CN112861955A - Risk model strategy generation system and method - Google Patents
Risk model strategy generation system and method Download PDFInfo
- Publication number
- CN112861955A CN112861955A CN202110135369.3A CN202110135369A CN112861955A CN 112861955 A CN112861955 A CN 112861955A CN 202110135369 A CN202110135369 A CN 202110135369A CN 112861955 A CN112861955 A CN 112861955A
- Authority
- CN
- China
- Prior art keywords
- model
- variable
- strategy
- stability
- fluctuation
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/211—Selection of the most significant subset of features
- G06F18/2113—Selection of the most significant subset of features by ranking or filtering the set of features, e.g. using a measure of variance or of feature cross-correlation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- 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
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/08—Insurance
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Business, Economics & Management (AREA)
- General Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Accounting & Taxation (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Finance (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- Technology Law (AREA)
- General Business, Economics & Management (AREA)
- Feedback Control In General (AREA)
Abstract
The invention discloses a risk model strategy generating system and a method, wherein the risk model strategy generating system comprises the following steps: the system comprises a variable stability checker, a variable stability intervener and a multi-model fusion device; the variable stability checker is used for screening stability of bottom variable and automatically evaluating the stability condition of the variable; the fluctuation of the variable selected by the model does not exceed a set threshold value; the variable stability intervener is used for applying the output result of the variable stability checker to the process of model establishment; the multi-model fusion device is used for selecting a plurality of models to be fused, each model corresponds to one use scene, and different models are used for different scenes. The risk model strategy generation system and method provided by the invention can automatically update the risk model strategy and improve the accuracy of risk control.
Description
Technical Field
The invention belongs to the technical field of risk control, relates to a risk control system, and particularly relates to a risk model strategy generation system and method.
Background
Risk control is a key of finance, and with the development of the times, the general trend in the field of wind control is that the informatization, modeling and intellectualization degrees are higher and higher.
The auditing mode process of the financial risk control is as follows: firstly, manual paper quality audit is developed to manual on-line audit, then the paper quality audit is developed to a part of machine audit, the rest part of machine audit is subjected to manual audit, and then the paper quality audit is developed to full-automatic audit; however, the fully automatic auditing also begins to generate a bottleneck in actual business, namely the iteration speed is not fast enough. The reason for this is that the construction of risk models and strategies still requires calculation and evaluation by human, and this work requires a lot of calculations, empirical judgment and decision making, and is difficult to be simply implemented by machine learning.
In view of the above, there is a need to design a new risk control model to overcome at least some of the above-mentioned disadvantages of the existing risk control models.
Disclosure of Invention
The invention provides a risk model strategy generation system and method, which can automatically update a risk model strategy and improve the accuracy of risk control.
In order to solve the technical problem, according to one aspect of the present invention, the following technical solutions are adopted:
a risk model policy generation system, the risk model policy generation system comprising:
the variable stability checker is used for screening the stability of the bottom variable and automatically evaluating the stability condition of the variable; the fluctuation generated by the variable selected by the model does not exceed a set threshold value; if the fluctuation generated by the selected variable exceeds a set threshold value, deleting the corresponding variable;
the variable stability intervener is used for applying the output result of the variable stability checker to the process of establishing the model;
the multi-model fusion device is used for selecting a plurality of models to fuse, each model corresponds to one use scene, and different models are used for different scenes;
the model strategy generator can receive a feedback report of the model strategy challenger and regenerate a new model strategy according to the received feedback report;
if a variable with fluctuation quantity larger than a set threshold value is encountered, the variable is subjected to elimination or data cleaning treatment, and the influence of the fluctuation on the model strategy is eliminated or reduced;
the model strategy generator operates at least once and forms a model rule pool by combining approved models and rules; according to a strategy target, the fusion in the model rule pool firstly makes a strategy through a single model and then combines the strategies, or sets a main model to make a large strategy, and then makes a strategy of specific details through a detailed model and a rule;
the specific intervention method of the variable stability intervener comprises at least one of the following modes: the weight of the influencing variable in the model, the probability of the influencing variable entering the model, the sequence of the influencing variable entering the model, and the cleaning and processing method of the influencing variable in the model.
According to another aspect of the invention, the following technical scheme is adopted: a risk model policy generation system, the risk model policy generation system comprising:
the variable stability checker is used for screening the stability of the bottom variable and automatically evaluating the stability condition of the variable; the fluctuation generated by the variable selected by the model does not exceed a set threshold value; if the fluctuation generated by the selected variable exceeds a set threshold value, deleting the corresponding variable;
the variable stability intervener is used for applying the output result of the variable stability checker to the process of establishing the model;
the multi-model fusion device is used for selecting a plurality of models to be fused, each model corresponds to one use scene, and different models are used for different scenes.
As an embodiment of the present invention, the model policy generator can receive a feedback report from the model policy challenger, and regenerate a new model policy according to the received feedback report;
if the variable with the fluctuation amount larger than the set threshold value is encountered, the variable is subjected to elimination or data cleaning treatment, and the influence of the fluctuation on the model strategy is eliminated or reduced.
As an embodiment of the present invention, the model policy generator operates at least once, combining approved models and rules to form a model rule pool; the fusion in the model rule pool firstly makes a strategy through a single model and then combines the strategies according to a strategy target, or sets a main model to make a large strategy, and then makes a strategy of specific details through a detailed model and a rule.
As an embodiment of the present invention, the specific intervention method of the variable stability intervener includes at least one of the following manners: the weight of the influencing variable in the model, the probability of the influencing variable entering the model, the sequence of the influencing variable entering the model, and the cleaning and processing method of the influencing variable in the model.
According to another aspect of the invention, the following technical scheme is adopted: a risk model policy generation method, comprising:
the variable stability checker screens the stability of the bottom variable and automatically evaluates the stability condition of the variable; the fluctuation of the variable selected by the model does not exceed a set threshold value; if the fluctuation generated by the selected variable exceeds a set threshold value, deleting the corresponding variable;
the variable stability intervener applies the output result of the variable stability checker to the process of model establishment;
the multi-model fusion device selects a plurality of models for fusion, each model corresponds to one use scene, and different models are used for different scenes;
the specific intervention method of the variable stability intervener comprises at least one of the following modes: the weight of the influence variable in the model, the probability of the influence variable entering the model, the sequence of the influence variable entering the model, and the cleaning and processing method of the influence variable in the model;
the model strategy generator can receive a feedback report of the model strategy challenger and regenerate a new model strategy according to the received feedback report;
if a variable with fluctuation quantity larger than a set threshold value is encountered, the variable is subjected to elimination or data cleaning treatment, and the influence of the fluctuation on the model strategy is eliminated or reduced;
the model strategy generator operates at least once and forms a model rule pool by combining approved models and rules; the fusion in the model rule pool firstly makes a strategy through a single model and then combines the strategies according to a strategy target, or sets a main model to make a large strategy, and then makes a strategy of specific details through a detailed model and a rule.
According to another aspect of the invention, the following technical scheme is adopted: a risk model policy generation method, comprising:
the variable stability checker screens the stability of the bottom variable and automatically evaluates the stability condition of the variable; the fluctuation of the variable selected by the model does not exceed a set threshold value; if the fluctuation generated by the selected variable exceeds a set threshold value, deleting the corresponding variable;
the variable stability intervener applies the output result of the variable stability checker to the process of model establishment;
the multi-model fusion device selects a plurality of models for fusion, each model corresponds to one use scene, and different models are used for different scenes.
As an embodiment of the present invention, the specific intervention method of the variable stability intervener includes at least one of the following manners: the weight of the influencing variable in the model, the probability of the influencing variable entering the model, the sequence of the influencing variable entering the model, and the cleaning and processing method of the influencing variable in the model.
As an embodiment of the present invention, the model policy generator can receive a feedback report from the model policy challenger, and regenerate a new model policy according to the received feedback report;
if the variable with the fluctuation amount larger than the set threshold value is encountered, the variable is subjected to elimination or data cleaning treatment, and the influence of the fluctuation on the model strategy is eliminated or reduced.
As an embodiment of the present invention, the model policy generator operates at least once, combining approved models and rules to form a model rule pool; the fusion in the model rule pool firstly makes a strategy through a single model and then combines the strategies according to a strategy target, or sets a main model to make a large strategy, and then makes a strategy of specific details through a detailed model and a rule.
The invention has the beneficial effects that: the risk model strategy generation system and method provided by the invention can automatically update the risk model strategy and improve the accuracy of risk control.
Drawings
Fig. 1 is a schematic composition diagram of a risk model policy generation system according to an embodiment of the present invention.
Fig. 2 is a flowchart of a risk model policy generation method according to an embodiment of the present invention.
Fig. 3 is a schematic diagram of a risk control system according to an embodiment of the present invention.
Fig. 4 is a flowchart of a risk control method according to an embodiment of the invention.
Detailed Description
Preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings.
For a further understanding of the invention, reference will now be made to the preferred embodiments of the invention by way of example, and it is to be understood that the description is intended to further illustrate features and advantages of the invention, and not to limit the scope of the claims.
The description in this section is for several exemplary embodiments only, and the present invention is not limited only to the scope of the embodiments described. It is within the scope of the present disclosure and protection that the same or similar prior art means and some features of the embodiments may be interchanged.
The steps in the embodiments in the specification are only expressed for convenience of description, and the implementation manner of the present application is not limited by the order of implementation of the steps. The term "connected" in the specification includes both direct connection and indirect connection.
The invention discloses a risk model strategy generation system, and fig. 1 is a schematic composition diagram of the risk model strategy generation system in an embodiment of the invention; referring to fig. 1, the risk model policy generation system includes: a variable stability checker 11, a variable stability intervener 12 and a multi-model fusion 13.
The variable stability checker 11 is used for screening stability of bottom-layer variables and automatically evaluating stability conditions of the variables; the variables selected by the model produce fluctuations that do not exceed a set threshold. If the fluctuation generated by the selected variable exceeds a set threshold, the corresponding variable is deleted.
The variable stability intervener 12 is used to apply the output result of the variable stability checker 11 to the process of model building. In one embodiment, the specific intervention method of the variable stability intervener 12 includes at least one of the following ways: the weight of the influencing variable in the model, the probability of the influencing variable entering the model, the sequence of the influencing variable entering the model, and the cleaning and processing method of the influencing variable in the model.
The multi-model fusion device 13 is used for selecting a plurality of models for fusion, each model corresponds to one usage scenario, and different models are used for different scenarios.
In an embodiment of the present invention, the model policy generator can receive a feedback report from the model policy challenger, and regenerate a new model policy according to the received feedback report. If the variable with the fluctuation amount larger than the set threshold value is encountered, the variable is subjected to elimination or data cleaning treatment, and the influence of the fluctuation on the model strategy is eliminated or reduced.
In one embodiment, the model policy generator operates at least once, combining approved models and rules to form a model rule pool; the fusion in the model rule pool firstly makes a strategy through a single model and then combines the strategies according to a strategy target, or sets a main model to make a large strategy, and then makes a strategy of specific details through a detailed model and a rule.
The risk model strategy generation system can be used in a risk control system and used as a model strategy generator of the risk control system; FIG. 3 is a schematic diagram of a risk control system according to an embodiment of the present invention; referring to fig. 3, the risk control system includes: a model policy generator 10, a model policy tracker 20, and a model policy challenger 30.
The model strategy generator 10 is used for selecting sample data in the database, selecting and setting characteristic data, selecting variables, and generating models and strategies.
The model strategy tracker 20 is used for monitoring and tracking various indexes of the model and the strategy generated by the model strategy generator. In an embodiment, the indicators comprise an effect indicator comprising models KS, AUC, IV or/and a stability indicator comprising PSI; the model strategy tracker is used for dynamically tracking the index, and if the set index reaches a set threshold value, the model strategy challenger is triggered to work.
The model strategy challenger 30 is used for dynamically discovering the deficiencies of the model and the strategy and providing challenges. In one embodiment, the model strategy challenger carries out one-time all-round challenge on the model independently, finds weak variables and strong variables according to the variable change conditions, and generates a standard report.
In an embodiment of the present invention, the output format of the model policy challenger 30 is standardized, and such a standardized file is fed back to the model policy generator, so as to achieve the effect of the model policy challenger on the model policy generator.
The model strategy generator 10 receives the report from the model strategy challenger 30 and regenerates a new model strategy. If the variable with the fluctuation amount larger than the set threshold value is encountered, the variable is subjected to elimination or data cleaning treatment, and the influence of the fluctuation on the model strategy is eliminated or reduced.
In a usage scenario of the present invention, the model policy generator may operate more than once (operate at least once), and combine the models and rules still approved by the model policy tracker to form a model rule pool, and the fusion in the model rule pool is generally based on a policy target, and a policy is first formulated by a single model and then a policy is combined, or a policy is first formulated by a plurality of main models, and then a specific and finer policy is formulated by detailed models and rules, which can be taken as an embodiment of the present invention.
The invention also discloses a risk model strategy generation method, and fig. 2 is a flow chart of the risk model strategy generation method in an embodiment of the invention.
The risk model strategy generation method comprises the following steps:
step S11, the variable stability checker screens the stability of the bottom variable and automatically evaluates the stability condition of the variable; the fluctuation generated by the variable selected by the model does not exceed a set threshold value; if the fluctuation generated by the selected variable exceeds a set threshold, the corresponding variable is deleted.
Step S12, the variable stability intervener applies the output result of the variable stability checker to the process of model establishment.
In one embodiment, the specific intervention method of the variable stability intervener comprises at least one of the following ways: the weight of the influencing variable in the model, the probability of the influencing variable entering the model, the sequence of the influencing variable entering the model, and the cleaning and processing method of the influencing variable in the model.
Step S13, the multi-model fusion device selects a plurality of models to be fused, each model corresponds to one usage scenario, and different models are used for different scenarios.
In one embodiment, the model policy generator can receive a feedback report from the model policy challenger and regenerate a new model policy based on the received feedback report. If the variable with the fluctuation amount larger than the set threshold value is encountered, the variable is subjected to elimination or data cleaning treatment, and the influence of the fluctuation on the model strategy is eliminated or reduced.
In one embodiment, the model policy generator operates at least once, combining approved models and rules to form a model rule pool; the fusion in the model rule pool firstly makes a strategy through a single model and then combines the strategies according to a strategy target, or sets a main model to make a large strategy, and then makes a strategy of specific details through a detailed model and a rule.
The risk model policy generation method of the present invention may be used in a risk control method, and fig. 4 is a flowchart of a risk control method in an embodiment of the present invention; referring to fig. 4, the risk control method includes:
step S1, the model policy generator selects sample data in the database, selects setting feature data, selects variables, and generates a model and a policy.
In an embodiment of the present invention, the step of generating the model and the policy by the model policy generator specifically includes:
the variable stability checker screens the stability of the bottom variable and automatically evaluates the stability condition of the variable; the fluctuation generated by the variable selected by the model does not exceed a set threshold value; if the fluctuation generated by the selected variable exceeds a set threshold value, deleting the corresponding variable;
and the variable stability intervener applies the output result of the variable stability checker to the model building process. In one embodiment, the specific intervention method of the variable stability intervener comprises at least one of the following ways: the weight of the influencing variable in the model, the probability of the influencing variable entering the model, the sequence of the influencing variable entering the model, and the cleaning and processing method of the influencing variable in the model.
The multi-model fusion device selects a plurality of models for fusion, each model corresponds to one use scene, and different models are used for different scenes.
Step S2, the model policy tracker monitors and tracks various indexes of the model and the policy generated by the model policy generator.
In an embodiment, the indicators comprise an effect indicator comprising models KS, AUC, IV or/and a stability indicator comprising PSI; the model strategy tracker is used for dynamically tracking the index, and if the set index reaches a set threshold value, the model strategy challenger is triggered to work.
Step S3, the model strategy challenger dynamically discovers the deficiencies of the models and strategies and challenges them. In one embodiment, the model strategy challenger carries out one-time all-round challenge on the model independently, can find weak variables and strong variables mainly aiming at the variation condition of the variables, and generates a standard report.
In an embodiment of the present invention, the output format of the model policy challenger is standardized, and such a standardized file is fed back to the model policy generator, so as to achieve the effect of the model policy challenger on the model policy generator.
After receiving the report from the model strategy challenger 30, the model strategy generator 10 regenerates a new model strategy, wherein if a variable with a fluctuation amount greater than a set threshold is encountered, the variable is subjected to a rejection or data cleaning process to eliminate or reduce the influence of the fluctuation on the model strategy.
The model strategy generator 10 operates at least once, and forms a model rule pool by combining the models and rules still approved by the model strategy tracker 20, and the fusion in the model rule pool firstly makes strategies through a single model and then combines the strategies according to strategy targets
In summary, the risk model policy generation system and method provided by the invention can automatically update the risk model policy and improve the accuracy of risk control.
It should be noted that the present application may be implemented in software and/or a combination of software and hardware; for example, it may be implemented using Application Specific Integrated Circuits (ASICs), general purpose computers, or any other similar hardware devices. In some embodiments, the software programs of the present application may be executed by a processor to implement the above steps or functions. As such, the software programs (including associated data structures) of the present application can be stored in a computer-readable recording medium; such as RAM memory, magnetic or optical drives or diskettes, and the like. In addition, some steps or functions of the present application may be implemented using hardware; for example, as circuitry that cooperates with the processor to perform various steps or functions.
The technical features of the embodiments described above may be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the embodiments described above are not described, but should be considered as being within the scope of the present specification as long as there is no contradiction between the combinations of the technical features.
The description and applications of the invention herein are illustrative and are not intended to limit the scope of the invention to the embodiments described above. Effects or advantages referred to in the embodiments may not be reflected in the embodiments due to interference of various factors, and the description of the effects or advantages is not intended to limit the embodiments. Variations and modifications of the embodiments disclosed herein are possible, and alternative and equivalent various components of the embodiments will be apparent to those skilled in the art. It will be clear to those skilled in the art that the present invention may be embodied in other forms, structures, arrangements, proportions, and with other components, materials, and parts, without departing from the spirit or essential characteristics thereof. Other variations and modifications of the embodiments disclosed herein may be made without departing from the scope and spirit of the invention.
Claims (10)
1. A risk model policy generation system, the risk model policy generation system comprising:
the variable stability checker is used for screening the stability of the bottom variable and automatically evaluating the stability condition of the variable; the fluctuation generated by the variable selected by the model does not exceed a set threshold value; if the fluctuation generated by the selected variable exceeds a set threshold value, deleting the corresponding variable;
the variable stability intervener is used for applying the output result of the variable stability checker to the process of establishing the model;
the multi-model fusion device is used for selecting a plurality of models to fuse, each model corresponds to one use scene, and different models are used for different scenes;
the model strategy generator can receive a feedback report of the model strategy challenger and regenerate a new model strategy according to the received feedback report;
if a variable with fluctuation quantity larger than a set threshold value is encountered, the variable is subjected to elimination or data cleaning treatment, and the influence of the fluctuation on the model strategy is eliminated or reduced;
the model strategy generator operates at least once and forms a model rule pool by combining approved models and rules; according to a strategy target, the fusion in the model rule pool firstly makes a strategy through a single model and then combines the strategies, or sets a main model to make a large strategy, and then makes a strategy of specific details through a detailed model and a rule;
the specific intervention method of the variable stability intervener comprises at least one of the following modes: the weight of the influencing variable in the model, the probability of the influencing variable entering the model, the sequence of the influencing variable entering the model, and the cleaning and processing method of the influencing variable in the model.
2. A risk model policy generation system, the risk model policy generation system comprising:
the variable stability checker is used for screening the stability of the bottom variable and automatically evaluating the stability condition of the variable; the fluctuation generated by the variable selected by the model does not exceed a set threshold value; if the fluctuation generated by the selected variable exceeds a set threshold value, deleting the corresponding variable;
the variable stability intervener is used for applying the output result of the variable stability checker to the process of establishing the model;
the multi-model fusion device is used for selecting a plurality of models to be fused, each model corresponds to one use scene, and different models are used for different scenes.
3. The risk model policy generation system of claim 2, wherein:
the model strategy generator can receive a feedback report of the model strategy challenger and regenerate a new model strategy according to the received feedback report;
if the variable with the fluctuation amount larger than the set threshold value is encountered, the variable is subjected to elimination or data cleaning treatment, and the influence of the fluctuation on the model strategy is eliminated or reduced.
4. The risk model policy generation system of claim 2, wherein:
the model strategy generator operates at least once and forms a model rule pool by combining approved models and rules; the fusion in the model rule pool firstly makes a strategy through a single model and then combines the strategies according to a strategy target, or sets a main model to make a large strategy, and then makes a strategy of specific details through a detailed model and a rule.
5. The risk model policy generation system of claim 2, wherein:
the specific intervention method of the variable stability intervener comprises at least one of the following modes: the weight of the influencing variable in the model, the probability of the influencing variable entering the model, the sequence of the influencing variable entering the model, and the cleaning and processing method of the influencing variable in the model.
6. A risk model strategy generation method is characterized by comprising the following steps:
the variable stability checker screens the stability of the bottom variable and automatically evaluates the stability condition of the variable; the fluctuation of the variable selected by the model does not exceed a set threshold value; if the fluctuation generated by the selected variable exceeds a set threshold value, deleting the corresponding variable;
the variable stability intervener applies the output result of the variable stability checker to the process of model establishment;
the multi-model fusion device selects a plurality of models for fusion, each model corresponds to one use scene, and different models are used for different scenes;
the specific intervention method of the variable stability intervener comprises at least one of the following modes: the weight of the influence variable in the model, the probability of the influence variable entering the model, the sequence of the influence variable entering the model, and the cleaning and processing method of the influence variable in the model;
the model strategy generator can receive a feedback report of the model strategy challenger and regenerate a new model strategy according to the received feedback report;
if a variable with fluctuation quantity larger than a set threshold value is encountered, the variable is subjected to elimination or data cleaning treatment, and the influence of the fluctuation on the model strategy is eliminated or reduced;
the model strategy generator operates at least once and forms a model rule pool by combining approved models and rules; the fusion in the model rule pool firstly makes a strategy through a single model and then combines the strategies according to a strategy target, or sets a main model to make a large strategy, and then makes a strategy of specific details through a detailed model and a rule.
7. A risk model strategy generation method is characterized by comprising the following steps:
the variable stability checker screens the stability of the bottom variable and automatically evaluates the stability condition of the variable; the fluctuation of the variable selected by the model does not exceed a set threshold value; if the fluctuation generated by the selected variable exceeds a set threshold value, deleting the corresponding variable;
the variable stability intervener applies the output result of the variable stability checker to the process of model establishment;
the multi-model fusion device selects a plurality of models for fusion, each model corresponds to one use scene, and different models are used for different scenes.
8. The risk model policy generation method of claim 7, wherein:
the specific intervention method of the variable stability intervener comprises at least one of the following modes: the weight of the influencing variable in the model, the probability of the influencing variable entering the model, the sequence of the influencing variable entering the model, and the cleaning and processing method of the influencing variable in the model.
9. The risk model policy generation method of claim 7, wherein:
the model strategy generator can receive a feedback report of the model strategy challenger and regenerate a new model strategy according to the received feedback report;
if the variable with the fluctuation amount larger than the set threshold value is encountered, the variable is subjected to elimination or data cleaning treatment, and the influence of the fluctuation on the model strategy is eliminated or reduced.
10. The risk model policy generation method of claim 7, wherein:
the model strategy generator operates at least once and forms a model rule pool by combining approved models and rules; the fusion in the model rule pool firstly makes a strategy through a single model and then combines the strategies according to a strategy target, or sets a main model to make a large strategy, and then makes a strategy of specific details through a detailed model and a rule.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110135369.3A CN112861955A (en) | 2021-02-01 | 2021-02-01 | Risk model strategy generation system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110135369.3A CN112861955A (en) | 2021-02-01 | 2021-02-01 | Risk model strategy generation system and method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112861955A true CN112861955A (en) | 2021-05-28 |
Family
ID=75987392
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110135369.3A Pending CN112861955A (en) | 2021-02-01 | 2021-02-01 | Risk model strategy generation system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112861955A (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20180025290A1 (en) * | 2016-07-22 | 2018-01-25 | Edwards Lifesciences Corporation | Predictive risk model optimization |
| CN108109066A (en) * | 2017-12-11 | 2018-06-01 | 上海前隆信息科技有限公司 | A kind of credit scoring model update method and system |
| CN109063952A (en) * | 2018-06-15 | 2018-12-21 | 阿里巴巴集团控股有限公司 | Strategy generating and risk control method and device |
| 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 |
| CN111353850A (en) * | 2020-02-26 | 2020-06-30 | 支付宝(杭州)信息技术有限公司 | Risk identification strategy updating method and device and risk merchant identification method and device |
| CN111861704A (en) * | 2020-07-10 | 2020-10-30 | 深圳无域科技技术有限公司 | Wind control feature generation method and system |
-
2021
- 2021-02-01 CN CN202110135369.3A patent/CN112861955A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20180025290A1 (en) * | 2016-07-22 | 2018-01-25 | Edwards Lifesciences Corporation | Predictive risk model optimization |
| CN108109066A (en) * | 2017-12-11 | 2018-06-01 | 上海前隆信息科技有限公司 | A kind of credit scoring model update method and system |
| CN109063952A (en) * | 2018-06-15 | 2018-12-21 | 阿里巴巴集团控股有限公司 | Strategy generating and risk control method and device |
| 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 |
| CN111353850A (en) * | 2020-02-26 | 2020-06-30 | 支付宝(杭州)信息技术有限公司 | Risk identification strategy updating method and device and risk merchant identification method and device |
| CN111861704A (en) * | 2020-07-10 | 2020-10-30 | 深圳无域科技技术有限公司 | Wind control feature generation method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10943186B2 (en) | Machine learning model training method and device, and electronic device | |
| Le Tallec | Robust, risk-sensitive, and data-driven control of Markov decision processes | |
| CN112052071B (en) | Cloud software service resource allocation method combining reinforcement learning and machine learning | |
| Nassif et al. | Regression model for software effort estimation based on the use case point method | |
| CN108089893A (en) | Definite method, apparatus, terminal device and the storage medium of redundant resource | |
| CN110991871A (en) | Risk monitoring method, device, equipment and computer readable storage medium | |
| CN108615189A (en) | A method of based on the processing of block chain data collaborative | |
| CN109102377A (en) | A kind of non-performing asset auction disposition platform based on block chain technology | |
| CN109726821A (en) | Data balancing method, device, computer readable storage medium and electronic equipment | |
| CN116846619A (en) | Automatic network security risk assessment method, system and readable storage medium | |
| CN106815232A (en) | Catalog management method, apparatus and system | |
| CN114138597B (en) | Operating system performance tuning device, method, equipment and storage medium | |
| CN106789947B (en) | Internet of things node trust value evaluation and task delegation method based on environment | |
| CN113342418B (en) | Distributed machine learning task unloading method based on block chain | |
| CN112861955A (en) | Risk model strategy generation system and method | |
| CN117272330B (en) | Method and system for reinforcing and updating server system | |
| Moonen et al. | What are the effects of history length and age on mining software change impact? | |
| CN112749925A (en) | Risk control system and method | |
| DE102009041256A1 (en) | Method and apparatus for enabling a persistent application programming interface | |
| CN108306780B (en) | Cloud environment-based virtual machine communication quality self-optimization system and method | |
| CN111258911A (en) | Software test case generation method, system and storage medium based on data driving and multiple coverage strategies | |
| CN112749921A (en) | Mathematical modeling method, system, device and computer readable medium | |
| CN113298492B (en) | High-response process intelligent processing method and system based on process engine | |
| CN114880664A (en) | Information security risk event element relation and path association analysis method and device | |
| CN114968765A (en) | Software quality evaluation method and device and computer readable storage medium |
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 |