CN116563016A - Wind control management system and wind control management method - Google Patents
Wind control management system and wind control management method Download PDFInfo
- Publication number
- CN116563016A CN116563016A CN202210098773.2A CN202210098773A CN116563016A CN 116563016 A CN116563016 A CN 116563016A CN 202210098773 A CN202210098773 A CN 202210098773A CN 116563016 A CN116563016 A CN 116563016A
- Authority
- CN
- China
- Prior art keywords
- wind control
- data
- risk
- transaction
- result
- 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
- 238000007726 management method Methods 0.000 title claims description 49
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 19
- 238000010801 machine learning Methods 0.000 claims abstract description 14
- 238000000034 method Methods 0.000 claims abstract description 14
- 230000000694 effects Effects 0.000 claims description 16
- 238000012795 verification Methods 0.000 claims description 16
- 238000004590 computer program Methods 0.000 claims description 11
- 238000012549 training Methods 0.000 claims description 10
- 238000010276 construction Methods 0.000 claims description 5
- 238000012360 testing method Methods 0.000 claims description 5
- 238000011217 control strategy Methods 0.000 abstract description 27
- 238000012545 processing Methods 0.000 abstract description 8
- 230000006870 function Effects 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 3
- 238000011156 evaluation Methods 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000012790 confirmation Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000007477 logistic regression Methods 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 238000007637 random forest analysis Methods 0.000 description 2
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000003631 expected effect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
- 230000002087 whitening effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/04—Trading; Exchange, e.g. stocks, commodities, derivatives or currency exchange
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0635—Risk analysis of enterprise or organisation activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Development Economics (AREA)
- Theoretical Computer Science (AREA)
- Educational Administration (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Marketing (AREA)
- Finance (AREA)
- Accounting & Taxation (AREA)
- Game Theory and Decision Science (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Technology Law (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to a wind control management system capable of self-updating wind control strategies and a method thereof. The system comprises: the basic data acquisition system acquires wind control basic data; a risk data mart obtains characteristic factors of the transaction based on the transaction data and the wind control basic data; the risk modeling platform is used for executing modeling of the wind control model based on the characteristic factors of the transaction and a preset machine learning algorithm and outputting the wind control model; the risk scoring system scores the transaction data in real time to obtain a scoring result of the transaction; the real-time wind control system judges the transaction risk by combining the grading result of the transaction with a preset expert rule and outputs an interception result; and the result feedback system is used for feeding the interception feedback node back to the real-time wind control system and the risk data marts respectively. According to the invention, the closed loop processing and self-updating of the wind control system can be realized.
Description
Technical Field
The invention relates to a computer technology, in particular to a wind control management system capable of self-updating a wind control strategy and a wind control management method.
Background
There are two major main flows of common wind control systems in the industry at present: the system comprises an expert rule-based wind control engine and a machine learning-based scoring system.
The expert rule-based wind control engine is used for mainly analyzing the risk transaction which has occurred, and combining the characteristics to form an expert rule after acquiring the fraud characteristics, so that whether the risk exists in the transaction is judged to be non-black or white. Although suspicious transactions may be intercepted in real time after triggering rules, such systems lack the ability to learn rules themselves, requiring significant human resources to invest in making rule adjustments to cope with changes and shifts in risk.
Machine learning-based scoring systems typically require training through existing training samples to obtain an optimal model. The model decays over time, and therefore, a lot of human resources are also required to be input to iteratively optimize the model to be suitable for the latest risk situation. Furthermore, scoring systems are typically just systems that output a score for each transaction, and even if it finds true fraud, it is not possible to abort the transaction.
In summary, both the wind control engine based on expert rules and the scoring system based on machine learning have respective limitations, a large amount of human resources are required to be input to adjust and optimize the wind control rules and the models, and the wind control strategy self-updating of the system cannot be realized.
Disclosure of Invention
In view of the foregoing, the present invention aims to provide a wind control management system and a wind control management method capable of automatically updating a wind control policy according to a risk profile.
A wind control management system capable of self-updating a wind control policy according to an aspect of the present invention is characterized by comprising:
the basic data acquisition system acquires wind control basic data;
the risk data marts are accessed into transaction data and the wind control basic data from the basic data acquisition system, and feature factors of the transaction are obtained based on the transaction data and the wind control basic data;
the risk modeling platform is used for presetting a machine learning algorithm, accessing characteristic factors of the transaction from the risk data marts, executing modeling of a wind control model based on the characteristic factors of the transaction and the preset machine learning algorithm, and outputting the wind control model;
the risk scoring system scores the transaction data in real time based on the wind control model output by the risk modeling platform and outputs the scoring result of the transaction;
the real-time wind control system is used for presetting expert rules, accessing the grading result of the transaction output by the risk grading system, judging the transaction risk by combining the grading result of the transaction with the preset expert rules, and outputting an interception result; and
and the result feedback system is connected with the interception result output by the real-time wind control system, confirms the accuracy of the interception result and feeds back an interception feedback node representing the accuracy of the interception result to the real-time wind control system and the risk data marts respectively.
Optionally, the risk data marketplace comprises: a base data layer; and a derived data layer derived from the base data layer.
Optionally, the derivative data layer includes:
the event library is used for storing risk events;
the list library is used for storing a main list related to the risk event;
the feature library is used for combining the wind control basic data, the risk event and the main body list to form feature data;
a tag library for forming tag data based on the feature data,
the feature data of the feature library and the tag data of the tag library form feature factors of the transaction.
Optionally, the real-time wind control system updates the risk policy based on obtaining the interception feedback result.
Optionally, the updating of the risk policy by the real-time wind control system based on the obtained interception feedback result includes:
blacking the subject list identified as fraudulent transactions; and
white-adding the main body list intercepted by mistake, and the like.
Optionally, the risk data marketplace updates the feature factor based on obtaining the interception feedback result and pushes the updated feature factor to the modeling platform.
Optionally, the risk modeling platform performs the following actions:
determining positive sample data and negative sample data;
completing data set construction, including training data sets, verification data sets and test data sets;
performing model effect verification by selecting the feature factors or the combination of the feature factors and selecting a model algorithm; and
and according to the model effect verification result, finishing characteristic factor and model parameter adjustment, and outputting the wind control model.
The wind control management method capable of self-updating the wind control strategy in one aspect of the invention is characterized by comprising the following steps:
a basic data acquisition step, namely acquiring wind control basic data;
a risk data mart forming step, namely accessing transaction data and the wind control basic data, and obtaining characteristic factors of the transaction based on the transaction data and the wind control basic data;
a wind control model modeling step, namely accessing characteristic factors from the transaction, executing modeling of a wind control model based on the characteristic factors of the transaction and a preset machine learning algorithm, and outputting the wind control model;
a risk scoring step, namely scoring the transaction data in real time based on the wind control model, and outputting a scoring result of the transaction;
the wind control executing step is used for judging the transaction risk according to preset expert rules and combining the grading result of the transaction and outputting an interception result; and
and a result feedback step of confirming the accuracy of the interception result and feeding back the interception feedback result representing the accuracy of the interception result to the wind control execution step and the risk data mart forming step.
Optionally, in the wind control executing step, updating a risk policy based on the obtained interception feedback result.
Optionally, the updating the risk policy based on the obtaining the interception feedback result in the wind control executing step includes:
blacking the subject list identified as fraudulent transactions; and
white-adding the main body list intercepted by mistake, and the like.
Optionally, in the risk data mart forming step, updating a feature factor based on obtaining the interception feedback result.
Optionally, the wind control model modeling step includes:
determining positive sample data and negative sample data;
completing data set construction, including training data sets, verification data sets and test data sets;
performing model effect verification by selecting the feature factors or the combination of the feature factors and selecting a model algorithm; and
and finishing characteristic factor and model parameter adjustment according to the model effect verification result, and outputting the wind control model.
A computer readable medium of an aspect of the present invention has stored thereon a computer program which, when executed by a processor, implements the method of wind control management of a self-updatable wind control strategy.
The computer equipment comprises a storage module, a processor and a computer program which is stored on the storage module and can run on the processor, wherein the wind control management method capable of updating the wind control strategy automatically is realized when the processor executes the computer program.
Drawings
FIG. 1 is a block diagram illustrating the configuration of a self-updatable air management system according to the present invention.
FIG. 2 is a flow chart illustrating a method of wind control management for a self-updatable wind control strategy according to the present invention.
Detailed Description
The following presents a simplified summary of the invention in order to provide a basic understanding of the invention. It is not intended to identify key or critical elements of the invention or to delineate the scope of the invention.
For the purposes of brevity and explanation, the principles of the present invention are described herein primarily with reference to exemplary embodiments thereof. However, those skilled in the art will readily recognize that the same principles are equally applicable to, and may be implemented in, all types of self-renewable wind control strategy wind control management systems and wind control management methods, and that any such variations do not depart from the true spirit and scope of the present patent application.
Also, in the following description, reference is made to the accompanying drawings that illustrate specific exemplary embodiments. Electrical, mechanical, logical and structural changes may be made to these embodiments without departing from the spirit and scope of the present invention. Furthermore, while a feature of the invention may have been disclosed with respect to only one of several implementations/embodiments, such feature may be combined with one or more other features of the other implementations/embodiments, as may be desired and/or advantageous for any given or identifiable function. The following description is, therefore, not to be taken in a limiting sense, and the scope of the present invention is defined by the appended claims and their equivalents.
Terms such as "comprising" and "including" mean that the technical solution of the present invention does not exclude the presence of other elements (modules) and steps than those directly and explicitly described in the description and claims.
The wind control management system capable of self-updating the wind control strategy is formed by constructing a basic data acquisition system, a risk data mart, a risk modeling platform, a risk scoring system, a transaction real-time interception system and a result feedback system, and connecting the systems in series to form a set of closed-loop processing system, so that the wind control system capable of automatically updating the wind control strategy according to a risk form is formed.
The wind control management method capable of self-updating the wind control strategy is characterized in that basic data acquisition, collection, feature factor calculation, modeling, grading, transaction interception and interception accuracy confirmation are carried out, and then the confirmed transaction is updated to basic data, so that feature factor update, model update and wind control strategy update are realized, and a risk management method for closed-loop processing is formed.
The wind control management system and the wind control management method capable of self-updating the wind control strategy of the invention can be applied to wind control of transaction data as an example, and the following description mainly takes the transaction data as a data processing object.
FIG. 1 is a block diagram illustrating the configuration of a self-updatable air management system according to the present invention.
As shown in fig. 1, the wind control management system of the self-updatable wind control strategy of the present invention includes:
the basic data acquisition system 100, the risk data mart 200, the risk modeling platform 300, the risk scoring system 400, the real-time wind-controlled transaction interception system 500 and the result feedback system 600 are connected in series.
Next, the respective functions of the basic data acquisition system 100, the risk data bazaar 200, the risk modeling platform 300, the risk scoring system 400, the real-time wind-controlled transaction interception system 500, and the result feedback system 600 will be described.
The underlying data collection system 100 is used to collect wind control underlying data, such as wind control underlying data (which may also be understood as "negative sample data") including collection of blacklists of fraudulent transaction details, card numbers, merchant numbers, equipment numbers, etc., and to push the collected wind control underlying data to the risk event marketplace 200.
Risk data marketplace 200 may be divided into a base data layer and a derivative data layer.
The risk data mart 200 gathers the transaction data and the wind control basic data (i.e. "negative sample data") to a big data platform, and then processes the transaction data and the wind control basic data (i.e. "negative sample data") to count the data characteristics of the negative samples so as to obtain characteristic factors of the card number, the merchant number, the equipment number in the time dimension, the region dimension, the transaction type and other dimensions, thereby forming the risk data mart.
The basic data layer is used for accessing transaction data, for example, including online and offline transaction details, mobile payment transaction details, and the like, and also accessing wind control basic data, and mainly includes fraud transaction details, card numbers, merchant numbers, black-gray lists of equipment numbers, and the like, which are acquired by the radix data acquisition system 100.
The derived data layer may be divided into four libraries, such as an event library, a list library, a feature library, and a tag library, respectively, as an example.
The event library stores all risk events including fraud report, monitoring cases, supervision complaints, information leakage and the like;
the list library contains a main list related to the risk event, such as card number, card BIN, equipment ID, mobile phone number, IP address, merchant ID, terminal ID, and other information;
the feature library is a combination feature of different dimensions such as a card, equipment, a merchant, a mobile phone number and the like which are formed by combining the basic data, the event library and the list library;
the tag library is a tag which is formed into dimensions of regions, institutions, merchants, terminals, cards, equipment and the like according to different characteristic combinations.
The feature library and the tag library may mainly convey feature factors and tag data for the risk modeling platform 300.
The risk modeling platform 300 is pre-configured with several commonly used machine learning algorithms, including, for example, logistic regression, neural networks, random forests, GBDT, etc., and builds a corresponding baseline model in conjunction with prior modeling experience. The feature factors in the risk data mart 200 are butted to the risk modeling platform 300, then model effect verification and evaluation are carried out through feature factor combination and multi-algorithm model training, and finally the feature factors and specific parameters used by the model are confirmed through feature importance analysis and model parameter adjustment and optimization, so that the off-line modeling process is completed.
The risk modeling platform 300 specifically performs the following actions including:
(1) Explicitly modeling a target, and determining an application scene and an expected effect;
(2) Sample data definition, determining positive sample data and negative sample data;
(3) Completing data set construction, including training data sets, verification data sets and test data sets;
(4) Selecting proper characteristic factors and a model algorithm to verify the model effect; and
(5) And comparing the model effect with the baseline model effect to finish the adjustment of the characteristic factors and the model parameters.
The risk scoring system 400 achieves a real-time scoring function for transactions through feature factor invocation and model algorithm invocation and sends real-time scoring result pairs into the real-time wind control system 500.
The real-time wind control system 500 is used for realizing real-time interception of transactions, and mainly refers to a transaction real-time interception function based on expert rules. The real-time wind control system 500 comprehensively judges the risk of the transaction by acquiring the real-time scoring result of the risk scoring system 400 (for example, acquiring the real-time scoring result of each transaction) of the risk scoring system 400 and combining with preset expert rules, and outputs the intercepting result to be transmitted to the result feedback system 600, so that the coverage rate and the accuracy rate of the intercepting of the risk transaction can be effectively improved.
The result feedback system 600 is configured to receive an interception result (i.e. a "transaction decision" in the figure) from the real-time wind control system 500, and the result feedback system 600 confirms the accuracy of the interception result through a customer service outbound call, a short message notification, etc., and then transmits an interception feedback result indicating whether the interception is accurate to the real-time wind control system 500 and the risk data mart 200.
The purpose of transmitting the interception feedback result to the real-time wind control system 500, so that the real-time wind control system 500 is synchronized to implement updating of the risk policy, includes:
blackening the information such as the card number and the equipment number which are confirmed to be fraudulent transactions; and
and (3) whitening information such as the card number, the equipment number and the like intercepted by mistake.
The purpose of transmitting the interception feedback result to the risk data bazaar 200 is to synchronize to the risk data bazaar 200 to update the feature factors, and push the updated feature factors to the modeling platform 300 to update the model. Further, the updated model is pushed to the scoring system 400, and the scoring system 400 pushes the new score to the real-time wind control system 500, so as to form a risk management system of closed-loop processing.
The above description has been made of the wind control management system of the self-updatable wind control strategy of the present invention. Next, a method for managing wind control with self-renewable wind control policy according to the present invention will be described.
FIG. 2 is a flow chart illustrating a method of wind control management for a self-updatable wind control strategy according to the present invention.
As shown in fig. 2, the wind control management method of the self-renewable wind control strategy of the present invention is characterized by comprising the following steps:
basic data acquisition step S100: collecting wind control basic data, such as negative sample data, such as fraudulent transaction details, card numbers, merchant numbers, black gray lists of equipment numbers, and the like;
risk data mart formation step S200: accessing transaction data and the wind control basic data, acquiring characteristic factors of the transaction based on the transaction data and the wind control basic data, for example, collecting basic transaction data and negative sample data, and counting characteristic factors of negative sample data, such as card numbers, merchant numbers, equipment numbers in time dimension, region dimension, transaction types and other dimensions, through processing the basic data and the negative sample data to form a risk data bazaar;
wind control model modeling step S300: accessing characteristic factors from the transaction, executing modeling of a wind control model based on the characteristic factors of the transaction and a preset machine learning algorithm, outputting the wind control model, specifically, for example, according to the characteristic factors in a risk data mart, using the preset machine learning algorithm such as logistic regression, random forest, neural network and the like, performing model effect verification and evaluation through characteristic factor combination and multi-algorithm model training, and finally, optimizing and confirming that a model uses the characteristic factors and specific parameters through characteristic importance analysis and model parameter adjustment, completing a modeling process, and outputting a risk model;
risk scoring step S400: scoring the transaction data in real time based on the wind control model to obtain a scoring result of the transaction;
the wind control executes step S500: judging the transaction risk according to preset expert rules and the grading result of the transaction, and outputting an interception result, for example, taking the grading result of the transaction as a factor in the expert rules to participate in the configuration of the wind control rules as an example; and
result feedback step S600: and according to the interception result, confirming the accuracy of the interception result, and feeding back an interception feedback result representing the accuracy of the interception result to the wind control executing step and the risk data mart forming step.
In the wind control executing step S500, the risk policy is further updated based on the obtained interception feedback result. The updating includes: blacking the subject list identified as fraudulent transactions; and whitewashing the main body list intercepted by mistake, etc.
In the step S200 of forming the risk data mart, the feature factor is further updated based on obtaining the interception feedback result.
According to the wind control management system and the wind control management method capable of self-updating the wind control strategy, disclosed by the invention, the risk management system for closed-loop processing is formed by collecting and collecting basic data, calculating the characteristic factors, modeling, grading, intercepting the transaction, confirming the interception accuracy, and updating the basic data of the confirmed transaction, so that the characteristic factor updating, the model updating and the wind control strategy updating are realized.
In the wind control management system and the wind control management method capable of self-updating the wind control strategy, the machine learning model can be dynamically adjusted and optimized according to the risk form: the modeling process is completed by butting the characteristic factors in the risk data marts to a risk modeling platform, then performing model effect verification and evaluation through characteristic factor combination and multi-algorithm model training, and finally confirming that the model uses the characteristic factors and specific parameters through characteristic importance analysis and model parameter adjustment and optimization. And meanwhile, according to the accuracy confirmation after transaction interception, the characteristic factors in the risk data marts are updated and transmitted to the risk modeling platform, so that the effect of updating the model is realized.
In the wind control management system and the wind control management method capable of self-updating the wind control strategy, expert rules in the real-time wind control system can be dynamically adjusted: and combining the expert rules with the real-time scoring results, and updating the expert rules and the scoring model according to the interception feedback results representing the interception accuracy after transaction interception, thereby realizing the effect of automatically updating the wind control strategy.
As described above, according to the wind control management system and the wind control management method capable of self-updating the wind control strategy, the problem that the traditional wind control system needs to rely on a large amount of human resources to frequently adjust the wind control strategy can be solved, and the closed-loop processing and self-updating of the wind control system are realized.
The invention also provides a computer readable medium, on which a computer program is stored, characterized in that the computer program, when executed by a processor, implements the method for managing wind control of a self-updatable wind control strategy.
The invention also provides a computer device, which comprises a storage module, a processor and a computer program stored on the storage module and capable of running on the processor, and is characterized in that the wind control management method capable of updating the wind control strategy by itself is realized when the processor executes the computer program.
The above examples mainly illustrate the wind control management system and the wind control management method of the self-updatable wind control strategy of the present invention. Although only a few specific embodiments of the present invention have been described, those skilled in the art will appreciate that the present invention may be embodied in many other forms without departing from the spirit or scope thereof. Accordingly, the present examples and embodiments are to be considered as illustrative and not restrictive, and the invention is intended to cover various modifications and substitutions without departing from the spirit and scope of the invention as defined by the appended claims.
Claims (14)
1. A wind control management system, comprising:
the basic data acquisition system acquires wind control basic data;
the risk data marts are accessed into transaction data and the wind control basic data from the basic data acquisition system, and feature factors of the transaction are obtained based on the transaction data and the wind control basic data;
the risk modeling platform is used for presetting a machine learning algorithm, accessing characteristic factors of the transaction from the risk data marts, executing modeling of a wind control model based on the characteristic factors of the transaction and the preset machine learning algorithm, and outputting the wind control model;
the risk scoring system scores the transaction data in real time based on the wind control model output by the risk modeling platform and outputs the scoring result of the transaction;
the real-time wind control system is used for presetting expert rules, accessing the grading result of the transaction output by the risk grading system, judging the transaction risk by combining the grading result of the transaction with the preset expert rules, and outputting an interception result; and
and the result feedback system is connected with the interception result output by the real-time wind control system, confirms the accuracy of the interception result and feeds back an interception feedback node representing the accuracy of the interception result to the real-time wind control system and the risk data marts respectively.
2. The wind control management system of claim 1, wherein,
the risk data marketplace includes: a base data layer; and a derived data layer derived from the base data layer.
3. The wind control management system of claim 1, wherein,
the derivative data layer comprises:
the event library is used for storing risk events;
the list library is used for storing a main list related to the risk event;
the feature library is used for combining the wind control basic data, the risk event and the main body list to form feature data;
a tag library for forming tag data based on the feature data,
the feature data of the feature library and the tag data of the tag library form feature factors of the transaction.
4. The wind control management system of claim 1, wherein,
and the real-time wind control system updates the risk strategy based on the acquired interception feedback result.
5. The wind control management system of claim 4, wherein,
the updating of the risk strategy by the real-time wind control system based on the interception feedback result comprises the following steps:
blacking the subject list identified as fraudulent transactions; and
white-adding the main body list intercepted by mistake, and the like.
6. The wind control management system of claim 1, wherein,
and the risk data mart updates the characteristic factors based on the acquired interception feedback result and pushes the updated characteristic factors to the modeling platform.
7. The wind control management system of claim 1, wherein,
the risk modeling platform performs the following actions:
determining positive sample data and negative sample data;
completing data set construction, including training data sets, verification data sets and test data sets;
performing model effect verification by selecting the feature factors or the combination of the feature factors and selecting a model algorithm; and
and according to the model effect verification result, finishing characteristic factor and model parameter adjustment, and outputting the wind control model.
8. The wind control management method is characterized by comprising the following steps of:
a basic data acquisition step, namely acquiring wind control basic data;
a risk data mart forming step, namely accessing transaction data and the wind control basic data, and obtaining characteristic factors of the transaction based on the transaction data and the wind control basic data;
a wind control model modeling step, namely accessing characteristic factors from the transaction, executing modeling of a wind control model based on the characteristic factors of the transaction and a preset machine learning algorithm, and outputting the wind control model;
a risk scoring step, namely scoring the transaction data in real time based on the wind control model, and outputting a scoring result of the transaction;
the wind control executing step is used for judging the transaction risk according to preset expert rules and combining the grading result of the transaction and outputting an interception result; and
and a result feedback step of confirming the accuracy of the interception result and feeding back the interception feedback result representing the accuracy of the interception result to the wind control execution step and the risk data mart forming step.
9. The method for managing air control according to claim 8, wherein,
and in the wind control executing step, updating a risk strategy based on the interception feedback result.
10. The method for managing air control according to claim 4, wherein,
the updating of the risk policy based on the interception feedback result in the wind control executing step comprises the following steps:
blacking the subject list identified as fraudulent transactions; and
white-adding the main body list intercepted by mistake, and the like.
11. The method for managing air control according to claim 8, wherein,
and in the risk data mart forming step, updating the characteristic factors based on the acquired interception feedback result.
12. The method for managing air control according to claim 8, wherein,
the wind control model modeling step comprises the following steps:
determining positive sample data and negative sample data;
completing data set construction, including training data sets, verification data sets and test data sets;
performing model effect verification by selecting the feature factors or the combination of the feature factors and selecting a model algorithm; and
and finishing characteristic factor and model parameter adjustment according to the model effect verification result, and outputting the wind control model.
13. A computer readable medium having a computer program stored thereon, characterized in that,
the computer program when executed by a processor implements the wind control management method of any one of claims 8 to 12.
14. A computer device comprising a memory module, a processor and a computer program stored on the memory module and executable on the processor, characterized in that the processor implements the wind control management method according to any one of claims 8-12 when executing the computer program.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210098773.2A CN116563016A (en) | 2022-01-27 | 2022-01-27 | Wind control management system and wind control management method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210098773.2A CN116563016A (en) | 2022-01-27 | 2022-01-27 | Wind control management system and wind control management method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116563016A true CN116563016A (en) | 2023-08-08 |
Family
ID=87496977
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210098773.2A Pending CN116563016A (en) | 2022-01-27 | 2022-01-27 | Wind control management system and wind control management method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116563016A (en) |
-
2022
- 2022-01-27 CN CN202210098773.2A patent/CN116563016A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11620652B1 (en) | Systems, methods, and apparatuses for implementing user customizable risk management tools with statistical modeling and recommendation engine | |
| CN110232565B (en) | Resource clearing method, device, computer equipment and storage medium | |
| CN113011973B (en) | Method and equipment for financial transaction supervision model based on intelligent contract data lake | |
| WO2020052168A1 (en) | Anti-fraud model generation and application method, device and equipment, and storage medium | |
| CN112860676B (en) | Data cleaning method applied to big data mining and business analysis and cloud server | |
| CN107360117A (en) | The method, apparatus and system of data processing | |
| CN112418659A (en) | Cross-border commodity tracing method, system and device based on block chain system | |
| CN116777633A (en) | Financial asset management system based on data management | |
| CN111767192A (en) | Service data detection method, device, equipment and medium based on artificial intelligence | |
| CN107871213B (en) | Transaction behavior evaluation method, device, server and storage medium | |
| CN113744054A (en) | Anti-fraud method, device and equipment | |
| CN116563016A (en) | Wind control management system and wind control management method | |
| CN111738824A (en) | Method, device and system for screening financial data processing modes | |
| CN113837481B (en) | Financial big data management system based on block chain | |
| CN112561696B (en) | Machine learning-based money backwashing system and method | |
| CN115801306A (en) | Data processing method and server applied to artificial intelligence | |
| CN110087230B (en) | Data processing method, data processing device, storage medium and electronic equipment | |
| CN112651704B (en) | Positioning-based meeting judgment method and server | |
| CN110232612A (en) | Product method for pushing, device, computer equipment and storage medium | |
| CN112801561B (en) | User relationship determination method and device, storage medium and electronic equipment | |
| CN113570333B (en) | Process design method suitable for integration | |
| CN117114589B (en) | Inventory management system of cross-border enterprise marketing products based on Internet | |
| CN115860904A (en) | Credit abnormal group mining method and device | |
| KR20090079398A (en) | System for risk estimation and method for operating the same | |
| CN115564422A (en) | Automatic butt-joint management method and system for bill staging service |
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 |