition in the history data, y' i denotes the target number of the i-th partition predicted, and α denotes the accuracy of control.
2. The method of claim 1, wherein after searching the historical data for target historical information matching the selection keyword and before the importing the target historical information into a gradient-hoisting tree gbdt model to predict whether a target object in the target partition continues to use the target resource, the method further comprises:
performing missing value processing on the target historical information to obtain first target information;
Performing outlier processing on the first target information to obtain second target information;
discretizing the second target information to obtain third target information;
And importing the target history information into a gradient-lifting tree gbdt model to predict whether a target object in the target partition continues to use the target resource, including:
and importing the third target information into a gradient promotion tree gbdt model to predict whether the target object in the target partition continues to use the target resource.
3. The method according to claim 2, wherein after said searching for target history information matching the selection keyword from the history data and before said missing value processing of the target history information to obtain first target information, the method further comprises:
judging whether the data format of the target history information is a designated format or not;
If the target historical information is not in the appointed format, converting the data format of the target historical information into the appointed format;
importing target history information with a data format being the specified format into a hive database;
And performing missing value processing on the target history information to obtain first target information, including:
and in the hive database, carrying out missing value processing on the target historical information with the data format being the specified format so as to obtain first target information.
4. A method according to claim 2 or 3, wherein after said importing the third target information into the gradient-lifting tree gbdt model to predict whether a target object in the target partition continues to use the target resource, and before counting the target number of predictions that target object in the target partition continues to use the target resource, the method further comprises:
Marking a predicted result of the target object in the target partition, which continuously uses the target resource, by a first identifier;
marking a predicted result that the target object in the target partition does not continue to use the target resource by using a second identifier;
wherein the first and second identifiers are both of a computer-readable identifier type;
And counting the target number of the predicted results of the target object in the target partition continuing to use the target resource, including:
classifying all prediction results according to the identification types;
Counting the number of the predicted results marked with the first mark, and taking the number of the predicted results marked with the first mark as the target number of the predicted results of the target resource, which is continuously used by the target object in the target partition.
5. A resource allocation apparatus, characterized in that the resource allocation apparatus comprises:
the receiving unit is used for receiving an input resource processing request, wherein the resource processing request carries a selection keyword;
a search unit for searching for target history information matching the selected keyword from the history data; the target history information at least comprises target objects in a target partition and target resources used by the target objects in the target partition;
the first prediction unit is used for importing the target history information into a gradient lifting tree gbdt model so as to predict whether a target object in the target partition continuously uses the target resource;
the statistics unit is used for counting the target quantity of the prediction results of the target resource which is continuously used by the target object in the target partition;
An evaluation unit for evaluating the target number by using an approximation function of the absolute value loss function to obtain an evaluation parameter;
An optimizing unit, configured to optimize the gbdt model according to the evaluation parameter;
A second prediction unit, configured to predict a resource usage of the target resource by the target object in the target partition using the optimized gbdt model;
The allocation unit is used for allocating the target resources to the target objects of the target partition according to the resource use condition and the current information of the target objects in the target partition;
The method for allocating the target resource to the target object of the target partition according to the resource use condition and the current information of the target object in the target partition comprises the following steps: determining the quantity to be allocated of the target resource according to the resource use condition; determining a weight value of each target object in the target partition according to the current information of the target object in the target partition; according to the sequence of the weight values from large to small, distributing the target resources with the quantity to be distributed to the target objects of the target partition;
evaluating the target number with an approximation function of the absolute value loss function to obtain an evaluation parameter, comprising: acquiring the historical quantity of the target resources of the target partition in the historical data; substituting the target quantity and the historical quantity into an approximation function of an absolute value loss function for evaluation so as to obtain an evaluation parameter;
The approximation function of the absolute value loss function is:
where i denotes a partition number, n denotes a partition number, y i denotes the history number of the i-th partition in the history data, y' i denotes the target number of the i-th partition predicted, and α denotes the accuracy of control.
6. An electronic device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the method of any one of claims 1-4 when the computer program is executed.
7. A computer-readable storage medium, characterized in that it stores a computer program that causes a computer to execute a resource allocation method according to any one of claims 1 to 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910822837.7A CN110659133B (en) | 2019-09-02 | 2019-09-02 | Resource allocation method and allocation device, storage medium and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910822837.7A CN110659133B (en) | 2019-09-02 | 2019-09-02 | Resource allocation method and allocation device, storage medium and electronic equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110659133A CN110659133A (en) | 2020-01-07 |
| CN110659133B true CN110659133B (en) | 2024-05-14 |
Family
ID=69036620
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910822837.7A Active CN110659133B (en) | 2019-09-02 | 2019-09-02 | Resource allocation method and allocation device, storage medium and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110659133B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113672375B (en) * | 2020-05-15 | 2023-05-19 | 中国联合网络通信集团有限公司 | Resource allocation prediction method, device, equipment and storage medium |
| CN111985726B (en) * | 2020-08-31 | 2023-04-18 | 重庆紫光华山智安科技有限公司 | Resource quantity prediction method and device, electronic equipment and storage medium |
| CN112381234B (en) * | 2020-11-09 | 2024-05-14 | 北京达佳互联信息技术有限公司 | Resource distribution method, device, equipment and computer readable storage medium |
| CN113761335A (en) * | 2020-11-19 | 2021-12-07 | 北京沃东天骏信息技术有限公司 | Resource processing method and device |
| CN114648257A (en) * | 2022-05-23 | 2022-06-21 | 德州市民政局 | Information processing method, device and equipment |
| CN114943456B (en) * | 2022-05-31 | 2024-05-07 | 北京邮电大学 | Resource scheduling method and device, electronic equipment and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103365700A (en) * | 2013-06-28 | 2013-10-23 | 福建师范大学 | Cloud computing virtualization environment-oriented resource monitoring and adjustment system |
| CN107645731A (en) * | 2017-09-21 | 2018-01-30 | 北京邮电大学 | Load-balancing method based on self-organizing resource allocation in a kind of non-orthogonal multiple access system |
| CN109857551A (en) * | 2019-01-09 | 2019-06-07 | 平安科技（深圳）有限公司 | Dispatching method and device, the electronic equipment of Service Source based on cloud computing |
| CN109995573A (en) * | 2018-01-02 | 2019-07-09 | 中国移动通信有限公司研究院 | Determine method, the computational resource allocation method and device for predicting the sample space of computing resource |
| CN109993428A (en) * | 2019-03-28 | 2019-07-09 | 第四范式（北京）技术有限公司 | Resource allocation methods and device |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20170193413A1 (en) * | 2016-01-04 | 2017-07-06 | Bank Of America Corporation | Predictive utilization of resources and alarm system |
| US10579494B2 (en) * | 2018-01-05 | 2020-03-03 | Nec Corporation | Methods and systems for machine-learning-based resource prediction for resource allocation and anomaly detection |
-
2019
- 2019-09-02 CN CN201910822837.7A patent/CN110659133B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103365700A (en) * | 2013-06-28 | 2013-10-23 | 福建师范大学 | Cloud computing virtualization environment-oriented resource monitoring and adjustment system |
| CN107645731A (en) * | 2017-09-21 | 2018-01-30 | 北京邮电大学 | Load-balancing method based on self-organizing resource allocation in a kind of non-orthogonal multiple access system |
| CN109995573A (en) * | 2018-01-02 | 2019-07-09 | 中国移动通信有限公司研究院 | Determine method, the computational resource allocation method and device for predicting the sample space of computing resource |
| CN109857551A (en) * | 2019-01-09 | 2019-06-07 | 平安科技（深圳）有限公司 | Dispatching method and device, the electronic equipment of Service Source based on cloud computing |
| CN109993428A (en) * | 2019-03-28 | 2019-07-09 | 第四范式（北京）技术有限公司 | Resource allocation methods and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110659133A (en) | 2020-01-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110659133B (en) | Resource allocation method and allocation device, storage medium and electronic equipment | |
| US11210613B2 (en) | Method and system for semi-supervised semantic task management from semi-structured heterogeneous data streams | |
| WO2020186786A1 (en) | File processing method and apparatus, computer device and storage medium | |
| WO2020233320A1 (en) | Reminding task allocation method and apparatus, computer device, and storage medium | |
| CN109614238B (en) | Target object identification method, device and system and readable storage medium | |
| WO2021012790A1 (en) | Page data generation method and apparatus, computer device, and storage medium | |
| CN109509017B (en) | User retention prediction method and device based on big data analysis | |
| US10740336B2 (en) | Computerized methods and systems for grouping data using data streams | |
| CN108256718B (en) | Policy service task allocation method and device, computer equipment and storage equipment | |
| CN107622326B (en) | User classification and available resource prediction method, device and equipment | |
| CN110471763B (en) | Scheduling method, system, medium and electronic equipment based on shared object pool | |
| CN110717509B (en) | Data sample analysis method and device based on tree splitting algorithm | |
| CN111400126A (en) | Network service abnormal data detection method, device, equipment and medium | |
| CN111047048B (en) | Energized model training and merchant energizing method and device, and electronic equipment | |
| CN112183953A (en) | Method and device for allocating customer service resources, electronic equipment and storage medium | |
| CN109857967B (en) | Report subscription method and system based on big data | |
| CN114743132A (en) | Target algorithm selection method and device, electronic equipment and storage medium | |
| US20160267586A1 (en) | Methods and devices for computing optimized credit scores | |
| CN105162931A (en) | Method and device for classifying communication numbers | |
| CN110457365B (en) | Time sequence parallelism-based decision tree generation method and device and electronic equipment | |
| CN111143608A (en) | Information pushing method and device, electronic equipment and storage medium | |
| CN111078984B (en) | Network model issuing method, device, computer equipment and storage medium | |
| CN115018608A (en) | Risk prediction method and device and computer equipment | |
| CN112860416A (en) | Annotating task assignment strategy method and device | |
| CN111553749A (en) | Activity push strategy configuration method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |