CN111507829A - Overseas credit card wind control model iteration method, device, equipment and storage medium - Google Patents
Overseas credit card wind control model iteration method, device, equipment and storage medium Download PDFInfo
- Publication number
- CN111507829A CN111507829A CN202010322281.8A CN202010322281A CN111507829A CN 111507829 A CN111507829 A CN 111507829A CN 202010322281 A CN202010322281 A CN 202010322281A CN 111507829 A CN111507829 A CN 111507829A
- Authority
- CN
- China
- Prior art keywords
- data
- credit card
- model
- control model
- wind control
- 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
- 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
- G06Q40/03—Credit; Loans; Processing thereof
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Software Systems (AREA)
- General Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Data Mining & Analysis (AREA)
- Strategic Management (AREA)
- Marketing (AREA)
- Economics (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Technology Law (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Development Economics (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
Abstract
The invention relates to the technical field of credit card wind control, in particular to an overseas credit card wind control model iteration method, a device, equipment and a storage medium, which comprises the following steps: s10: acquiring a preset overseas credit card wind control model, and acquiring sub-model data from the overseas credit card wind control model; s20: extracting features to be identified from each sub-model data; s30: acquiring real-time transaction data according to the to-be-identified characteristics of the overseas credit card wind control model, and performing effect judgment on the real-time transaction data to obtain a judgment result; s40: and updating the sub-model data in the overseas credit card wind control model by adopting machine learning according to the judgment result. The method and the system have the effects of updating the overseas credit card wind control system in real time and ensuring the accuracy of risk detection.
Description
Technical Field
The invention relates to the technical field of credit card wind control, in particular to an overseas credit card wind control model iteration method, device, equipment and storage medium.
Background
In recent years, with the rapid development of economy and the continuous deepening of open to the external economy, the network payment service is rapidly developed, various industries gradually pay attention to the credit card service, particularly the rapid development of overseas mobile internet economy, so that various internet service industries are involved in the credit card service in different modes, and a credit card payment mode is provided for providing goods and services.
In the existing use scene, most merchants access a bank payment system and use the own wind control mechanism of the bank, and meanwhile, some merchants also have the payment risk of dealing with overseas credit cards by adopting the wind control system with self-set rules.
The above prior art solutions have the following drawbacks:
although some merchants adopt the wind control system with the self-set rules to deal with the payment risks of overseas credit cards, the effect is acceptable in a short term, but along with the large quantity of users and the complex diversity of user behaviors, the credit card fraud prevention and control system with the set rule mode cannot adapt to the ever-increasing mass data and the constantly-changing user payment behaviors, so that the judgment criteria are not perfect, misjudgment and missed judgment are easy to occur, the payment experience of the users is greatly discounted, and the economic benefit of the companies is further influenced.
Disclosure of Invention
The invention aims to provide an overseas credit card wind control model iteration method, an overseas credit card wind control model iteration device, overseas credit card wind control model iteration equipment and a storage medium, wherein the overseas credit card wind control system can be updated in real time, and the accuracy of risk detection is ensured.
The above object of the present invention is achieved by the following technical solutions:
an overseas credit card wind control model iteration method, comprising the following steps:
s10: acquiring a preset overseas credit card wind control model, and acquiring sub-model data from the overseas credit card wind control model;
s20: extracting features to be identified from each sub-model data;
s30: acquiring real-time transaction data according to the to-be-identified characteristics of the overseas credit card wind control model, and performing effect judgment on the real-time transaction data to obtain a judgment result;
s40: and updating the sub-model data in the overseas credit card wind control model by adopting machine learning according to the judgment result.
By adopting the technical scheme, the sub-model data are extracted from the trained overseas credit card wind control model, and the characteristics to be identified are extracted from the sub-model data, so that the real-time transaction data can be acquired and analyzed when the overseas credit card wind control model is actually used; meanwhile, effect judgment is carried out on the real-time transaction data, machine learning, model optimization and dynamic model regulation are adopted according to the judgment result, each abnormal model is perfected, and system decision-making power is improved.
The present invention in a preferred example may be further configured to: step S10 includes:
s11: obtaining historical user payment data, wherein the historical user payment data comprises equipment and network data, social data, user behavior data and business data;
s12: preprocessing the historical payment data to obtain data to be subjected to feature extraction;
s13: performing feature extraction on the data to be subjected to feature extraction through regularization to obtain a data set to be trained;
s14: and respectively modeling the equipment and the data sets to be trained corresponding to the network data, the social data, the user behavior data and the business data, and then training to obtain the credit card wind control model for detecting the risk level of the client.
By adopting the technical scheme, the historical user payment data is obtained according to the latitude of the equipment and network data, the social data, the user behavior data and the business data, and the historical user payment data can be grouped and classified when being obtained, so that the subsequent feature extraction is facilitated; by preprocessing the acquired historical user payment data, the integrity of the historical user data can be ensured to adopt regularized processing when the characteristics are extracted, the overfitting condition can be avoided, and the finally obtained credit card wind control model is improved to be wider in application range; the credit card wind control model is obtained by respectively carrying out training after modeling on the equipment and the data sets to be trained corresponding to the network data, the social data, the user behavior data and the business data, can be suitable for the current use scene of overseas credit cards, can quickly identify the risk level of the user when the user carries out cross-country credit card consumption, and further can reduce the risk of bad accounts and ensure the benefits of merchants.
The present invention in a preferred example may be further configured to: step S30 includes:
s31: obtaining model test data according to the overseas credit card wind control model;
s32: and comparing the real-time transaction data with the model test data to obtain the judgment result.
By adopting the technical scheme, the accurate judgment result can be obtained by comparing the model test data with the real-time transaction data.
The present invention in a preferred example may be further configured to: step S40 includes:
s41: obtaining abnormal data from the judgment result, and obtaining corresponding sub-model data according to the abnormal data to serve as a sub-model to be adjusted;
s42: and adjusting parameters of the submodel to be adjusted.
By adopting the technical scheme, the sub-model data is obtained from the abnormal data obtained from the judgment result, the abnormal part in the credit card wind control model can be adjusted in a targeted manner, and the efficiency of updating iteration of credit card wind control is improved.
The present invention in a preferred example may be further configured to: step S42 includes:
s421: acquiring corresponding rule weight and rule threshold from the submodel to be adjusted;
s422: and adjusting the rule weight and the rule threshold according to the judgment result.
By adopting the technical scheme, the rule weight and the rule threshold value are dynamically adjusted through machine learning, and the upgrading and updating period of the overseas credit card pneumatic control model is shortened through a real-time machine learning means.
The second aim of the invention is realized by the following technical scheme:
an overseas credit card pneumatic control model iteration apparatus, the overseas credit card pneumatic control model iteration apparatus comprising:
the system comprises a sub-model acquisition module, a sub-model acquisition module and a sub-model management module, wherein the sub-model acquisition module is used for acquiring a preset overseas credit card wind control model and acquiring sub-model data from the overseas credit card wind control model;
the identification feature acquisition module is used for extracting features to be identified from each sub-model data;
the judging module is used for acquiring real-time transaction data according to the to-be-identified characteristics of the overseas credit card wind control model, and performing effect judgment on the real-time transaction data to obtain a judgment result;
and the tuning module is used for updating the sub-model data in the overseas credit card wind control model by adopting machine learning according to the judgment result.
By adopting the technical scheme, the sub-model data are extracted from the trained overseas credit card wind control model, and the characteristics to be identified are extracted from the sub-model data, so that the real-time transaction data can be acquired and analyzed when the overseas credit card wind control model is actually used; meanwhile, effect judgment is carried out on the real-time transaction data, machine learning, model optimization and dynamic model regulation are adopted according to the judgment result, each abnormal model is perfected, and system decision-making power is improved.
The third object of the invention is realized by the following technical scheme:
a computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, the processor implementing the steps of the above-described overseas credit card pneumatic control model iteration method when executing the computer program.
The fourth object of the invention is realized by the following technical scheme:
a computer readable storage medium storing a computer program which, when executed by a processor, implements the steps of the above-described overseas credit card pneumatic control model iteration method.
In summary, the invention includes at least one of the following beneficial technical effects:
1. by extracting sub-model data from the trained overseas credit card wind control model and extracting the characteristics to be identified from the sub-model data, real-time transaction data can be acquired and analyzed when the overseas credit card wind control model is actually used;
2. the effect of the real-time transaction data is judged, and machine learning, model tuning and model rule dynamic adjustment are adopted according to the judgment result, so that each abnormal model is perfected, and the decision-making power of the system is improved;
3. and dynamically adjusting the rule weight and the rule threshold value through machine learning, and shortening the upgrading and updating period of the overseas credit card wind control model through a real-time machine learning means.
Drawings
FIG. 1 is a flow chart of an iterative method of overseas credit card atmospheric control model in one embodiment of the present invention;
FIG. 2 is a flowchart illustrating an implementation of step S10 in the overseas credit card pneumatic control model iteration method according to an embodiment of the present invention;
FIG. 3 is a flowchart illustrating an implementation of step S30 in the overseas credit card pneumatic control model iteration method according to an embodiment of the present invention;
FIG. 4 is a flowchart illustrating the implementation of step S40 in the overseas credit card pneumatic control model iteration method according to an embodiment of the present invention;
FIG. 5 is a flowchart illustrating the implementation of step S42 in the overseas credit card pneumatic control model iteration method according to an embodiment of the present invention;
FIG. 6 is a schematic block diagram of an overseas credit card pneumatic control model iteration apparatus in an embodiment of the present invention;
FIG. 7 is a schematic diagram of a computer device according to an embodiment of the invention.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings.
The first embodiment is as follows:
in an embodiment, as shown in fig. 1, the present invention discloses an overseas credit card wind control model iteration method, which specifically includes the following steps:
s10: and acquiring a preset overseas credit card wind control model, and acquiring sub-model data from the overseas credit card wind control model.
In this embodiment, the overseas credit card pneumatic control model is a model that is trained in advance, and is used for detecting the risk level of the user when the user consumes the credit card, and making a corresponding decision according to the risk level. Submodel data refers to submodels used in overseas credit card pneumatic models to detect data of different dimensions.
Specifically, the overseas credit card pneumatic control model is trained in advance according to historical data, and corresponding sub-model data are obtained according to the latitude of the data obtained from the overseas credit card pneumatic control model.
S20: features to be identified are extracted from each sub-model data.
In this embodiment, the feature to be recognized refers to a feature that each sub-model data needs to be extracted from a large amount of data.
Specifically, when the overseas credit card wind control model is trained, the content of data which needs to be extracted specifically is set when each sub-model data is trained to be actually used, and when each sub-model data is actually used, features are extracted from corresponding data in data generated in actual transactions, and the features which need to be extracted are used as the features to be identified.
S30: and acquiring real-time transaction data according to the to-be-identified characteristics of the overseas credit card wind control model, and performing effect judgment on the real-time transaction data to obtain a judgment result.
In this embodiment, the real-time transaction data refers to data of a result obtained by detecting data generated during a transaction through an overseas credit card pneumatic control model when a user conducts a transaction in an overseas e-commerce transaction platform.
Specifically, after the overseas credit card pneumatic control model is loaded on an actual overseas e-commerce platform, when a transaction is actually generated, the feature to be identified corresponding to each sub-model data in the overseas credit card pneumatic control model is used as a placeholder, and detection data and decision results corresponding to each sub-model data are obtained from data generated in the actual transaction, such as software and hardware information of a user, information of a network where the user is located, historical records of the user or data of registration records, browsing and payment, and the like, so as to obtain the real-time transaction data.
Further, data of a comparison standard corresponding to the real-time transaction data is obtained, and the data of the comparison standard is a result after actual transaction. And comparing the decision result in the real-time transaction data with the result after actual transaction, and judging to obtain the judgment result.
S40: and updating the sub-model data in the overseas credit card wind control model by adopting machine learning according to the judgment result.
Specifically, if the result obtained from the judgment result is inconsistent, that is, the data actually generating the transaction is analyzed through the oversea credit card climate control model, a decision for allowing the user to perform the transaction is obtained, and the transaction information of the user generates bad accounts from the actual result, or the user is judged or users with the same attribute are judged to be users with medium or low risk, and the decision is made to allow the user to perform the transaction, and the probability of the bad accounts of the user obtained from the actual result is high, it is proved that a new fraud or a similar fraud means may occur, so that the parameters in the oversea credit card climate control model need to be adjusted. It should be noted that the probability is higher, which may be determined by human, or a related probability threshold may be set, and if the probability of the bad account exceeds the threshold, the probability is determined to be higher, which is not limited herein.
Further, when the result is inconsistent, the actual transaction data corresponding to the situation is correspondingly split according to the sub-model data in the overseas credit card pneumatic control model, so that the split data is the data which needs to be analyzed by each sub-model data.
Further, each sub-model data performs machine learning on the obtained data to adjust parameters of the sub-model data in real time, for example, if it is determined that an overfitting phenomenon occurs, parameters of contents of the sub-model data for preventing the overfitting are adjusted.
In the embodiment, by extracting sub-model data from a trained overseas credit card wind control model and extracting features to be identified from the sub-model data, real-time transaction data can be acquired and analyzed when the overseas credit card wind control model is actually used; meanwhile, effect judgment is carried out on the real-time transaction data, machine learning, model optimization and dynamic model regulation are adopted according to the judgment result, each abnormal model is perfected, and system decision-making power is improved.
In an embodiment, as shown in fig. 2, in step S10, acquiring a preset overseas credit card pneumatic control model specifically includes the following steps:
s11: obtaining historical user payment data, wherein the historical user payment data comprises equipment and network data, social data, user behavior data and business data.
In this embodiment, the historical user payment data refers to data generated when a user conducts credit card transactions in a multinational e-commerce transaction platform before the relevant model is initially trained. The device and network data refer to the device for payment and the data of the network environment when the user pays the credit card. The social data refers to data such as social associated account numbers, address book information and call records of the users. The user behavior data refers to the data of the registration login of the user and the operation behavior of the E-commerce transaction platform. The business data includes business event records, historical payment data and the like. The business event data refers to the time of triggering payment after the user places an order, whether an order is cancelled or not, whether the order is placed for multiple times in a short time or whether a record of triggering the operation of applying refund or the like exists after the payment. The historical payment data specifically refers to the data of financial data flow generated by payment or refund behaviors after the user places an order.
Specifically, the latitude obtained by the data is set according to the equipment and network data, social data, user behavior data and business data, and historical user payment data of each user is obtained in a preset database through the latitude.
S12: and preprocessing the historical payment data to obtain data to be subjected to feature extraction.
In this embodiment, the data to be feature extracted refers to a data set that needs to extract important features for training.
Specifically, the historical payment data is preprocessed, incomplete data in the historical payment data is cleaned, then the incomplete data is stored in a unified mode and converted into a data format capable of being mined, and the data to be subjected to feature extraction is obtained.
S13: and performing feature extraction on the data to be subjected to feature extraction through regularization to obtain a data set to be trained.
In this embodiment, the data set to be trained refers to a feature set of a model that needs to be trained to detect user risk.
The feature extraction method comprises the steps of extracting key features in a feature extraction function in a data set to be feature extracted, wherein L1 regularization terms are added in the feature extraction function, and the problem that after feature selection is trained, an obtained model is over-fitted is avoided.
S14: and respectively modeling and training the equipment and a data set to be trained corresponding to the network data, the social data, the user behavior data and the business data to obtain a credit card wind control model for detecting the risk level of the client.
In this embodiment, the credit card wind control model refers to a model for detecting a risk level of a user when the user triggers a transaction.
Specifically, weak models are respectively constructed for data types in equipment and network data, social data, user behavior data and business data, and the correlation between the weak models is counted to obtain the credit card wind control model.
In the embodiment, historical user payment data are obtained according to the latitude of equipment and network data, social data, user behavior data and business data, and when the historical user payment data are obtained, the historical user payment data can be grouped and classified, so that subsequent feature extraction is facilitated; by preprocessing the acquired historical user payment data, the integrity of the historical user data can be ensured to adopt regularized processing when the characteristics are extracted, the overfitting condition can be avoided, and the finally obtained credit card wind control model is improved to be wider in application range; the credit card wind control model is obtained by respectively carrying out training after modeling on the equipment and the data sets to be trained corresponding to the network data, the social data, the user behavior data and the business data, can be suitable for the current use scene of overseas credit cards, can quickly identify the risk level of the user when the user carries out cross-country credit card consumption, and further can reduce the risk of bad accounts and ensure the benefits of merchants.
In an embodiment, as shown in fig. 3, in step S30, the method specifically includes the following steps of obtaining real-time transaction data according to the feature to be identified of the overseas credit card pneumatic control model, and performing effect judgment on the real-time transaction data to obtain a judgment result:
s31: and obtaining model test data according to the overseas credit card wind control model.
In this embodiment, the model test data is data for testing the decision output by the overseas credit card pneumatic control model using historical transaction data.
Specifically, data for training and testing each submodel data in the overseas credit card wind control model when the overseas credit card wind control model is trained is acquired as the model test data. Wherein the model test data also includes data on attributes of users who have bad accounts, fraud and other abnormal situations in historical payment situations.
S32: and comparing the real-time transaction data with the model test data to obtain a judgment result.
Specifically, the attribute of the user in the real-time transaction data is compared and judged with the attribute of the historical user under the condition that the attribute of the user is the same as the attribute of each model test data, and the judgment result is obtained.
In an embodiment, as shown in fig. 4, in step S40, that is, according to the determination result, the updating of the sub-model data in the overseas credit card pneumatic control model by using machine learning specifically includes the following steps:
s41: and acquiring abnormal data from the judgment result, and acquiring corresponding sub-model data according to the abnormal data to serve as the sub-model to be adjusted.
In this embodiment, the submodel to be adjusted refers to a submodel that needs to be adjusted and updated iteratively in the submodel data.
Specifically, from the information of the attribute of the user in the determination result, data inconsistent with the attribute information of the corresponding historical user in the model test data is searched for as abnormal data.
Further, if abnormal data occurs, it is indicated that the sub-model data used for acquiring and analyzing the abnormal data has a phenomenon that the attribute of the current user does not match, and the sub-module data of the model test data having the phenomenon that does not match is used as the sub-model to be adjusted.
S42: and adjusting parameters of the submodel to be adjusted.
Specifically, each parameter in the submodel to be adjusted is adjusted, and a new oversea credit card pneumatic control model formed after adjustment is tested, so that the new oversea credit card pneumatic control model can detect that the user with the user attribute is a risk client, and the user is refused to make a consumption decision.
In an embodiment, as shown in fig. 5, in step S42, performing parameter adjustment on the submodel to be adjusted specifically includes the following steps:
s421: and acquiring the corresponding rule weight and the rule threshold value from the submodel to be adjusted.
In this embodiment, the rule weight is a ratio of an influence of a decision rule set in the submodel to be adjusted on a final output decision in the oversea credit card pneumatic control model. The rule threshold is information of a threshold value adopted when calculating the characteristics of the data of the attributes of the user in the submodel to be adjusted.
Specifically, the corresponding rule weight and the rule threshold are obtained from the submodel to be adjusted before adjustment.
S422: and adjusting the rule weight and the rule threshold according to the judgment result.
Specifically, based on the machine learning means, the rule weight and the rule threshold value in each submodel to be adjusted are adjusted according to the abnormal data appearing in the judgment result, the overseas credit card wind control model after the submodel to be adjusted is tested, and the rule weight and the rule threshold value in the submodel to be adjusted can be further adjusted according to the test result. The update cycle of the overseas credit card pneumatic control model can be shortened by adjusting the rule weights and the rule threshold values by means of machine learning, and in the embodiment, the update cycle of 10 weeks can be shortened to 2 weeks.
It should be understood that, the sequence numbers of the steps in the foregoing embodiments do not imply an execution sequence, and the execution sequence of each process should be determined by its function and inherent logic, and should not constitute any limitation to the implementation process of the embodiments of the present invention.
Example two:
in one embodiment, an overseas credit card wind control model iteration device is provided, and the overseas credit card wind control model iteration device corresponds to the overseas credit card wind control model iteration method in the above embodiment one to one. As shown in fig. 6, the overseas credit card pneumatic control model iteration device comprises a sub-model acquisition module 10, an identification feature acquisition module 20, a judgment module 30 and an adjusting and optimizing module 40. The functional modules are explained in detail as follows:
the sub-model acquisition module 10 is used for acquiring a preset overseas credit card wind control model and acquiring sub-model data from the overseas credit card wind control model;
the identification feature obtaining module 20 is configured to extract features to be identified from each sub-model data;
the judging module 30 is used for acquiring real-time transaction data according to the to-be-identified characteristics of the overseas credit card wind control model, and performing effect judgment on the real-time transaction data to obtain a judgment result;
and the tuning module 40 is used for updating the sub-model data in the overseas credit card wind control model by adopting machine learning according to the judgment result.
Preferably, the sub-model acquisition module 10 comprises:
the historical data acquisition submodule 11 is configured to acquire historical user payment data, where the historical user payment data includes device and network data, social data, user behavior data, and service data;
the data preprocessing submodule 12 is used for preprocessing the historical payment data to obtain data to be subjected to feature extraction;
the feature extraction submodule 13 is configured to perform feature extraction on data to be feature extracted through regularization to obtain a data set to be trained;
and the model training submodule 14 is used for respectively modeling and training a device and a data set to be trained corresponding to the network data, the social data, the user behavior data and the business data, so as to obtain a credit card wind control model for detecting the risk level of the client.
Preferably, the judging module 30 includes:
the test data acquisition submodule 31 is used for acquiring model test data according to the overseas credit card wind control model;
and the judgment submodule 32 is configured to compare the real-time transaction data with the model test data to obtain a judgment result.
Preferably, the tuning module 40 includes:
a submodel acquiring submodule 41 to be adjusted, configured to acquire abnormal data from the determination result, and acquire corresponding submodel data according to the abnormal data, where the corresponding submodel data is used as a submodel to be adjusted;
and the parameter adjusting and optimizing module 42 is used for adjusting the parameters of the submodel to be adjusted.
Preferably, the parameter tuning sub-module 42 includes:
a parameter obtaining unit 421, configured to obtain a rule weight and a rule threshold from the submodel to be adjusted;
and an iteration unit 422, configured to adjust the rule weight and the rule threshold according to the determination result.
For specific limitations of the overseas credit card pneumatic control model iteration device, reference may be made to the above limitations of the overseas credit card pneumatic control model iteration method, which are not described herein again. The modules in the above-mentioned overseas credit card pneumatic control model iteration device can be wholly or partially realized by software, hardware and a combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.
Example three:
in one embodiment, a computer device is provided, which may be a server, the internal structure of which may be as shown in fig. 7. The computer device includes a processor, a memory, a network interface, and a database connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The database of the computer device is used to store historical payment data. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program when executed by a processor implements an overseas credit card pneumatic control model iterative method.
In one embodiment, a computer device is provided, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, the processor implementing the following steps when executing the computer program:
s10: acquiring a preset overseas credit card wind control model, and acquiring sub-model data from the overseas credit card wind control model;
s20: extracting features to be identified from each sub-model data;
s30: acquiring real-time transaction data according to the to-be-identified characteristics of the overseas credit card wind control model, and performing effect judgment on the real-time transaction data to obtain a judgment result;
s40: and updating the sub-model data in the overseas credit card wind control model by adopting machine learning according to the judgment result.
Example four:
in one embodiment, a computer-readable storage medium is provided, having a computer program stored thereon, which when executed by a processor, performs the steps of:
s10: acquiring a preset overseas credit card wind control model, and acquiring sub-model data from the overseas credit card wind control model;
s20: extracting features to be identified from each sub-model data;
s30: acquiring real-time transaction data according to the to-be-identified characteristics of the overseas credit card wind control model, and performing effect judgment on the real-time transaction data to obtain a judgment result;
s40: and updating the sub-model data in the overseas credit card wind control model by adopting machine learning according to the judgment result.
It will be understood by those of ordinary skill in the art that all or a portion of the processes of the methods of the embodiments described above may be implemented by a computer program that may be stored on a non-volatile computer-readable storage medium, which when executed, may include the processes of the embodiments of the methods described above, wherein any reference to memory, storage, database, or other medium used in the embodiments provided herein may include non-volatile and/or volatile memory.
It will be apparent to those skilled in the art that, for convenience and brevity of description, only the above-mentioned division of the functional units and modules is illustrated, and in practical applications, the above-mentioned function distribution may be performed by different functional units and modules according to needs, that is, the internal structure of the apparatus is divided into different functional units or modules to perform all or part of the above-mentioned functions.
The above-mentioned embodiments are only used for illustrating the technical solutions of the present invention, and not for limiting the same; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; such modifications and substitutions do not substantially depart from the spirit and scope of the embodiments of the present invention, and are intended to be included within the scope of the present invention.
Claims (10)
1. An overseas credit card wind control model iteration method is characterized by comprising the following steps:
s10: acquiring a preset overseas credit card wind control model, and acquiring sub-model data from the overseas credit card wind control model;
s20: extracting features to be identified from each sub-model data;
s30: acquiring real-time transaction data according to the to-be-identified characteristics of the overseas credit card wind control model, and performing effect judgment on the real-time transaction data to obtain a judgment result;
s40: and updating the sub-model data in the overseas credit card wind control model by adopting machine learning according to the judgment result.
2. The iterative method of the oversea credit card pneumatic control model of claim 1, wherein the step S10 comprises:
s11: obtaining historical user payment data, wherein the historical user payment data comprises equipment and network data, social data, user behavior data and business data;
s12: preprocessing the historical payment data to obtain data to be subjected to feature extraction;
s13: performing feature extraction on the data to be subjected to feature extraction through regularization to obtain a data set to be trained;
s14: and respectively modeling the equipment and the data sets to be trained corresponding to the network data, the social data, the user behavior data and the business data, and then training to obtain the credit card wind control model for detecting the risk level of the client.
3. The iterative method of the oversea credit card pneumatic control model of claim 1, wherein the step S30 comprises:
s31: obtaining model test data according to the overseas credit card wind control model;
s32: and comparing the real-time transaction data with the model test data to obtain the judgment result.
4. The iterative method of the oversea credit card pneumatic control model of claim 1, wherein the step S40 comprises:
s41: obtaining abnormal data from the judgment result, and obtaining corresponding sub-model data according to the abnormal data to serve as a sub-model to be adjusted;
s42: and adjusting parameters of the submodel to be adjusted.
5. The iterative method of the oversea credit card pneumatic control model of claim 1, wherein the step S42 comprises:
s421: acquiring corresponding rule weight and rule threshold from the submodel to be adjusted;
s422: and adjusting the rule weight and the rule threshold according to the judgment result.
6. An overseas credit card wind control model iteration device, characterized in that the overseas credit card wind control model iteration device comprises:
the system comprises a sub-model acquisition module, a sub-model acquisition module and a sub-model management module, wherein the sub-model acquisition module is used for acquiring a preset overseas credit card wind control model and acquiring sub-model data from the overseas credit card wind control model;
the identification feature acquisition module is used for extracting features to be identified from each sub-model data;
the judging module is used for acquiring real-time transaction data according to the to-be-identified characteristics of the overseas credit card wind control model, and performing effect judgment on the real-time transaction data to obtain a judgment result;
and the tuning module is used for updating the sub-model data in the overseas credit card wind control model by adopting machine learning according to the judgment result.
7. The oversea credit card pneumatic control model iteration device of claim 6, wherein the sub-model acquisition module comprises:
the historical data acquisition submodule is used for acquiring historical user payment data, and the historical user payment data comprises equipment and network data, social data, user behavior data and business data;
the data preprocessing submodule is used for preprocessing the historical payment data to obtain data to be subjected to feature extraction;
the characteristic extraction submodule is used for extracting the characteristics of the data to be subjected to characteristic extraction through regularization to obtain a data set to be trained;
and the model training submodule is used for respectively carrying out training after modeling the equipment and the data set to be trained corresponding to the network data, the social data, the user behavior data and the business data to obtain the credit card wind control model for detecting the risk level of the client.
8. The iterative means of the overseas credit card pneumatic control model of claim 6, wherein the means for determining comprises:
the test data acquisition sub-module is used for acquiring model test data according to the overseas credit card wind control model;
and the judging submodule is used for comparing the real-time transaction data with the model test data to obtain the judging result.
9. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that the processor, when executing the computer program, implements the steps of the overseas credit card pneumatic control model iterative method of any of claims 1 to 5.
10. A computer readable storage medium storing a computer program, wherein the computer program when executed by a processor implements the steps of the oversea credit card pneumatic control model iterative method of any of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010322281.8A CN111507829A (en) | 2020-04-22 | 2020-04-22 | Overseas credit card wind control model iteration method, device, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010322281.8A CN111507829A (en) | 2020-04-22 | 2020-04-22 | Overseas credit card wind control model iteration method, device, equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111507829A true CN111507829A (en) | 2020-08-07 |
Family
ID=71876314
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010322281.8A Pending CN111507829A (en) | 2020-04-22 | 2020-04-22 | Overseas credit card wind control model iteration method, device, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111507829A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112017029A (en) * | 2020-08-31 | 2020-12-01 | 中国银行股份有限公司 | Information prompting method and device |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107067324A (en) * | 2017-04-18 | 2017-08-18 | 上海翼翎数据信息技术有限公司 | A kind of utilization network packet capturing data realize the method and system of transaction risk control |
| CN107886431A (en) * | 2017-10-18 | 2018-04-06 | 上海瀚银信息技术有限公司 | Financial air control system based on big data and artificial intelligence |
| CN108154430A (en) * | 2017-12-28 | 2018-06-12 | 上海氪信信息技术有限公司 | A kind of credit scoring construction method based on machine learning and big data technology |
| CN108346048A (en) * | 2017-01-23 | 2018-07-31 | 阿里巴巴集团控股有限公司 | A kind of method, Risk Identification Method and the device of adjustment risk parameter |
| CN109389486A (en) * | 2018-08-27 | 2019-02-26 | 深圳壹账通智能科技有限公司 | Loan air control rule adjustment method, apparatus, equipment and computer storage medium |
| CN109767110A (en) * | 2019-01-04 | 2019-05-17 | 中国银行股份有限公司 | A kind of risk control system optimization method, device, equipment and storage medium |
| CN109840838A (en) * | 2018-12-26 | 2019-06-04 | 天翼电子商务有限公司 | Air control rule model system with double engines, control method and server |
| CN109840680A (en) * | 2018-12-19 | 2019-06-04 | 平安国际融资租赁有限公司 | Service request processing method, device, computer equipment and storage medium |
| CN110232612A (en) * | 2019-05-20 | 2019-09-13 | 深圳壹账通智能科技有限公司 | Product method for pushing, device, computer equipment and storage medium |
| CN110264340A (en) * | 2019-06-12 | 2019-09-20 | 重庆无界领智普惠商务信息咨询有限公司 | A kind of P2P net loan customers' credit methods of marking and system based on machine learning |
| CN110599004A (en) * | 2019-08-23 | 2019-12-20 | 阿里巴巴集团控股有限公司 | Risk control method, equipment, medium and device |
| US20190392441A1 (en) * | 2018-06-25 | 2019-12-26 | Apple Inc. | Customizing authorization request schedules with machine learning models |
| CN110991650A (en) * | 2019-11-25 | 2020-04-10 | 第四范式（北京）技术有限公司 | Method and device for training card maintenance identification model and identifying card maintenance behavior |
-
2020
- 2020-04-22 CN CN202010322281.8A patent/CN111507829A/en active Pending
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108346048A (en) * | 2017-01-23 | 2018-07-31 | 阿里巴巴集团控股有限公司 | A kind of method, Risk Identification Method and the device of adjustment risk parameter |
| CN107067324A (en) * | 2017-04-18 | 2017-08-18 | 上海翼翎数据信息技术有限公司 | A kind of utilization network packet capturing data realize the method and system of transaction risk control |
| CN107886431A (en) * | 2017-10-18 | 2018-04-06 | 上海瀚银信息技术有限公司 | Financial air control system based on big data and artificial intelligence |
| CN108154430A (en) * | 2017-12-28 | 2018-06-12 | 上海氪信信息技术有限公司 | A kind of credit scoring construction method based on machine learning and big data technology |
| US20190392441A1 (en) * | 2018-06-25 | 2019-12-26 | Apple Inc. | Customizing authorization request schedules with machine learning models |
| CN109389486A (en) * | 2018-08-27 | 2019-02-26 | 深圳壹账通智能科技有限公司 | Loan air control rule adjustment method, apparatus, equipment and computer storage medium |
| CN109840680A (en) * | 2018-12-19 | 2019-06-04 | 平安国际融资租赁有限公司 | Service request processing method, device, computer equipment and storage medium |
| CN109840838A (en) * | 2018-12-26 | 2019-06-04 | 天翼电子商务有限公司 | Air control rule model system with double engines, control method and server |
| CN109767110A (en) * | 2019-01-04 | 2019-05-17 | 中国银行股份有限公司 | A kind of risk control system optimization method, device, equipment and storage medium |
| CN110232612A (en) * | 2019-05-20 | 2019-09-13 | 深圳壹账通智能科技有限公司 | Product method for pushing, device, computer equipment and storage medium |
| CN110264340A (en) * | 2019-06-12 | 2019-09-20 | 重庆无界领智普惠商务信息咨询有限公司 | A kind of P2P net loan customers' credit methods of marking and system based on machine learning |
| CN110599004A (en) * | 2019-08-23 | 2019-12-20 | 阿里巴巴集团控股有限公司 | Risk control method, equipment, medium and device |
| CN110991650A (en) * | 2019-11-25 | 2020-04-10 | 第四范式（北京）技术有限公司 | Method and device for training card maintenance identification model and identifying card maintenance behavior |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112017029A (en) * | 2020-08-31 | 2020-12-01 | 中国银行股份有限公司 | Information prompting method and device |
| CN112017029B (en) * | 2020-08-31 | 2023-09-08 | 中国银行股份有限公司 | Information prompting method and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108876133B (en) | Risk assessment processing method, device, server and medium based on business information | |
| CN112417439B (en) | Account detection method, device, server and storage medium | |
| CN109858737B (en) | Grading model adjustment method and device based on model deployment and computer equipment | |
| CN111369342A (en) | Loan approval method, device, equipment and storage medium based on machine learning | |
| CN110728323B (en) | Target type user identification method and device, electronic equipment and storage medium | |
| CN109816200B (en) | Task pushing method, device, computer equipment and storage medium | |
| CN112633962B (en) | Service recommendation method and device, computer equipment and storage medium | |
| CN112837069B (en) | Block chain and big data based secure payment method and cloud platform system | |
| CN109740869A (en) | Data checking method, device, computer equipment and storage medium | |
| CN109670931A (en) | Behavioral value method, apparatus, equipment and the storage medium of loan user | |
| CN112529575A (en) | Risk early warning method, equipment, storage medium and device | |
| CN111091408A (en) | User identification model creating method and device and identification method and device | |
| CN111641594B (en) | Method, system, medium and device for detecting fraudulent user based on page behavior | |
| CN111507829A (en) | Overseas credit card wind control model iteration method, device, equipment and storage medium | |
| CN117495544A (en) | Sandbox-based wind control evaluation method, sandbox-based wind control evaluation system, sandbox-based wind control evaluation terminal and storage medium | |
| CN115578045A (en) | Tender invitation auditing method, electronic equipment and related products | |
| CN107871213B (en) | Transaction behavior evaluation method, device, server and storage medium | |
| CN112685478B (en) | Information processing method for cloud service and user portrait mining and cloud server | |
| CN110570301B (en) | Risk identification method, device, equipment and medium | |
| CN109711984B (en) | Pre-loan risk monitoring method and device based on collection urging | |
| CN113487320A (en) | Fraud transaction detection method, device, computer equipment and storage medium | |
| CN111652712A (en) | Pre-credit analysis method, device, equipment and storage medium based on geographic information | |
| CN112085594B (en) | Identity verification method, device and readable storage medium | |
| CN111429270A (en) | Overseas credit card wind control model acquisition method, device, equipment and storage medium | |
| CN114581693B (en) | User behavior mode distinguishing method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200807 |
|
| RJ01 | Rejection of invention patent application after publication |