CN113128810B - Training method, device, equipment and storage medium of scoring model - Google Patents

Training method, device, equipment and storage medium of scoring model Download PDF

Info

Publication number
CN113128810B
CN113128810B CN201911413990.0A CN201911413990A CN113128810B CN 113128810 B CN113128810 B CN 113128810B CN 201911413990 A CN201911413990 A CN 201911413990A CN 113128810 B CN113128810 B CN 113128810B
Authority
CN
China
Prior art keywords
cloud platform
scoring
value
tenant
determining
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
Application number
CN201911413990.0A
Other languages
Chinese (zh)
Other versions
CN113128810A (en
Inventor
钟维坚
何庆
徐海勇
陶涛
尚晶
陈立峰
唐显莉
汪新勇
陈沁�
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
China Mobile Communications Group Co Ltd
China Mobile Information Technology Co Ltd
Original Assignee
China Mobile Communications Group Co Ltd
China Mobile Information Technology Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by China Mobile Communications Group Co Ltd, China Mobile Information Technology Co Ltd filed Critical China Mobile Communications Group Co Ltd
Priority to CN201911413990.0A priority Critical patent/CN113128810B/en
Publication of CN113128810A publication Critical patent/CN113128810A/en
Application granted granted Critical
Publication of CN113128810B publication Critical patent/CN113128810B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
    • G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • G06F17/16—Matrix or vector computation, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
    • G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
    • 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
    • G06F9/5011—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resources being hardware resources other than CPUs, Servers and Terminals
    • G06F9/5016—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resources being hardware resources other than CPUs, Servers and Terminals the resource being the memory
    • 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
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/06—Buying, selling or leasing transactions
    • G06Q30/0645—Rental transactions; Leasing transactions

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Human Resources & Organizations (AREA)
  • Software Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Strategic Management (AREA)
  • Economics (AREA)
  • Development Economics (AREA)
  • Accounting & Taxation (AREA)
  • Mathematical Physics (AREA)
  • Marketing (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Pure & Applied Mathematics (AREA)
  • Mathematical Optimization (AREA)
  • Mathematical Analysis (AREA)
  • Computational Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • Finance (AREA)
  • Educational Administration (AREA)
  • Operations Research (AREA)
  • Computing Systems (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • Artificial Intelligence (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Game Theory and Decision Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Biology (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Evolutionary Computation (AREA)
  • Algebra (AREA)
  • Databases & Information Systems (AREA)
  • Medical Treatment And Welfare Office Work (AREA)

Abstract

The embodiment of the invention discloses a training method, device, equipment and storage medium of a scoring model. The method comprises the following steps: acquiring information of a plurality of cloud platform tenants, wherein the information of each cloud platform tenant comprises a plurality of index data, and a first grading value and a second grading value of each cloud platform tenant determined based on the plurality of index data; training a scoring model based at least on information of a plurality of cloud platform tenants, wherein the scoring model is used for determining model scoring values of the cloud platform tenants according to the information of the cloud platform tenants. According to the embodiment of the invention, the scoring model for evaluating the cloud platform tenant value is trained through the cloud platform tenant index data, and the tenant condition can be rapidly judged based on the cloud platform tenant scoring value output by the trained scoring model.

Description

Training method, device, equipment and storage medium of scoring model
Technical Field
The invention belongs to the field of information processing, and particularly relates to a training method, device and equipment of a scoring model and a storage medium.
Background
In the operation management work of the cloud platform, in order to ensure effective utilization of platform resources, the value of a platform tenant is often required to be analyzed, and the rationality of the tenant application resources is evaluated.
The existing analysis mode is to directly check detailed operation data such as the resource utilization rate and the resource occupation of tenants, make decisions by experience, and lack unified, visual and scientific data indexes to guide operation work and influence the working efficiency.
Therefore, how to simply and efficiently analyze the tenant value of the platform becomes a problem to be solved.
Disclosure of Invention
The embodiment of the invention provides a training method, device, equipment and storage medium of a scoring model, which can rapidly and accurately analyze the tenant value of a cloud platform.
In a first aspect, a training method of a scoring model is provided, where the scoring model is used for determining a model scoring value of a cloud platform tenant according to information of the cloud platform tenant, and the method includes: acquiring information of a plurality of cloud platform tenants, wherein the information of each cloud platform tenant comprises a plurality of index data, and a first grading value and a second grading value of each cloud platform tenant determined based on the plurality of index data; a scoring model is trained based at least on information of a plurality of cloud platform tenants.
In one possible implementation, determining the first scoring value based on the plurality of metric data includes: determining an evaluation factor set according to the plurality of index data, wherein factors in the evaluation factor set comprise at least one dimension of evaluation factors; determining a fuzzy comprehensive judgment matrix according to the evaluation factor set; and determining a first grading value according to the fuzzy comprehensive judgment matrix.
In one possible implementation, determining the second scoring value based on the plurality of metric data includes: determining a factor weight set corresponding to the evaluation factor set; and determining a second grading value according to the fuzzy comprehensive judgment matrix and the factor weight set.
In one possible implementation, training a scoring model based at least on information of a plurality of cloud platform tenants, includes: respectively determining a first predictive scoring value of each cloud platform user in the cloud platform users according to a plurality of index data in the information of the cloud platform tenants; a scoring model is trained based on the plurality of metric data and the first predictive scoring value.
In one possible implementation, training a scoring model based on a plurality of metric data and a first predictive scoring value includes: respectively determining a second predictive score value of each cloud platform user in the plurality of cloud platform tenants according to the plurality of index data and the first predictive score value; and training a scoring model according to the second predicted scoring value and the second scoring value of each cloud platform tenant.
In one possible implementation, the plurality of metric data includes a first resource occupancy metric, a first resource utilization metric, and a first per Zico technical value metric; the plurality of index data is determined by a second index, the second index is determined by information of the plurality of cloud platform tenants, and the second index comprises a second resource occupation index, a second resource utilization index and a second economic and technological value index.
In one possible implementation, the second resource occupancy indicator comprises at least one of: the memory of the sea Du Pu of the tenant application in the preset time period, the number of cores of the central processing unit of the tenant application in the preset time period and the memory of the tenant application in the preset time period.
In one possible implementation, the second resource utilization index includes at least one of: the storage utilization rate of the tenant in the preset time period and the average utilization rate of the central processing unit of the tenant in the preset time period.
In one possible implementation, the second economic and technological value indicator includes at least one of: the number of times of calling an application program interface (Application Program Interface, API) in a preset time period, the number of times of accessing an application in the preset time period and the number of multiplexing applications in the preset time period.
In a second aspect, there is provided a scoring method comprising:
acquiring information of a cloud platform tenant to be scored, wherein the information of the cloud platform tenant comprises a plurality of index data; inputting information of the cloud platform tenant to be scored into a scoring model to obtain a second scoring value; determining an objective scoring value according to the plurality of index data; determining a comprehensive score value of the cloud platform tenant according to the second score value and the objective score value; wherein the scoring model is derived based on the training method in any one of the possible implementations of the first aspect.
In one possible implementation, determining the objective scoring value based on the plurality of metric data includes: determining standard deviation according to the plurality of index data; determining a weight value corresponding to each index data in the plurality of index data based on the standard deviation; determining a feature matrix according to the plurality of index data and the weight value corresponding to each index data; an objective score value is determined based on the ideal solution and the feature matrix.
In a third aspect, a training apparatus for a scoring model is provided, where the scoring model is used to determine a model scoring value of a cloud platform tenant according to information of the cloud platform tenant, and the apparatus includes:
The system comprises an acquisition module, a storage module and a storage module, wherein the acquisition module is used for acquiring information of a plurality of cloud platform tenants, and the information of each cloud platform tenant comprises a plurality of index data, a first grading value and a second grading value of each cloud platform tenant, wherein the first grading value and the second grading value are determined based on the plurality of index data; and the training module is used for training the scoring model at least based on the information of the cloud platform tenants.
In a fourth aspect, there is provided a scoring apparatus comprising:
The cloud platform tenant information evaluation module is used for acquiring information of cloud platform tenants to be scored, wherein the information of the cloud platform tenants comprises a plurality of index data; the scoring model module is used for inputting the information of the cloud platform tenant to be scored into a scoring model to obtain a second scoring value; wherein the scoring model is derived based on the training method in any one of the possible implementations of the first aspect; the objective scoring module is used for determining objective scoring values according to the plurality of index data; and the comprehensive scoring module is used for determining the comprehensive scoring value of the cloud platform tenant according to the second scoring value and the objective scoring value.
In a fifth aspect, there is provided an apparatus comprising: a processor and a memory storing computer program instructions; the processor when executing the computer program instructions implements the method as in the first to second aspects, or any of the possible implementations of the first to second aspects.
In a sixth aspect, there is provided a computer storage medium having stored thereon computer program instructions which, when executed by a processor, implement a method as in the first to second aspects, or any possible implementation of the first to second aspects.
Based on the method, the device, the equipment and the computer storage medium provided by the embodiment of the invention, the scoring model for evaluating the tenant value of the cloud platform is trained through the tenant index data of the cloud platform, and the tenant condition can be rapidly judged based on the cloud platform tenant scoring value output by the trained scoring model, so that the operation efficiency is improved, and the interference of human factors can be avoided.
Drawings
In order to more clearly illustrate the technical solution of the embodiments of the present invention, the drawings that are needed to be used in the embodiments of the present invention will be briefly described, and it is possible for a person skilled in the art to obtain other drawings according to these drawings without inventive effort.
FIG. 1 is a flowchart of a training method of a scoring model according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of a second resource occupancy indicator according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of a second resource utilization index according to an embodiment of the present invention;
FIG. 4 is a second economic and technological value index provided by an embodiment of the present invention;
FIG. 5 is a flowchart of a scoring method according to an embodiment of the present invention;
FIG. 6 is a flowchart illustrating a scoring method according to an embodiment of the present invention;
Fig. 7 is a schematic structural diagram of a tenant health evaluation device according to an embodiment of the present invention;
FIG. 8 is a schematic structural diagram of a training device according to an embodiment of the present invention;
Fig. 9 is a schematic structural diagram of a scoring device according to an embodiment of the present invention;
fig. 10 is a schematic diagram of an exemplary hardware architecture provided by an embodiment of the present invention.
Detailed Description
Features and exemplary embodiments of various aspects of the present invention will be described in detail below, and in order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention will be described in further detail below with reference to the accompanying drawings and the detailed embodiments. It should be understood that the specific embodiments described herein are merely configured to illustrate the invention and are not configured to limit the invention. It will be apparent to one skilled in the art that the present invention may be practiced without some of these specific details. The following description of the embodiments is merely intended to provide a better understanding of the invention by showing examples of the invention.
It is noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
In the operation management work of the cloud platform, in order to ensure effective utilization of resources of the cloud platform, the value of a platform tenant is often required to be analyzed, and the rationality of the tenant application resources is evaluated. The previous working mode in the aspect is to directly check detailed operation data such as the resource utilization rate and the resource occupation of tenants, make judgment according to experience, and lack unified, visual and scientific data indexes to guide operation work, so that the efficiency and the accuracy are not high.
In order to score the cloud platform tenant with efficiency and accuracy, the embodiment of the invention provides a training method of a scoring model, and the training method of the scoring model provided by the embodiment of the invention is described below.
Fig. 1 is a schematic flow chart of a training method of a scoring model according to an embodiment of the present invention. As shown in fig. 1, the execution subject of the method is a server, the scoring model trained by the method is used for determining a model scoring value of a cloud platform tenant according to information of the cloud platform tenant, and the method may include S101-S102, which are specifically shown as follows:
S101, acquiring information of a plurality of cloud platform tenants, wherein the information of each cloud platform tenant comprises a plurality of index data, and a first grading value and a second grading value of each cloud platform tenant determined based on the plurality of index data.
In one embodiment, the plurality of metric data includes a first resource occupancy metric, a first resource utilization metric, and a first per Zico technical value metric; the plurality of index data is determined by a second index, the second index is determined by information of the plurality of cloud platform tenants, and the second index comprises a second resource occupation index, a second resource utilization index and a second economic and technological value index.
Wherein the plurality of index data are determined by a second index which is an index for acquiring and determining 40-dimensional characteristics based on tenant resource occupation, resource utilization, value and meaning, and the 40-dimensional characteristics are determined as primary characteristicsThe second index includes a second resource occupation index, a second resource utilization index, and a second economic and technological value index, and these three indexes are described below respectively.
The first, second resource occupancy indicator comprises at least one of: the memory of the sea Du Pu of the tenant application in the preset time period, the number of cores of the central processing unit of the tenant application in the preset time period and the memory of the tenant application in the preset time period. The specific second resource occupancy index can be seen in FIG. 2
Second, the second resource utilization index includes at least one of: the storage utilization rate of the tenant in the preset time period and the average utilization rate of the central processing unit of the tenant in the preset time period. The specific second resource occupancy index can be seen in FIG. 3
Third, the second economic and technological value index comprises at least one of the following: the number of times of calling an application program interface (Application Program Interface, API) in a preset time period, the number of times of accessing an application in the preset time period and the number of multiplexing applications in the preset time period. The specific second resource occupancy index can be seen in FIG. 4
After the second index is determined, a plurality of index data may be determined based on the second index, the index data including: the first resource occupancy indicator, the first resource utilization indicator, and the first warp Zico technical value indicator. The following is an example of three aspects, and the overall transformation process will be described in detail later.
In a first aspect, a first resource occupancy indicator is determined from a second resource occupancy indicator.
For example, a first resource occupancy indicatorRepresenting the common data storage period, the smaller and the better this feature is, because the memory resource of the interface machine is occupied.
In a second aspect, the first resource utilization index is based on the second resource utilization index.
For example, a first resource utilization indexThe change degree of the memory and the CPU utilization rate of the tenant is respectively represented, and the conversion mode is as follows:
In a third aspect, a first warp Zico technical value indicator is based on a second economic and technological value indicator.
Wherein, the first warp Zico technical value index
Thus, the second index of 40 dimensions, namely the second resource occupation index, the second resource utilization index and the second economic and technological value index. And sequentially processing and outputting index data of 27 dimensions, namely a first resource occupation index, a first resource utilization index and a first warp Zico technical value index.
Based on the determined index data, a first scoring value may be determined, where the first scoring value is actually a five-item score based on five dimensions of resource utilization, resource rationality, economic value of the tenant, scientific and innovative value of the tenant, politics, and administrative significance, and the step of determining the first scoring value may include: determining an evaluation factor set according to the plurality of index data, wherein factors in the evaluation factor set comprise at least one dimension of evaluation factors; determining a fuzzy comprehensive judgment matrix according to the evaluation factor set; and determining a first grading value according to the fuzzy comprehensive judgment matrix.
First, determining an evaluation factor set according to a plurality of index data may specifically include: and determining evaluation factors Fset = { f1, f2, f3, f4, f5} = { resource utilization, resource rationality, tenant economic value, tenant technological innovation value, politics and management significance. The evaluation factor set FES { =fes 1,fes2,..,fes5 } = { a, b, c, d, e }, and the corresponding score as= {95,85,70,50,20}, are determined.
Secondly, determining the fuzzy comprehensive judgment matrix according to the evaluation factor set specifically can comprise: the relevant specialists of operation work score each dimension of the tenant, the number of specialists is rn, the score is ER r,l epsilon FES, the score of the r-th specialist on the first factor of the tenant is represented, and the score set is ER r,l:
the computing expert scores the type ratio of each evaluation factor of the tenant
Thereby determining a fuzzy comprehensive judgment matrix SR:
Where rr is used to represent each type of evaluation, namely fes 1,fes2,..,fes5. The SR is a matrix generated in the process of calculating the evaluation score and is used for representing the probability that the tenant evaluates to a certain class score in each evaluation dimension.
And finally, determining a first grading value according to the fuzzy comprehensive judgment matrix.
dimScored＝as×SR
The second scoring value may be further determined based on the determined index data, and the step of determining the second scoring value may include: determining a factor weight set corresponding to the evaluation factor set; and determining a second grading value according to the fuzzy comprehensive judgment matrix and the factor weight set.
The method comprises the steps of determining a factor weight set corresponding to an evaluation factor set, wherein the factor weight set can be determined by using a hierarchical analysis method, and the hierarchical analysis method is a decision method for carrying out qualitative and quantitative analysis on the basis of decomposing elements related to decision into levels of targets, criteria, schemes and the like. The factor weight set corresponding to the evaluation factor set determined in the invention is fw= { FW1, FW2, …, FW5}.
The second score value determined according to the fuzzy comprehensive judgment matrix and the factor weight set may be fscore =index (max (fw×sr T)), that is, the fuzzy comprehensive judgment is performed on the user, and it is determined that the tenant belongs to the membership degree of the various scores, and the final fuzzy score fscore of the tenant.
S102, training a scoring model at least based on information of a plurality of cloud platform tenants.
In one embodiment, a first predictive score value of each of the cloud platform users is determined according to a plurality of index data in information of a plurality of cloud platform tenants; a scoring model is trained based on the plurality of metric data and the first predictive scoring value.
The first predicted score value Y d of each cloud platform user in the cloud platform users is respectively determined according to a plurality of index data in the information of the cloud platform tenants, the first predicted score value Y d is compared with the first score value dimScore d calculated in the previous step, the parameters of a score model are optimized according to the difference value between the first predicted score value Y d and the first score value dimScore d, and the first predicted score value Y d＝modeld (X') is output based on the score model.
Respectively determining a second predictive score value of each cloud platform user in the plurality of cloud platform tenants according to the plurality of index data and the first predictive score value; and training a scoring model according to the second predicted scoring value and the second scoring value of each cloud platform tenant.
And respectively determining a second predicted score value es of each cloud platform user in the plurality of cloud platform tenants according to the plurality of index data and the first predicted score value Y d, comparing the second predicted score value es with the second score value fscore calculated in the previous step, optimizing parameters of the scoring model according to the difference value between the second predicted score value es and the second score value fscore, guiding the precision of the scoring model to reach a preset precision condition, and outputting a second predicted score value es=model fuzzy (x') based on the scoring model.
Therefore, according to the training method of the scoring model provided by the embodiment of the invention, the scoring model for evaluating the value of the cloud platform tenant is trained through the index data of the cloud platform tenant, and the situation of the tenant can be rapidly judged based on the score value of the cloud platform tenant output by the trained scoring model, so that the operation efficiency can be improved, and the interference of human factors can be avoided.
Fig. 5 is a schematic flow chart of a scoring method provided in an embodiment of the present invention, where an execution subject of the method is a server, and the method may include S501-S504, which specifically include the following steps:
s501, acquiring information of cloud platform tenants to be scored, wherein the information of the cloud platform tenants comprises a plurality of index data.
S502, inputting information of the cloud platform tenant to be scored into a scoring model to obtain a second scoring value.
First, according to index data, first predicted score values yd of cloud platform tenants on 5 factors are obtained respectively
yd＝modeld(x′),d＝1,2,…,5
Secondly, new tenant data x '= { x', y d }, d=1, 2, …,5 are constructed according to the first predictive score value and the index data
Finally, x "is input into the scoring model to determine a second scoring value es.
S503, determining objective grading values according to the plurality of index data.
The objective scoring value is determined according to the plurality of index data, wherein the objective scoring value is obtained by performing weight setting on each index of the tenant by adopting a standard deviation coefficient weighting method and performing objective scoring on the health degree of the tenant by using a TOPSIS ideal solution. The method comprises the following specific steps:
Step 1: confirm that tenant data x= { X i},i＝1,2,…,n,xi contains features
Step 2: determining tenant feature data X' that includes features
Step 3: after dimensionless, weighting each feature by using a standard deviation political trickery method:
Step 3.1: calculating the average value of each feature of the feature matrix X':
Wherein j is used to represent the category of the feature, and i is used to represent the value of the feature of the tenant data.
Step 3.2: solving standard deviation of each characteristic index:
step 3.3: weighting each feature:
Step 4: calculating a weighted feature matrix:
X″＝X′×W
Step 5: finding the best worst solution, finding the best score for each feature as Find the worst score of each feature asThe optimal solution is that the worst solution is:
step 6: calculating the Euclidean distance between each tenant and the optimal worst vector:
step 7: computing relative closeness of tenants:
Step 8: determining an objective score value:
and S504, determining the comprehensive grading value of the cloud platform tenant according to the second grading value and the objective grading value.
The second score value output based on the scoring model is es i, and the objective score value os i output by the objective scoring method is the overall health of the tenant:
In sum, based on index data of cloud platform tenants in tenant resource utilization, resource rationality, tenant economic value, tenant technological innovation value, politics and management significance, the second grading value and the objective grading value output based on the grading model are determined, the grading value of the cloud platform tenants is comprehensively output from two aspects, reliability and stability of analysis results are improved, the problems that the previous evaluation of the health degree of the cloud platform tenants is too subjective and lacks applicability are solved, and the cloud platform tenant analysis method is well suitable for cloud platform tenant analysis work.
Fig. 6 is a flow chart illustrating a scoring method according to an embodiment of the present invention, where the method shown in fig. 6 includes three parts, namely, collection processing of tenant features, establishment of an expert scoring model, and establishment of an objective scoring model.
First, a health model index is confirmed.
Confirming that the health degree model index needs to be subjected to feature processing, wherein the feature processing needs to determine the type and range of the operation features of the collection tenant, and processes and converts the features through feature engineering, and the specific steps are as follows:
Step 1: feature processing requires determining the type and scope of collection tenant operating features. The method collects the total 40-dimension characteristics based on three aspects of tenant resource occupation, resource utilization, value and meaning, and determines the 40-dimension characteristics as first-level characteristics Detailed first-order features/>As shown in fig. 2-4.
Step 2: the aspect of converting the resource occupation is characterized by a secondary characteristic.
Step 2.1: according to the data scale of other companies and the data of various resource scales, a regression prediction model is established, so that the degree of the data scale of the tenant reflected by the secondary characteristics and the applied resource scale are equal:
Step 2.1.1: data of 31 provincial companies and various resource sizes are collected and moved, and the data comprises the characteristics shown in table 1:
TABLE 1 data Scale of companies and construction of various resource Scale data
Step 2.1.2: constructing a training set X, and then:
The x-feature comprises
Step 2.1.3: building training label Y
Step 2.1.3: construction of a Linear regression Model f (X, Y) based on X, Y
Step 2.1.4: to-be-evaluated tenant x 1, predicting the data size by using the constructed regression model:
step 2.1.5: evaluating the actual data volume of the tenant and comparing the actual data volume with the expected processable data scale of the tenant application resource, thereby determining the secondary characteristic
Step 2.2: determining secondary featuresRepresenting the common data storage period, the smaller and the better this feature is, because the memory resource of the interface machine is occupied.
Step 2.3: processing to obtain secondary featuresThe ratios of the physical storage, the physical CPU resource, the physical memory resource and other resources applied by the tenant are respectively represented as follows:
step 2.4: judging whether the tenant has application tool service or not, and determining the secondary characteristics
Step 3: the aspect features of converting resource utilization are secondary features:
step 3.1: the resource utilization standard value utel_base=0.5 is determined.
Step 3.2: processing to obtain secondary featuresThe method respectively represents the reasonability of the CPU utilization rate of the tenant, the reasonability of the memory utilization rate and the reasonability of the storage utilization rate, and the conversion mode is as follows:
Step 3.3: determining secondary features The change degree of the memory and the CPU utilization rate of the tenant is respectively represented, and the conversion mode is as follows:
step 3.4: determining CPU, storage, and memory unit price as cpu_price=1000, volumn_price=150, raw_price=70, thereby determining tenant Respectively representing the resource waste cost generated by the utilization rate of storage, CPU and memory:
Step 4: the conversion tenant value and meaning aspect features are secondary features:
step 4.1: determining a total value cost of resources occupied by tenants:
Step 4.2: determining secondary features
Step 5: if the feature processing method is determined to be Ftransform, after the 40-dimensional tenant features are processed by the Ftransform method, 27-dimensional secondary features are output, specifically, a table is shown below, wherein OD represents the optimal feature direction, if od=1, the larger the feature value is, the better tenant evaluation is, and the opposite is the case when od= -1 is adopted. A detailed 27-dimensional secondary feature, as shown in table 2:
Table 2 27 dimensional secondary features
Secondly, an expert scoring model is established, namely, expert scoring is carried out by using collected tenant data, the expert evaluation method is combined with the fuzzy comprehensive evaluation method and the SVM support vector machine, the fuzzy comprehensive evaluation method is firstly utilized to obtain scores of five important tenant health degree dimensions determined by a scheme on the basis of expert scoring, the fuzzy comprehensive score is obtained, and then the SVM is utilized to train and learn an expert scoring mechanism, so that the capability of automatically obtaining the scores of the five dimensions and the fuzzy comprehensive score is realized.
The specific steps for training expert scoring models are as follows:
Step 1: determining evaluation factor Fset = { f1, f2, f3, f4, f5} = { resource utilization, resource rationality, tenant economic value, tenant technological innovation value, political and administrative significance }
Step2: the determination factor evaluation set fes= { FES 1,fes2,..,fes5 } = { a, b, c, d, e }, and the corresponding score as= {95,85,70,50,20}
Step 3: using analytic hierarchy process, an evaluation factor weight fw= { FW 1,fw2,…,fw5 }, is determined
Step 4: the relevant specialists of operation work score each dimension of the tenant, the number of specialists is rn, the score is ER r,l epsilon FES, the score of the r-th specialist on the first factor of the tenant is represented, and the score set is ER rl:
step 5: the computing expert scores the type ratio of each evaluation factor of the tenant
Thereby determining a fuzzy comprehensive judgment matrix SR:
step 6: determining a tenant factor score ds:
dimScored＝as×SR
Step 7: performing fuzzy comprehensive evaluation to determine membership degrees of tenants belonging to various scores and final fuzzy scores fscore of tenants:
fscore＝index(max(FW×SRT))
Step 8: training a regression model based on an SVM algorithm, and learning an expert scoring mechanism:
step 8.1: tenant data After feature processing and normalization, the/> According to scoring results of 5 dimensions, training sets are normalized and then respectively training 5 regression model d(X′,dimScored), d=1, 2, …,5
Step 8.2: obtaining regression model output
Yd＝modeld(X′),d＝1,2,…,5
Step 8.3: splicing Y d to the original tenant dataset X' to obtain X
Step 8.4: model fuzzy (X', fscore) was trained on fuzzy scoring regression models
Step 9: when it is required to determine its expert score for tenant x:
Step 9.1: tenant data After feature processing and normalization, the/> Then, input X 'to model d, d=1, 2, …,5, get the tenant's score on 5 factors on Fset
yd＝modeld(x′),d＝1,2,…,5
Step 9.2: building new tenant data x "= { x', y d }, d=1, 2, …,5
Step 9.3: output tenant expert evaluation score es=model fuzzy (x')
Finally, an objective scoring model is established, namely, after each index data of the tenant is weighted by using a standard deviation political trickery method, the health score of the tenant is evaluated by adopting a TOPSIS ideal solution, and the specific process is referred to the method and the step shown in the S503.
The flowchart for realizing the scoring method shown in fig. 6 is that firstly, the evaluation is performed from the multi-dimensions of the tenant's own resource occupation, data occupation, utilization, etc. by combining with the life cycle of the tenant, an evaluation model is built, and the tenant health score is output.
Secondly, a fuzzy comprehensive evaluation method is utilized in the expert scoring method, after the expert scores each tenant index, a fuzzy comprehensive judgment matrix is established after the weight of each index is determined, fuzzy evaluation is calculated, and the fuzzy comprehensive evaluation method requires the expert to score the tenant index, so that the step is omitted in the follow-up work, the fuzzy evaluation result is trained by an SVM classification algorithm, and the expert scoring rule knowledge is learned.
Then, the objective scoring model is to set weights of the indexes of the tenants by adopting a standard deviation coefficient weighting method, and then objectively score the health degree of the tenants by using a TOPSIS ideal solution.
Finally, determining that the expert scoring method outputs the tenant score es i and the objective scoring method outputs the score os i, and then the overall health of the tenant is as follows:
Therefore, the analysis results output by the objective scoring model and the expert scoring model are reliable and stable, the problems of excessive subjectivity and lack of applicability of the traditional health evaluation method are changed, and the analysis method is well applicable to cloud platform tenant analysis work.
Fig. 7 is a schematic structural diagram of a tenant health degree evaluation device according to an embodiment of the present invention.
The overall architecture of the tenant health evaluation device shown in fig. 7 includes: the system comprises a data source, a health degree analysis module and a visualization module.
Wherein the data source comprises: cloud platform data comprising information such as the utilization condition of resources, the occupation condition of resources and the like of tenants directly acquired from a cloud platform; tenant data including tenant technological value, management meaning and other tenant project background information; expert knowledge base data.
The health degree analysis module is characterized by three sub-modules, namely, a feature processing method, an objective scoring method and an expert scoring method. And after the tenant data source passes through the feature processing sub-module, outputting and synthesizing by the objective scoring sub-module and the expert scoring sub-module by the two scoring modules to obtain the tenant health score.
The visualization module visualizes the analysis result of the tenant health in the modes of World Wide Web (Web), mobile Application program (app) and client side respectively, and assists operators to analyze the tenant.
According to the tenant health degree assessment device provided by the embodiment of the invention, the scoring model for evaluating the tenant value of the cloud platform can be trained through the tenant index data of the cloud platform, and the tenant condition can be rapidly judged based on the cloud platform tenant scoring value output by the trained scoring model, so that the operation efficiency can be improved, and interference caused by human factors can be avoided.
Fig. 8 is a schematic structural diagram of a training device according to an embodiment of the present invention, and as shown in fig. 8, the device 800 may include:
the obtaining module 810 is configured to obtain information of a plurality of cloud platform tenants, where the information of each cloud platform tenant includes a plurality of index data, and a first score value and a second score value of each cloud platform tenant determined based on the plurality of index data.
The obtaining module 810 is specifically configured to determine an evaluation factor set according to the multiple index data, where factors in the evaluation factor set include at least one dimension of evaluation factor; determining a fuzzy comprehensive judgment matrix according to the evaluation factor set; and determining a first grading value according to the fuzzy comprehensive judgment matrix.
The obtaining module 810 is specifically configured to determine a factor weight set corresponding to the evaluation factor set; and determining a second grading value according to the fuzzy comprehensive judgment matrix and the factor weight set.
Wherein, the plurality of index data in the above-mentioned embodiment of the present invention includes a first resource occupation index, a first resource utilization index and a first Zico technical value index; the plurality of index data is determined by a second index, the second index is determined by information of the plurality of cloud platform tenants, and the second index comprises a second resource occupation index, a second resource utilization index and a second economic and technological value index.
The second resource occupation index in the above-mentioned embodiment of the present invention includes at least one of the following: the memory of the sea Du Pu of the tenant application in the preset time period, the number of cores of the central processing unit of the tenant application in the preset time period and the memory of the tenant application in the preset time period.
Wherein, the second resource utilization index in the above-mentioned embodiment of the present invention includes at least one of the following: the storage utilization rate of the tenant in the preset time period and the average utilization rate of the central processing unit of the tenant in the preset time period.
The second economic and technological value index in the above-mentioned embodiment of the present invention includes at least one of the following: the method comprises the steps of calling times of an application program interface in a preset time period, access times of an application in the preset time period and the number of application multiplexing in the preset time period.
The training module 820 is configured to train the scoring model based at least on information of the plurality of cloud platform tenants.
The training module 820 is specifically configured to determine a first prediction score value of each cloud platform user of the cloud platform users according to a plurality of index data in the information of the plurality of cloud platform tenants; a scoring model is trained based on the plurality of metric data and the first predictive scoring value.
The training module 820 is specifically configured to determine a second predicted score value of each cloud platform user in the plurality of cloud platform tenants according to the plurality of index data and the first predicted score value; and training a scoring model according to the second predicted scoring value and the second scoring value of each cloud platform tenant.
Each module of the training device provided in this embodiment may implement the method in fig. 1, and for brevity description, will not be repeated here. According to the training device provided by the embodiment, the scoring model for evaluating the cloud platform tenant value is trained through the cloud platform tenant index data, and the tenant condition can be rapidly judged based on the cloud platform tenant scoring value output by the trained scoring model, so that the operation efficiency can be improved, and interference caused by human factors can be avoided.
Fig. 9 is a schematic structural diagram of a scoring device according to an embodiment of the present invention, and as shown in fig. 9, the device 900 may include:
the acquiring module 910 is configured to acquire information of a cloud platform tenant to be scored, where the information of the cloud platform tenant includes a plurality of index data.
The scoring model module 920 is configured to input information of the cloud platform tenant to be scored into a scoring model, and obtain a second scoring value.
An objective scoring module 930 for determining objective scoring values based on the plurality of metric data.
And the comprehensive scoring module 940 is configured to determine a comprehensive scoring value of the cloud platform tenant according to the second scoring value and the objective scoring value.
The modules of the scoring device provided in this embodiment may implement the method in fig. 5, and are not described herein for brevity. The scoring device provided by the embodiment of the invention combines the scoring values respectively output by the objective scoring module and the scoring model, comprehensively analyzes and determines the analysis result of the cloud platform group user, has reliable and stable analysis result, changes the problems of excessively subjective scoring and lack of applicability in the past, and can be better suitable for the analysis work of the cloud platform tenant.
Fig. 10 shows an exemplary hardware architecture diagram provided by an embodiment of the present invention.
The processing device may include a processor 1001 and a memory 1002 storing computer program instructions.
The processor 1001 may include a central processing unit (Central Processing Unit, CPU), or Application SPECIFIC INTEGRATED Circuit (ASIC), or may be configured as one or more integrated circuits that implement embodiments of the present invention.
Memory 1002 may include mass storage for data or instructions. By way of example, and not limitation, memory 1002 may include a hard disk drive (HARD DISK DRIVE, HDD), floppy disk drive, flash memory, optical disk, magneto-optical disk, magnetic tape, or universal serial bus (Universal Serial Bus, USB) drive, or a combination of two or more of the foregoing. The memory 1002 may include removable or non-removable (or fixed) media, where appropriate. Memory 1002 may be internal or external to the integrated gateway disaster recovery device, where appropriate. In a particular embodiment, the memory 1002 is a non-volatile solid state memory. In a particular embodiment, the memory 1002 includes Read Only Memory (ROM). The ROM may be mask programmed ROM, programmable ROM (PROM), erasable PROM (EPROM), electrically Erasable PROM (EEPROM), electrically rewritable ROM (EAROM), or flash memory, or a combination of two or more of these, where appropriate.
The processor 1001 implements any of the methods of the embodiments shown in fig. 1-6 described above by reading and executing computer program instructions stored in the memory 1002.
In one example, the processing device may also include a communication interface 1003 and a bus 1010. As shown in fig. 10, the processor 1001, the memory 1002, and the communication interface 1003 are connected to each other by a bus 1010, and perform communication with each other.
The communication interface 1003 is mainly used for implementing communication among the modules, devices, units and/or apparatuses in the embodiment of the invention.
Bus 1010 includes hardware, software, or both, that couples the components of the communication incident processing device to one another. By way of example, and not limitation, the buses may include an Accelerated Graphics Port (AGP) or other graphics bus, an Enhanced Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), a HyperTransport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an infiniband interconnect, a Low Pin Count (LPC) bus, a memory bus, a micro channel architecture (MCa) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a video electronics standards association local (VLB) bus, or other suitable bus, or a combination of two or more of the above. Bus 1010 may include one or more buses, where appropriate. Although embodiments of the invention have been described and illustrated with respect to a particular bus, the invention contemplates any suitable bus or interconnect.
The processing device may perform the training method of the scoring model in the embodiment of the present invention, thereby implementing the method described in connection with fig. 1.
The processing device may perform the scoring method in the embodiment of the present invention, thereby implementing the method described in connection with fig. 5.
In addition, in combination with the scoring method in the above embodiment, the embodiment of the present invention may be implemented by providing a computer storage medium. The computer storage medium has stored thereon computer program instructions; the computer program instructions, when executed by a processor, implement the training method of any of the scoring models in the above embodiments.
In addition, in combination with the scoring method in the above embodiment, the embodiment of the present invention may be implemented by providing a computer storage medium. The computer storage medium has stored thereon computer program instructions; the computer program instructions, when executed by a processor, implement any of the scoring methods of the embodiments described above.
It should be understood that the invention is not limited to the particular arrangements and instrumentality described above and shown in the drawings. For the sake of brevity, a detailed description of known methods is omitted here. In the above embodiments, several specific steps are described and shown as examples. The method processes of the present invention are not limited to the specific steps described and shown, but various changes, modifications and additions, or the order between steps may be made by those skilled in the art after appreciating the spirit of the present invention.
Functional blocks shown in the above-described structural block diagrams may be implemented in software, and elements of the present invention are programs or code segments used to perform the required tasks. The program or code segments may be stored in a machine readable medium or transmitted over transmission media or communication links by a data signal carried in a carrier wave. A "machine-readable medium" may include any medium that can store or transfer information. Examples of machine-readable media include electronic circuitry, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, fiber optic media, radio Frequency (RF) links, and the like. The code segments may be downloaded via computer networks such as the internet, intranets, etc.
It should also be noted that the exemplary embodiments mentioned in this disclosure describe some methods or systems based on a series of steps or devices. The present invention is not limited to the order of the above-described steps, that is, the steps may be performed in the order mentioned in the embodiments, or may be performed in a different order from the order in the embodiments, or several steps may be performed simultaneously.
In the foregoing, only the specific embodiments of the present invention are described, and it will be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working processes of the systems, modules and units described above may refer to the corresponding processes in the foregoing method embodiments, which are not repeated herein. It should be understood that the scope of the present invention is not limited thereto, and any equivalent modifications or substitutions can be easily made by those skilled in the art within the technical scope of the present invention, and they should be included in the scope of the present invention.

Claims (10)

1. A training method of a scoring model, wherein the scoring model is used for determining a model scoring value of a cloud platform tenant according to information of the cloud platform tenant, and the method comprises:
acquiring information of a plurality of cloud platform tenants, wherein the information of each cloud platform tenant comprises a plurality of index data, and a first grading value and a second grading value of each cloud platform tenant determined based on the plurality of index data;
Training the scoring model based at least on information of the plurality of cloud platform tenants;
wherein determining the first and second scoring values based on the plurality of metric data comprises:
Determining an evaluation factor set according to the plurality of index data, wherein factors in the evaluation factor set comprise evaluation factors of at least one dimension; determining a fuzzy comprehensive judgment matrix according to the evaluation factor set; determining the first grading value according to the fuzzy comprehensive judgment matrix;
determining a factor weight set corresponding to the evaluation factor set; determining the second grading value according to the fuzzy comprehensive judgment matrix and the factor weight set;
the training the scoring model based at least on information of the plurality of cloud platform tenants includes:
respectively determining a first predictive scoring value of each cloud platform tenant in the cloud platform tenants according to the index data in the information of the cloud platform tenants;
training the scoring model based on the plurality of metric data and the first predictive scoring value.
2. The method of claim 1, wherein the training the scoring model based on the plurality of metric data and the first predictive scoring value comprises:
Respectively determining a second predictive score value of each cloud platform tenant in the plurality of cloud platform tenants according to the plurality of index data and the first predictive score value;
And training the scoring model according to the second predicted scoring value and the second scoring value of each cloud platform tenant.
3. The method of any one of claims 1 to 2, wherein the plurality of metric data includes a first resource occupancy metric, a first resource utilization metric, and a first per Zico technical value metric; the plurality of index data are determined by second indexes, the second indexes are determined by information of the plurality of cloud platform tenants, and the second indexes comprise a second resource occupation index, a second resource utilization index and a second economic and technological value index.
4. A method according to claim 3, wherein the second resource occupancy indicator comprises at least one of: the memory of the sea Du Pu of the tenant application in the preset time period, the number of cores of the central processing unit of the tenant application in the preset time period and the memory of the tenant application in the preset time period.
5. A scoring method, the method comprising:
Acquiring information of a cloud platform tenant to be scored, wherein the information of the cloud platform tenant comprises a plurality of index data;
Inputting the information of the cloud platform tenant to be scored into a scoring model to obtain a second scoring value;
determining an objective scoring value according to the plurality of index data;
determining a comprehensive grading value of the cloud platform tenant according to the second grading value and the objective grading value;
wherein the scoring model is derived based on the training method of any one of claims 1 to 4.
6. The method according to claim 5, wherein the determining objective score values from the plurality of index data comprises:
determining a standard deviation according to the plurality of index data;
determining a weight value corresponding to each index data in the plurality of index data based on the standard deviation;
Determining a feature matrix according to the index data and the weight value corresponding to each index data;
the objective scoring values are determined based on an ideal solution and the feature matrix.
7. A training apparatus of a scoring model for determining a model scoring value of a cloud platform tenant according to information of the cloud platform tenant, the apparatus comprising:
the system comprises an acquisition module, a judgment module and a judgment module, wherein the acquisition module is used for acquiring information of a plurality of cloud platform tenants, the information of each cloud platform tenant comprises a plurality of index data, and a first grading value and a second grading value of each cloud platform tenant are determined based on the plurality of index data;
The training module is used for training the scoring model at least based on the information of the cloud platform tenants;
the acquisition module is specifically configured to determine an evaluation factor set according to the multiple index data, where factors in the evaluation factor set include at least one dimension of evaluation factor; determining a fuzzy comprehensive judgment matrix according to the evaluation factor set; determining the first grading value according to the fuzzy comprehensive judgment matrix; determining a factor weight set corresponding to the evaluation factor set; determining the second grading value according to the fuzzy comprehensive judgment matrix and the factor weight set;
The training module is specifically configured to determine a first prediction score value of each cloud platform tenant in the cloud platform tenants according to the plurality of index data in the information of the plurality of cloud platform tenants; training the scoring model based on the plurality of metric data and the first predictive scoring value.
8. A scoring apparatus, comprising:
the cloud platform tenant information evaluation module is used for acquiring information of cloud platform tenants to be scored, wherein the information of the cloud platform tenants comprises a plurality of index data;
The scoring model module is used for inputting the information of the cloud platform tenant to be scored into a scoring model to obtain a second scoring value;
wherein the scoring model is trained based on the training device of the scoring model of claim 7;
The objective scoring module is used for determining objective scoring values according to the plurality of index data; and the comprehensive scoring module is used for determining the comprehensive scoring value of the cloud platform tenant according to the second scoring value and the objective scoring value.
9. An apparatus, the apparatus comprising: a processor and a memory storing computer program instructions;
The processor, when executing the computer program instructions, implements the method of any of claims 1-6.
10. A computer storage medium having stored thereon computer program instructions which, when executed by a processor, implement the method of any of claims 1-6.
CN201911413990.0A 2019-12-31 2019-12-31 Training method, device, equipment and storage medium of scoring model Active CN113128810B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201911413990.0A CN113128810B (en) 2019-12-31 2019-12-31 Training method, device, equipment and storage medium of scoring model

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201911413990.0A CN113128810B (en) 2019-12-31 2019-12-31 Training method, device, equipment and storage medium of scoring model

Publications (2)

Publication Number Publication Date
CN113128810A CN113128810A (en) 2021-07-16
CN113128810B true CN113128810B (en) 2024-05-28

Family

ID=76770366

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201911413990.0A Active CN113128810B (en) 2019-12-31 2019-12-31 Training method, device, equipment and storage medium of scoring model

Country Status (1)

Country Link
CN (1) CN113128810B (en)

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104883369A (en) * 2015-05-29 2015-09-02 天津大学 Cloud configuration safety assessment method
CN105471644A (en) * 2015-11-30 2016-04-06 中电科华云信息技术有限公司 Method for flexile regulation and control of network flow of tenement in cloud environment
CN105574685A (en) * 2016-02-02 2016-05-11 浙江工业大学 Subjective and objective combination-based cloud service evaluation method
CN106131158A (en) * 2016-06-30 2016-11-16 上海天玑科技股份有限公司 Resource scheduling device based on cloud tenant's credit rating under a kind of cloud data center environment
CN107122894A (en) * 2017-04-14 2017-09-01 华中师范大学 The Education Informatization Level appraisal procedure and system of a kind of combination weighting
CN107862455A (en) * 2017-11-02 2018-03-30 国网天津市电力公司 A kind of subjective and objective mixing appraisal procedure of power system cloud platform construction
CN108694486A (en) * 2017-04-07 2018-10-23 深圳市体数科科技有限公司 A kind of driving behavior intelligent Evaluation method and apparatus based on cloud model
CN110378545A (en) * 2018-04-13 2019-10-25 北京三快在线科技有限公司 Trade company's credit estimation method and device, storage medium, electronic equipment

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10026049B2 (en) * 2013-05-09 2018-07-17 Rockwell Automation Technologies, Inc. Risk assessment for industrial systems using big data

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104883369A (en) * 2015-05-29 2015-09-02 天津大学 Cloud configuration safety assessment method
CN105471644A (en) * 2015-11-30 2016-04-06 中电科华云信息技术有限公司 Method for flexile regulation and control of network flow of tenement in cloud environment
CN105574685A (en) * 2016-02-02 2016-05-11 浙江工业大学 Subjective and objective combination-based cloud service evaluation method
CN106131158A (en) * 2016-06-30 2016-11-16 上海天玑科技股份有限公司 Resource scheduling device based on cloud tenant's credit rating under a kind of cloud data center environment
CN108694486A (en) * 2017-04-07 2018-10-23 深圳市体数科科技有限公司 A kind of driving behavior intelligent Evaluation method and apparatus based on cloud model
CN107122894A (en) * 2017-04-14 2017-09-01 华中师范大学 The Education Informatization Level appraisal procedure and system of a kind of combination weighting
CN107862455A (en) * 2017-11-02 2018-03-30 国网天津市电力公司 A kind of subjective and objective mixing appraisal procedure of power system cloud platform construction
CN110378545A (en) * 2018-04-13 2019-10-25 北京三快在线科技有限公司 Trade company's credit estimation method and device, storage medium, electronic equipment

Also Published As

Publication number Publication date
CN113128810A (en) 2021-07-16

Similar Documents

Publication Publication Date Title
CN109587713B (en) Network index prediction method and device based on ARIMA model and storage medium
US10846620B2 (en) Machine learning-based patent quality metric
CN108280542B (en) User portrait model optimization method, medium and equipment
Shepperd Software project economics: a roadmap
US20190251458A1 (en) System and method for particle swarm optimization and quantile regression based rule mining for regression techniques
CN110059894B (en) Equipment state evaluation method, device, system and storage medium
CN112561320A (en) Training method of mechanism risk prediction model, mechanism risk prediction method and device
CN113869768A (en) Method, device, equipment and readable medium for evaluating strength of industrial chain
CN110989043B (en) Air quality index grade probability forecasting method and device and storage medium
CN110310012B (en) Data analysis method, device, equipment and computer readable storage medium
WO2017073446A1 (en) Device, method, and program for calculating explanatory variable value
CN113128810B (en) Training method, device, equipment and storage medium of scoring model
CN112860672A (en) Method and device for determining label weight
JP5167596B2 (en) Data set selection device and experimental design system
US20190205804A1 (en) Information processing device, information processing method and computer readable medium
CN114925895A (en) Maintenance equipment prediction method, terminal and storage medium
CN111160419B (en) Deep learning-based electronic transformer data classification prediction method and device
CN114330720A (en) Knowledge graph construction method and device for cloud computing and storage medium
CN113850483A (en) Enterprise credit risk rating system
Mathur et al. Prognostication of Weather Patterns using Meteorological Data and ML Techniques
CN113807654B (en) Evaluation method of network operation index, electronic equipment and storage medium
CN109905340B (en) Feature optimization function selection method and device and electronic equipment
Trappey et al. Intelligent Product Quality Prediction for Highly Customized Complex Production Adopting Ensemble Learning Model
CN114549032A (en) Stock processing method and device, computer storage medium and electronic equipment
Horvathova et al. USING MULTI-LEVEL DEA MODEL AND VERIFYING ITS RESULTS WITH LOGIT MODEL

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