CN109901979A - Model optimization intelligent evaluation method, server and computer readable storage medium - Google Patents
Model optimization intelligent evaluation method, server and computer readable storage medium Download PDFInfo
- Publication number
- CN109901979A CN109901979A CN201910067476.XA CN201910067476A CN109901979A CN 109901979 A CN109901979 A CN 109901979A CN 201910067476 A CN201910067476 A CN 201910067476A CN 109901979 A CN109901979 A CN 109901979A
- Authority
- CN
- China
- Prior art keywords
- model
- assessed
- judge
- result
- intelligent evaluation
- 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.)
- Pending
Links
Abstract
The present invention relates to a kind of risk evaluation model technologies, disclose a kind of model optimization intelligent evaluation method, this method comprises: the judgment rule of setting model optimization, the judgment rule includes judge index, judgment threshold and measurement period;In the measurement period, the investigation result for receiving the output result of each model push to be assessed and being fed back after being checked to the output result, according to the judge index and the output result and investigation as a result, each model to be assessed of statistics is directed to total calculated result of the judge index in the measurement period;Judge whether the model to be assessed needs to optimize according to statistical result and the corresponding judgment threshold of the judge index.The present invention also provides a kind of server and computer readable storage mediums.Whether model optimization intelligent evaluation method, server and computer readable storage medium provided by the invention, which can need the model runed to optimize, judges automatically.
Description
Technical field
The present invention relates to risk evaluation model technical field more particularly to a kind of model optimization intelligent evaluation methods, service
Device and computer readable storage medium.
Background technique
Model refers to through the methods of statistical analysis, physical and mathematical modeling, all kinds of algorithms and frame application, carries out to Various types of data
Classification, feature refine, mining analysis, and combines expertise etc., the mathematical expression or regular collection of formation, be risk control and
The offers support such as operational decision making.Model operation during, operation conditions monitoring, it is found aiming at the problem that or newly
Exploitation demand needs to carry out model regular or irregular optimization and updates.
Currently, judging whether a model currently needs to optimize, the experience for being mainly based upon developer carries out people
Work judgement, cannot achieve intelligentized automatic assessment.
Summary of the invention
In view of this, the present invention proposes a kind of model optimization intelligent evaluation method, server and computer-readable storage medium
Matter carries out intelligentized automatic assessment to whether model needs to optimize to solve the problems, such as how to realize.
Firstly, to achieve the above object, the present invention proposes that a kind of model optimization intelligent evaluation method, this method include step
It is rapid:
The judgment rule of model optimization is set, and the judgment rule includes judge index, judgment threshold and measurement period;
In the measurement period, receive each model to be assessed push output result and to the output result into
The investigation fed back after row investigation is as a result, according to the judge index and the output result and investigation as a result, statistics is each to be evaluated
Estimate total calculated result that model is directed to the judge index in the measurement period;And
According to statistical result and the corresponding judgment threshold of the judge index judge the model to be assessed whether need
Optimize.
Optionally, this method further comprises the steps of:
Whether the judge index for monitoring the model to be assessed there are unusual fluctuations;
When there are unusual fluctuations, cause of fluctuation is analyzed;
Judge whether the model to be assessed needs to optimize according to cause of fluctuation.
Optionally, the judge index includes that model is efficient and/or situation of having a showdown, wherein the model effective percentage is
Investigation feedback significant figure/model always goes out to count, and the situation of having a showdown is to combine investigation according to preset red, yellow, blue board auditing system
Result is checked in feedback determination.
Optionally, the unusual fluctuations refer to the amplitude that the judge index declines in the calculated result of the measurement period
More than preset threshold.
Optionally, described to judge whether the model to be assessed needs the step of optimizing to include: according to cause of fluctuation
When the cause of fluctuation is that model configures threshold value failure, judge that the model needs to be assessed optimize;
When the cause of fluctuation is that process has control, judge that the model to be assessed does not need to optimize;
When the cause of fluctuation is business change, it is judged as suggestion development model again.
In addition, to achieve the above object, the present invention also provides a kind of server, including memory, processor, the storages
The model optimization intelligent evaluation system that can be run on the processor, the model optimization intelligent evaluation system are stored on device
It realizes when being executed by the processor such as the step of above-mentioned model optimization intelligent evaluation method.
Further, to achieve the above object, the present invention also provides a kind of computer readable storage medium, the computers
Readable storage medium storing program for executing is stored with model optimization intelligent evaluation system, and the model optimization intelligent evaluation system can be by least one
It manages device to execute, so that at least one described processor is executed such as the step of above-mentioned model optimization intelligent evaluation method.
Compared to the prior art, model optimization intelligent evaluation method proposed by the invention, server and computer-readable
Whether storage medium can need to carry out excellent according to preset judge index and threshold value automatically according to statistical result judgment models
Change, and consider data stability, whether monitoring judge index unusual fluctuations occurs, to further be judged according to cause of fluctuation
Whether model, which needs, optimizes, and realizes the intelligent evaluation of model optimization, reduces labor workload, improves accuracy of judgement
Degree, and model can be continued to monitor, discovery needs the model optimized in time, to improve the processing effect of model
Fruit.
Detailed description of the invention
Fig. 1 is the schematic diagram of the optional hardware structure of server one of the present invention;
Fig. 2 is the program module schematic diagram of model optimization intelligent evaluation system first embodiment of the present invention;
Fig. 3 is the program module schematic diagram of model optimization intelligent evaluation system second embodiment of the present invention;
Fig. 4 is the flow diagram of model optimization intelligent evaluation method first embodiment of the present invention;
Fig. 5 is the flow diagram of model optimization intelligent evaluation method second embodiment of the present invention;
The embodiments will be further described with reference to the accompanying drawings for the realization, the function and the advantages of the object of the present invention.
Specific embodiment
In order to make the objectives, technical solutions, and advantages of the present invention clearer, with reference to the accompanying drawings and embodiments, right
The present invention is further elaborated.It should be appreciated that described herein, specific examples are only used to explain the present invention, not
For limiting the present invention.Based on the embodiments of the present invention, those of ordinary skill in the art are not before making creative work
Every other embodiment obtained is put, shall fall within the protection scope of the present invention.
It should be noted that the description for being related to " first ", " second " etc. in the present invention is used for description purposes only, and cannot
It is interpreted as its relative importance of indication or suggestion or implicitly indicates the quantity of indicated technical characteristic.Define as a result, " the
One ", the feature of " second " can explicitly or implicitly include at least one of the features.In addition, the skill between each embodiment
Art scheme can be combined with each other, but must be based on can be realized by those of ordinary skill in the art, when technical solution
Will be understood that the combination of this technical solution is not present in conjunction with there is conflicting or cannot achieve when, also not the present invention claims
Protection scope within.
As shown in fig.1, being the schematic diagram of the optional hardware structure of server 2 one of the present invention.
In the present embodiment, the server 2 may include, but be not limited only to, and can be in communication with each other connection by system bus and deposit
Reservoir 11, processor 12, network interface 13.It should be pointed out that Fig. 1 illustrates only the server 2 with component 11-13, but
Be it should be understood that, it is not required that implement all components shown, the implementation that can be substituted is more or less component.
Wherein, the server 2 can be rack-mount server, blade server, tower server or cabinet-type clothes
Business device etc. calculates equipment, which can be independent server, be also possible to server set composed by multiple servers
Group.
The memory 11 include at least a type of readable storage medium storing program for executing, the readable storage medium storing program for executing include flash memory,
Hard disk, multimedia card, card-type memory (for example, SD or DX memory etc.), random access storage device (RAM), static random are visited
It asks memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), may be programmed read-only deposit
Reservoir (PROM), magnetic storage, disk, CD etc..In some embodiments, the memory 11 can be the server
2 internal storage unit, such as the hard disk or memory of the server 2.In further embodiments, the memory 11 can also be with
It is the plug-in type hard disk being equipped on the External memory equipment of the server 2, such as the server 2, intelligent memory card (Smart
Media Card, SMC), secure digital (Secure Digital, SD) card, flash card (Flash Card) etc..Certainly, described
Memory 11 can also both including the server 2 internal storage unit and also including its External memory equipment.In the present embodiment,
The memory 11 is installed on the operating system and types of applications software of the server 2 commonly used in storage, such as model is excellent
Change the program code etc. of intelligent evaluation system 200.In addition, the memory 11 can be also used for temporarily storing exported or
The Various types of data that person will export.
The processor 12 can be in some embodiments central processing unit (Central Processing Unit,
CPU), controller, microcontroller, microprocessor or other data processing chips.The processor 12 is commonly used in the control clothes
The overall operation of business device 2.In the present embodiment, the processor 12 for run the program code stored in the memory 11 or
Person handles data, such as runs the model optimization intelligent evaluation system 200 etc..
The network interface 13 may include radio network interface or wired network interface, which is commonly used in
Communication connection is established between the server 2 and other electronic equipments.
So far, oneself is through describing the hardware configuration and function of relevant device of the present invention in detail.In the following, above-mentioned introduction will be based on
It is proposed each embodiment of the invention.
Firstly, the present invention proposes a kind of model optimization intelligent evaluation system 200.
As shown in fig.2, being the Program modual graph of 200 first embodiment of model optimization intelligent evaluation system of the present invention.
In the present embodiment, the model optimization intelligent evaluation system 200 includes a series of is stored on memory 11
The mould of various embodiments of the present invention may be implemented when the computer program instructions are executed by processor 12 in computer program instructions
Type Intelligent Optimal evaluation operation.In some embodiments, the specific behaviour realized based on the computer program instructions each section
Make, model optimization intelligent evaluation system 200 can be divided into one or more modules.For example, the model is excellent in Fig. 2
Setup module 201, statistical module 202, judgment module 203 can be divided by changing intelligent evaluation system 200.Wherein:
The setup module 201, for the judgment rule of model optimization to be arranged.
Specifically, the judgment rule includes judge index, judgment threshold and measurement period etc..The judge index includes
Model effective percentage and situation etc. of having a showdown.Wherein, effective percentage=investigation feedback significant figure/model always goes out to count.For example, model A is exported
Abnormal data amount be 50, after artificial or intelligence investigation, the effective anomaly data volume of feedback is 13, then model A's is effective
Rate is 13/50=26%.Situation of having a showdown is to combine investigation feedback is determining to check according to preset red, yellow, blue board auditing system
As a result.Measurement period can be 1 month, 3 months, 1 year etc..For different judge index, different judgement thresholds can be set
Value and measurement period.For example, the efficient measurement period of model is 1 month, three phases that continued are lower than efficient threshold values, are determined as
It needs to optimize.In another example the measurement period for situation of having a showdown is to send out a warning or yellow card is less than 2 in 1 year, 1 year, need
It optimizes.
In addition, being directed to different models, different judgment rules can also be set according to the aspect of model.For example, for height
The model of risk, corresponding judgment rule are more stringent (such as efficient threshold requirement is more high).
The statistical module 202, is counted for periodically treating assessment models according to the judge index.
Specifically, in pre-set measurement period, the output result for receiving each model to be assessed push is (such as different
Regular data amount, corresponding exception main body of abnormal data etc.), and be directed to the output result and carry out after artificial or intelligence is checked instead
The investigation result (such as effective anomaly data volume etc.) of feedback, then according to set judge index and the output result and row
It looks into as a result, counting total calculated result that each model to be assessed is directed to the judge index in the measurement period.For example, according to mould
Total effective anomaly data volume of abnormal data amount that type A is exported in total in 1 month and investigation feedback, statistical model A this 1
Total effective percentage in a month.
The judgment module 203, for should be to according to statistical result and the judgement of the judge index corresponding judgment threshold
Whether assessment models, which need, optimizes.
Specifically, when count model to be assessed in the measurement period be directed to the judge index total calculated result after,
Judge whether the model to be assessed needs to optimize according to corresponding judgment threshold.For example, model A 1st month, the 2nd month, the 3rd
The total effective rate counted for a month is respectively 26%, 20%, 32%, and efficient threshold value is 30%, then model A 1st month and
2nd month effective percentage is lower than threshold value, but 3rd month is higher than threshold value, wouldn't need to optimize.In another example Model B the 1st
The moon, the 2nd month, the total effective rate counted for the 3rd month are respectively 25%, 19%, 22%, and efficient threshold value is 30%, then should
1-3 months effective percentage of Model B are below threshold value, reach condition of lasting three phase lower than efficient threshold values, need to carry out excellent
Change.
Model optimization intelligent evaluation system provided in this embodiment, can be according to model operation data effective percentage and default week
Whether the situation judgment models of having a showdown in the phase need to optimize, and determine to need if efficient or quantity of having a showdown is lower than preset threshold
Optimization.The intelligent evaluation for realizing model optimization, reduces labor workload, improves accuracy of judgement degree, and can be to mould
Type is continued to monitor, and discovery needs the model optimized in time, to improve the treatment effect of model.
As shown in fig.3, being the Program modual graph of 200 second embodiment of model optimization intelligent evaluation system of the present invention.This
In embodiment, the model optimization intelligent evaluation system 200 in addition to include first embodiment in the setup module 201,
It further include monitoring modular 204, analysis module 205 except statistical module 202, judgment module 203.
Whether the judge index that the monitoring modular 204 is used to monitor the model to be assessed there are unusual fluctuations.
Specifically, each model to be assessed is being counted in corresponding measurement period for total calculated result of the judge index
When, it is also necessary to the interpretation phase, whether the result there are unusual fluctuations.In the present solution, the unusual fluctuations refer to the judge index
It is more than preset threshold in the amplitude that the calculated result of the measurement period declines.For example, what is counted within model A 1st month is total effective
Rate is 65%, and the 2nd month total effective rate counted is 41%, has dropped 24%, has been more than preset threshold (20%), therefore sentence
There are unusual fluctuations in the effective percentage of disconnected model A, needs to remind and inspect, and judges whether to need to optimize.
The analysis module 205 is used for when there are unusual fluctuations, analyzes cause of fluctuation.
Specifically, when unusual fluctuations occurs in the judge index for monitoring the model to be assessed, by predetermined manner
Warning is issued, further judges whether the model to be assessed needs to optimize.There is the extraordinary wave firstly, it is necessary to analyze
Dynamic reason.The cause of fluctuation includes: 1) model configuration threshold value failure;2) business change;3) process has control, etc..
Fluctuation Cause Analysis can combine artificial investigation interview to carry out by intellectualized algorithm (such as automatic threshold optimization algorithm).
The judgment module 203 is also used to judge whether the model to be assessed needs to optimize according to cause of fluctuation.
Specifically, when the cause of fluctuation is that model configures threshold value failure, judge that the model to be assessed needs to carry out
Optimization;When the cause of fluctuation is that process has control, judge that the model to be assessed does not need to optimize;When the wave
When dynamic reason is business change, it is judged as suggestion development model again.
Model optimization intelligent evaluation system provided in this embodiment, can be according to preset judge index and the automatic root of threshold value
Whether result judgment models need to optimize according to statistics, and consider data stability, and it is different whether monitoring judge index occurs
Ordinary wave is dynamic, and whether to need to optimize according to the further judgment models of cause of fluctuation, the intelligence for realizing model optimization is commented
Estimate, reduce labor workload, improve accuracy of judgement degree, and can continue to monitor to model, discovery needs in time
The model optimized, to improve the treatment effect of model.
In addition, the present invention also proposes a kind of model optimization intelligent evaluation method.
As shown in fig.4, being the flow diagram of model optimization intelligent evaluation method first embodiment of the present invention.In this reality
It applies in example, the execution sequence of the step in flow chart shown in Fig. 4 can change according to different requirements, and certain steps can be with
It omits.This method comprises:
The judgment rule of model optimization is arranged in step S400.
Specifically, the judgment rule includes judge index, judgment threshold and measurement period etc..The judge index includes
Model effective percentage and situation etc. of having a showdown.Wherein, effective percentage=investigation feedback significant figure/model always goes out to count.For example, model A is exported
Abnormal data amount be 50, after artificial or intelligence investigation, the effective anomaly data volume of feedback is 13, then model A's is effective
Rate is 13/50=26%.Situation of having a showdown is to combine investigation feedback is determining to check according to preset red, yellow, blue board auditing system
As a result.Measurement period can be 1 month, 3 months, 1 year etc..For different judge index, different judgement thresholds can be set
Value and measurement period.For example, the efficient measurement period of model is 1 month, three phases that continued are lower than efficient threshold values, are determined as
It needs to optimize.In another example the measurement period for situation of having a showdown is to send out a warning or yellow card is less than 2 in 1 year, 1 year, need
It optimizes.
In addition, being directed to different models, different judgment rules can also be set according to the aspect of model.For example, for height
The model of risk, corresponding judgment rule are more stringent (such as efficient threshold requirement is more high).
Step S402 periodically treats assessment models according to the judge index and is counted.
Specifically, in pre-set measurement period, the output result for receiving each model to be assessed push is (such as different
Regular data amount, corresponding exception main body of abnormal data etc.), and be directed to the output result and carry out after artificial or intelligence is checked instead
The investigation result (such as effective anomaly data volume etc.) of feedback, then according to set judge index and the output result and row
It looks into as a result, counting total calculated result that each model to be assessed is directed to the judge index in the measurement period.For example, according to mould
Total effective anomaly data volume of abnormal data amount that type A is exported in total in 1 month and investigation feedback, statistical model A this 1
Total effective percentage in a month.
Whether step S404 judges the model to be assessed according to statistical result and the corresponding judgment threshold of the judge index
It needs to optimize.
Specifically, when count model to be assessed in the measurement period be directed to the judge index total calculated result after,
Judge whether the model to be assessed needs to optimize according to corresponding judgment threshold.For example, model A 1st month, the 2nd month, the 3rd
The total effective rate counted for a month is respectively 26%, 20%, 32%, and efficient threshold value is 30%, then model A 1st month and
2nd month effective percentage is lower than threshold value, but 3rd month is higher than threshold value, wouldn't need to optimize.In another example Model B the 1st
The moon, the 2nd month, the total effective rate counted for the 3rd month are respectively 25%, 19%, 22%, and efficient threshold value is 30%, then should
1-3 months effective percentage of Model B are below threshold value, reach condition of lasting three phase lower than efficient threshold values, need to carry out excellent
Change.
Model optimization intelligent evaluation method provided in this embodiment, can be according to model operation data effective percentage and default week
Whether the situation judgment models of having a showdown in the phase need to optimize, and determine to need if efficient or quantity of having a showdown is lower than preset threshold
Optimization.The intelligent evaluation for realizing model optimization, reduces labor workload, improves accuracy of judgement degree, and can be to mould
Type is continued to monitor, and discovery needs the model optimized in time, to improve the treatment effect of model.
As shown in figure 5, being the flow diagram of the second embodiment of model optimization intelligent evaluation method of the present invention.This implementation
In example, the step S500-S504 of the model optimization intelligent evaluation method and the step S400-S404 of first embodiment are similar
Seemingly, difference is that this method further includes step S506-S510.
Method includes the following steps:
The judgment rule of model optimization is arranged in step S500.
Specifically, the judgment rule includes judge index, judgment threshold and measurement period etc..The judge index includes
Model effective percentage and situation etc. of having a showdown.Wherein, effective percentage=investigation feedback significant figure/model always goes out to count.For example, model A is exported
Abnormal data amount be 50, after artificial or intelligence investigation, the effective anomaly data volume of feedback is 13, then model A's is effective
Rate is 13/50=26%.Situation of having a showdown is to combine investigation feedback is determining to check according to preset red, yellow, blue board auditing system
As a result.Measurement period can be 1 month, 3 months, 1 year etc..For different judge index, different judgement thresholds can be set
Value and measurement period.For example, the efficient measurement period of model is 1 month, three phases that continued are lower than efficient threshold values, are determined as
It needs to optimize.In another example the measurement period for situation of having a showdown is to send out a warning or yellow card is less than 2 in 1 year, 1 year, need
It optimizes.
In addition, being directed to different models, different judgment rules can also be set according to the aspect of model.For example, for height
The model of risk, corresponding judgment rule are more stringent (such as efficient threshold requirement is more high).
Step S502 periodically treats assessment models according to the judge index and is counted.
Specifically, in pre-set measurement period, the output result for receiving each model to be assessed push is (such as different
Regular data amount, corresponding exception main body of abnormal data etc.), and be directed to the output result and carry out after artificial or intelligence is checked instead
The investigation result (such as effective anomaly data volume etc.) of feedback, then according to set judge index and the output result and row
It looks into as a result, counting total calculated result that each model to be assessed is directed to the judge index in the measurement period.For example, according to mould
Total effective anomaly data volume of abnormal data amount that type A is exported in total in 1 month and investigation feedback, statistical model A this 1
Total effective percentage in a month.
Whether step S504 judges the model to be assessed according to statistical result and the corresponding judgment threshold of the judge index
It needs to optimize.
Specifically, when count model to be assessed in the measurement period be directed to the judge index total calculated result after,
Judge whether the model to be assessed needs to optimize according to corresponding judgment threshold.For example, model A 1st month, the 2nd month, the 3rd
The total effective rate counted for a month is respectively 26%, 20%, 32%, and efficient threshold value is 30%, then model A 1st month and
2nd month effective percentage is lower than threshold value, but 3rd month is higher than threshold value, wouldn't need to optimize.In another example Model B the 1st
The moon, the 2nd month, the total effective rate counted for the 3rd month are respectively 25%, 19%, 22%, and efficient threshold value is 30%, then should
1-3 months effective percentage of Model B are below threshold value, reach condition of lasting three phase lower than efficient threshold values, need to carry out excellent
Change.
Whether step S506, the judge index for monitoring the model to be assessed there are unusual fluctuations.
Specifically, each model to be assessed is being counted in corresponding measurement period for total calculated result of the judge index
When, it is also necessary to the interpretation phase, whether the result there are unusual fluctuations.In the present solution, the unusual fluctuations refer to the judge index
It is more than preset threshold in the amplitude that the calculated result of the measurement period declines.For example, what is counted within model A 1st month is total effective
Rate is 65%, and the 2nd month total effective rate counted is 41%, has dropped 24%, has been more than preset threshold (20%), therefore sentence
There are unusual fluctuations in the effective percentage of disconnected model A, needs to remind and inspect, and judges whether to need to optimize.
Step S508 analyzes cause of fluctuation when there are unusual fluctuations.
Specifically, when unusual fluctuations occurs in the judge index for monitoring the model to be assessed, by predetermined manner
Warning is issued, further judges whether the model to be assessed needs to optimize.There is the extraordinary wave firstly, it is necessary to analyze
Dynamic reason.The cause of fluctuation includes: 1) model configuration threshold value failure;2) business change;3) process has control, etc..
Fluctuation Cause Analysis can combine artificial investigation interview to carry out by intellectualized algorithm (such as automatic threshold optimization algorithm).
Step S510 judges whether the model to be assessed needs to optimize according to cause of fluctuation.
Specifically, when the cause of fluctuation is that model configures threshold value failure, judge that the model to be assessed needs to carry out
Optimization;When the cause of fluctuation is that process has control, judge that the model to be assessed does not need to optimize;When the wave
When dynamic reason is business change, it is judged as suggestion development model again.
Model optimization intelligent evaluation method provided in this embodiment, can be according to preset judge index and the automatic root of threshold value
Whether result judgment models need to optimize according to statistics, and consider data stability, and it is different whether monitoring judge index occurs
Ordinary wave is dynamic, and whether to need to optimize according to the further judgment models of cause of fluctuation, the intelligence for realizing model optimization is commented
Estimate, reduce labor workload, improve accuracy of judgement degree, and can continue to monitor to model, discovery needs in time
The model optimized, to improve the treatment effect of model.
The present invention also provides another embodiments, that is, provide a kind of computer readable storage medium, the computer
Readable storage medium storing program for executing is stored with model optimization intelligent evaluation program, and the model optimization intelligent evaluation program can be by least one
It manages device to execute, so that at least one described processor is executed such as the step of above-mentioned model optimization intelligent evaluation method.
In the above-described embodiments, can come wholly or partly by software, hardware, firmware or any combination thereof real
It is existing.When implemented in software, it can entirely or partly realize in the form of a computer program product.
The computer program product includes one or more computer instructions.Load and execute on computers the meter
When calculation machine program instruction, entirely or partly generate according to process or function described in the embodiment of the present invention.The computer can
To be general purpose computer, special purpose computer, computer network or other programmable devices.The computer instruction can be deposited
Storage in a computer-readable storage medium, or from a computer readable storage medium to another computer readable storage medium
Transmission, for example, the computer instruction can pass through wired (example from a web-site, computer, server or data center
Such as coaxial cable, optical fiber, Digital Subscriber Line (DSL)) or wireless (such as infrared, wireless, microwave) mode to another website
Website, computer, server or data center are transmitted.The computer readable storage medium can be computer and can deposit
Any usable medium of storage either includes that the data storages such as one or more usable mediums integrated server, data center are set
It is standby.The usable medium can be magnetic medium, (for example, floppy disk, hard disk, tape), optical medium (for example, DVD) or partly lead
Body medium (such as solid state hard disk Solid State Disk (SSD)) etc.
It is apparent to those skilled in the art that for convenience and simplicity of description, the system of foregoing description,
The specific work process of device and unit, can refer to corresponding processes in the foregoing method embodiment, and details are not described herein.
In several embodiments provided herein, it should be understood that disclosed system, device and method can be with
It realizes by another way.For example, the apparatus embodiments described above are merely exemplary, for example, the unit/mould
The division of block, only a kind of logical function partition, there may be another division manner in actual implementation, for example, multiple units or
Component can be combined or can be integrated into another system, or some features can be ignored or not executed.Another point is shown
The mutual coupling, direct-coupling or communication connection shown or discussed can be through some interfaces, between device or unit
Coupling or communication connection are connect, can be electrical property, mechanical or other forms.
The unit as illustrated by the separation member may or may not be physically separated, aobvious as unit
The component shown may or may not be physical unit, it can and it is in one place, or may be distributed over multiple
In network unit.It can select some or all of unit therein according to the actual needs to realize the mesh of this embodiment scheme
's.
It, can also be in addition, each functional unit in each embodiment of the application can integrate in one processing unit
It is that each unit physically exists alone, can also be integrated in one unit with two or more units.Above-mentioned integrated list
Member both can take the form of hardware realization, can also realize in the form of software functional units.
If the integrated unit is realized in the form of SFU software functional unit and sells or use as independent product
When, it can store in a computer readable storage medium.Based on this understanding, the technical solution of the application is substantially
The all or part of the part that contributes to existing technology or the technical solution can be in the form of software products in other words
It embodies, which is stored in a storage medium, including some instructions are used so that a computer
Equipment (can be personal computer, server or the network equipment etc.) executes the complete of each embodiment the method for the application
Portion or part steps.And storage medium above-mentioned includes: USB flash disk, mobile hard disk, read-only memory (ROM, Read-Only
Memory), random access memory (RAM, Random Access Memory), magnetic or disk etc. are various can store journey
The medium of sequence code.
It should be noted that the serial number of the above embodiments of the invention is only for description, do not represent the advantages or disadvantages of the embodiments.And
The terms "include", "comprise" herein or any other variant thereof is intended to cover non-exclusive inclusion, so that packet
Process, device, article or the method for including a series of elements not only include those elements, but also including being not explicitly listed
Other element, or further include for this process, device, article or the intrinsic element of method.Do not limiting more
In the case where, the element that is limited by sentence "including a ...", it is not excluded that including process, device, the article of the element
Or there is also other identical elements in method.
The above is only a preferred embodiment of the present invention, is not intended to limit the scope of the invention, all to utilize this hair
Equivalent structure or equivalent flow shift made by bright specification and accompanying drawing content is applied directly or indirectly in other relevant skills
Art field, is included within the scope of the present invention.
Claims (10)
1. a kind of model optimization intelligent evaluation method, which is characterized in that the method includes the steps:
The judgment rule of model optimization is set, and the judgment rule includes judge index, judgment threshold and measurement period;
In the measurement period, receives the output result of each model push to be assessed and the output result is arranged
The investigation fed back after looking into is as a result, according to the judge index and the output result and investigation as a result, each mould to be assessed of statistics
Type is directed to total calculated result of the judge index in the measurement period;And
It is excellent to judge whether the model to be assessed needs according to statistical result and the corresponding judgment threshold of the judge index
Change.
2. model optimization intelligent evaluation method as described in claim 1, which is characterized in that this method further comprises the steps of:
Whether the judge index for monitoring the model to be assessed there are unusual fluctuations;
When there are unusual fluctuations, cause of fluctuation is analyzed;
Judge whether the model to be assessed needs to optimize according to cause of fluctuation.
3. model optimization intelligent evaluation method as claimed in claim 1 or 2, which is characterized in that the judge index includes mould
Type effective percentage and/or situation of having a showdown, wherein the model effective percentage is that investigation feedback significant figure/model always goes out number, described to have a showdown
Situation is to check result according to preset red, yellow, blue board auditing system combination investigation feedback determination.
4. model optimization intelligent evaluation method as claimed in claim 2, which is characterized in that the unusual fluctuations, which refer to, described to be sentenced
The amplitude that severed finger is marked on the calculated result decline of the measurement period is more than preset threshold.
5. model optimization intelligent evaluation method as claimed in claim 2, which is characterized in that described to judge institute according to cause of fluctuation
Whether state model to be assessed needs the step of optimizing to include:
When the cause of fluctuation is that model configures threshold value failure, judge that the model needs to be assessed optimize;
When the cause of fluctuation is that process has control, judge that the model to be assessed does not need to optimize;
When the cause of fluctuation is business change, it is judged as suggestion development model again.
6. a kind of server, which is characterized in that the server includes memory, processor, and being stored on the memory can
The model optimization intelligent evaluation system run on the processor, the model optimization intelligent evaluation system is by the processor
Following steps are realized when execution:
The judgment rule of model optimization is set, and the judgment rule includes judge index, judgment threshold and measurement period;
In the measurement period, receives the output result of each model push to be assessed and the output result is arranged
The investigation fed back after looking into is as a result, according to the judge index and the output result and investigation as a result, each mould to be assessed of statistics
Type is directed to total calculated result of the judge index in the measurement period;And
It is excellent to judge whether the model to be assessed needs according to statistical result and the corresponding judgment threshold of the judge index
Change.
7. server as claimed in claim 6, which is characterized in that the model optimization intelligent evaluation system is by the processor
Step is also realized when execution:
Whether the judge index for monitoring the model to be assessed there are unusual fluctuations;
When there are unusual fluctuations, cause of fluctuation is analyzed;
Judge whether the model to be assessed needs to optimize according to cause of fluctuation.
8. server as claimed in claims 6 or 7, which is characterized in that the judge index includes that model is efficient and/or bright
The cards one holds condition, wherein the model effective percentage is that investigation feedback significant figure/model always goes out to count, and the situation of having a showdown is according to default
Red, yellow, blue board auditing system combine investigation feedback it is determining check result.
9. server as claimed in claim 7, which is characterized in that described to judge that the model to be assessed is according to cause of fluctuation
It is no to need the step of optimizing to include:
When the cause of fluctuation is that model configures threshold value failure, judge that the model needs to be assessed optimize;
When the cause of fluctuation is that process has control, judge that the model to be assessed does not need to optimize;
When the cause of fluctuation is business change, it is judged as suggestion development model again.
10. a kind of computer readable storage medium, which is characterized in that the computer-readable recording medium storage has model optimization
Intelligent evaluation system, the model optimization intelligent evaluation system can be executed by least one processor so that it is described at least one
Processor executes the step of model optimization intelligent evaluation method according to any one of claims 1 to 5.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910067476.XA CN109901979A (en) | 2019-01-24 | 2019-01-24 | Model optimization intelligent evaluation method, server and computer readable storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910067476.XA CN109901979A (en) | 2019-01-24 | 2019-01-24 | Model optimization intelligent evaluation method, server and computer readable storage medium |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109901979A true CN109901979A (en) | 2019-06-18 |
Family
ID=66944143
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910067476.XA Pending CN109901979A (en) | 2019-01-24 | 2019-01-24 | Model optimization intelligent evaluation method, server and computer readable storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109901979A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110866328A (en) * | 2019-10-12 | 2020-03-06 | 中国平安财产保险股份有限公司 | Data model replacing method and device, computer equipment and storage medium |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101443732A (en) * | 2006-05-12 | 2009-05-27 | 皇家飞利浦电子股份有限公司 | Method for changing over from a first adaptive data processing version to a second adaptive data processing version |
CN107783896A (en) * | 2017-02-16 | 2018-03-09 | 平安科技(深圳)有限公司 | The optimization method and device of a kind of data processing model |
CN107958268A (en) * | 2017-11-22 | 2018-04-24 | 用友金融信息技术股份有限公司 | The training method and device of a kind of data model |
CN109242135A (en) * | 2018-07-16 | 2019-01-18 | 阿里巴巴集团控股有限公司 | A kind of model method for running, device and service server |
-
2019
- 2019-01-24 CN CN201910067476.XA patent/CN109901979A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101443732A (en) * | 2006-05-12 | 2009-05-27 | 皇家飞利浦电子股份有限公司 | Method for changing over from a first adaptive data processing version to a second adaptive data processing version |
CN107783896A (en) * | 2017-02-16 | 2018-03-09 | 平安科技(深圳)有限公司 | The optimization method and device of a kind of data processing model |
CN107958268A (en) * | 2017-11-22 | 2018-04-24 | 用友金融信息技术股份有限公司 | The training method and device of a kind of data model |
CN109242135A (en) * | 2018-07-16 | 2019-01-18 | 阿里巴巴集团控股有限公司 | A kind of model method for running, device and service server |
Non-Patent Citations (1)
Title |
---|
中国平安保险(集团)股份有限公司: "《中国监察学会金融分会2013年度理论研讨优秀论文集》", 31 December 2013 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110866328A (en) * | 2019-10-12 | 2020-03-06 | 中国平安财产保险股份有限公司 | Data model replacing method and device, computer equipment and storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106201722A (en) | The load regulation method of server and system | |
CN109857592A (en) | Data recovery control method, server and storage medium | |
CN110955586A (en) | System fault prediction method, device and equipment based on log | |
CN110198305A (en) | It attends a banquet method for detecting abnormality, system, computer equipment and the storage medium of IP | |
CN109670690A (en) | Data information center monitoring and early warning method, system and equipment | |
CN108427669A (en) | Abnormal behaviour monitoring method and system | |
CN110138688A (en) | Dynamic adjusts method, apparatus, equipment and the readable storage medium storing program for executing of business interface | |
CN105379204A (en) | Methods and systems for selecting resources for data routing | |
US20160364467A1 (en) | Event notification system with cluster classification | |
CN110502395A (en) | Equipment running status appraisal procedure, terminal device and storage medium based on cluster | |
CN109409764A (en) | Production monitoring method and terminal device | |
CN115543577A (en) | Kubernetes resource scheduling optimization method based on covariates, storage medium and equipment | |
CN104077128B (en) | A kind of data processing method and device | |
CN109376981B (en) | Data processing mode determining method and device, server and data processing method | |
CN109901979A (en) | Model optimization intelligent evaluation method, server and computer readable storage medium | |
CN109886809A (en) | Transaction data intelligent analysis method, electronic device and computer readable storage medium | |
EP3391589B1 (en) | Autonomic method for managing a computing system | |
CN110533268A (en) | A kind of risk control method, air control system and terminal device | |
EP2797260A2 (en) | Risk mitigation in data center networks | |
CN106469105A (en) | A kind of monitoring rules screening technique and device | |
CN116760726A (en) | Supplier cheating detection method based on coding and decoding algorithm | |
CN112363826B (en) | Project resource comprehensive management system, method, terminal and storage medium | |
CN109783310A (en) | The Dynamic and Multi dimensional method for safety monitoring and its monitoring device of information technoloy equipment | |
CN111651652B (en) | Emotion tendency identification method, device, equipment and medium based on artificial intelligence | |
CN106874201A (en) | Method of testing 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 | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190618 |