CN113159924A - Method and device for determining trusted client object - Google Patents
Method and device for determining trusted client object Download PDFInfo
- Publication number
- CN113159924A CN113159924A CN202110484551.XA CN202110484551A CN113159924A CN 113159924 A CN113159924 A CN 113159924A CN 202110484551 A CN202110484551 A CN 202110484551A CN 113159924 A CN113159924 A CN 113159924A
- Authority
- CN
- China
- Prior art keywords
- client object
- target
- client
- credit
- determining
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 78
- 238000012549 training Methods 0.000 claims abstract description 70
- 238000013475 authorization Methods 0.000 claims abstract description 9
- 238000012545 processing Methods 0.000 claims description 18
- 230000006870 function Effects 0.000 description 5
- 238000011835 investigation Methods 0.000 description 5
- 238000007781 pre-processing Methods 0.000 description 5
- 238000004422 calculation algorithm Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 238000010276 construction Methods 0.000 description 3
- 238000003860 storage Methods 0.000 description 3
- 238000012546 transfer Methods 0.000 description 3
- 230000006399 behavior Effects 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000007637 random forest analysis Methods 0.000 description 2
- 238000007619 statistical method Methods 0.000 description 2
- 238000009825 accumulation Methods 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000010219 correlation analysis Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- 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
- G06Q20/00—Payment architectures, schemes or protocols
- G06Q20/38—Payment protocols; Details thereof
- G06Q20/40—Authorisation, e.g. identification of payer or payee, verification of customer or shop credentials; Review and approval of payers, e.g. check credit lines or negative lists
- G06Q20/401—Transaction verification
- G06Q20/4016—Transaction verification involving fraud or risk level assessment in transaction processing
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Human Resources & Organizations (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Accounting & Taxation (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Economics (AREA)
- Development Economics (AREA)
- Data Mining & Analysis (AREA)
- Finance (AREA)
- Marketing (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Entrepreneurship & Innovation (AREA)
- Educational Administration (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Technology Law (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Game Theory and Decision Science (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Medical Informatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Computer Security & Cryptography (AREA)
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
Abstract
The invention provides a method and a device for determining a trusted client object, wherein the method comprises the following steps: acquiring each target client object; acquiring first characteristic information of each target client object; the first characteristic information of each target client object is obtained based on the service data of the target client object in a target time period; identifying first characteristic information of the target client object by applying a credit identification model to obtain an identification result of the target client object; the credit recognition model is obtained by training based on a training sample set, and the training sample set comprises all training samples; each training sample comprises second characteristic information and a label of a client object to which the training sample belongs, wherein the second characteristic information is obtained based on business data of the client object in an observation period, and the label is obtained based on an expression result of the client object in an expression period; and determining the credit client object in each target client object based on the identification result of each target client object. The method provided by the invention can improve the accuracy of the credit authorization of the client.
Description
Technical Field
The invention relates to the technical field of data processing, in particular to a method and a device for determining a trusted client object.
Background
With the development of computer technology, more and more technologies are applied in the financial field, and the traditional financial industry is gradually changing to financial technology (Fintech), but higher requirements are also put forward on the credit granting technology due to the security requirements of the financial industry.
The trust granting means that a financial institution directly provides fund support for a client meeting conditions or makes a guarantee action for the credit of the client in related economic activities to a third party, at present, the trust granting logic of the financial client is mostly based on rules, namely, the current credit granting logic is calculated according to rule logics set by some business experiences, and then the trusted client is determined.
Disclosure of Invention
The invention aims to provide a method for determining a credit client object, which can improve the accuracy of client credit and the security of credit.
The invention also provides a device for determining the trusted client object, which is used for ensuring the realization and the application of the method in practice.
A method for determining a trusted client object, comprising:
acquiring a set of client objects to be processed; the set of customer objects comprises target customer objects;
acquiring first characteristic information of each target client object; the first characteristic information of each target client object is obtained based on the service data of the target client object in a target time interval in a preset service time range;
applying a credit authorization identification model to identify the first characteristic information of each target customer object to obtain an identification result of each target customer object; the credit granting identification model is obtained by training based on a training sample set, and the training sample set comprises all training samples; each training sample comprises second characteristic information and a label of a client object to which the training sample belongs, wherein the second characteristic information is obtained based on business data of an observation period of the client object in the business time range, and the label is obtained based on a performance result of the client object in a performance period of the business time range;
and determining a credit client object in each target client object based on the identification result of each target client object.
Optionally, the method for acquiring a set of customer objects to be processed includes:
acquiring an initial client object list, wherein the initial client object list comprises a plurality of client objects;
determining a credit rating for each of the customer subjects;
determining the client object with the credit rating not greater than a preset rating threshold as a target client object;
and forming the target client objects into a client object set to be processed.
Optionally, the obtaining the first characteristic information of each target client object includes:
performing characteristic engineering processing on the service data of the target client object in a target time interval in a preset service time range to obtain first characteristic data of preset credit feature dimensions;
forming a first characteristic width table of the target customer object by using the first characteristic data of each credit granting characteristic dimension;
and taking the first characteristic width table as first characteristic information of the target client object.
In the foregoing method, optionally, the identification result represents a default probability of a target customer object to which the identification result belongs within a preset prediction period, where the prediction period is a period next to the target period;
the determining a credit customer object in each target customer object based on the identification result of each target customer object includes:
and determining the target customer object with the default probability smaller than a preset probability threshold value in the target customer objects as a credit customer object.
Optionally, the determining, based on the recognition result of each target client object, a trusted client object in each target client object includes:
determining alternative credit client objects in the target client objects based on the identification result of each target client object;
acquiring the customer information of each alternative credit customer object;
and determining the alternative credit granting client object with client information meeting the preset credit granting conditions as the credit granting client object.
The above method, optionally, further includes:
and forming a trusted client object list by the trusted client objects, and outputting the trusted client object list.
A trusted client object determining apparatus, comprising:
the system comprises a first acquisition unit, a second acquisition unit and a processing unit, wherein the first acquisition unit is used for acquiring a client object set to be processed; the set of customer objects comprises target customer objects;
a second acquisition unit configured to acquire first feature information of each of the target client objects; the first characteristic information of each target client object is obtained based on the service data of the target client object in a target time interval in a preset service time range;
the identification unit is used for identifying the first characteristic information of each target client object by applying a credit authorization identification model to obtain an identification result of each target client object; the credit granting identification model is obtained by training based on a training sample set, and the training sample set comprises all training samples; each training sample comprises second characteristic information and a label of a client object to which the training sample belongs, wherein the second characteristic information is obtained based on business data of an observation period of the client object in the business time range, and the label is obtained based on a performance result of the client object in a performance period of the business time range;
and the determining unit is used for determining the credit client object in each target client object based on the identification result of each target client object.
The above apparatus, optionally, the first obtaining unit includes:
a first obtaining subunit, configured to obtain an initial client object list, where the initial client object list includes a plurality of client objects;
the first determining subunit is used for determining the credit granting score of each client object;
the second determining subunit is used for determining the client object with the credit granting score not greater than a preset score threshold as a target client object;
and the first execution subunit is used for forming the target client objects into a client object set to be processed.
The above apparatus, optionally, the second obtaining unit includes:
the characteristic processing subunit is used for performing characteristic engineering processing on the service data of the target client object in a target time interval in a preset service time range to obtain first characteristic data of preset credit granting characteristic dimensions;
the second execution subunit is configured to compose the first feature data of each trust feature dimension into a first feature width table of the target client object;
and the third execution subunit is used for taking the first feature width table as the first feature information of the target client object.
The above apparatus, optionally, the determining unit includes:
a third determining subunit, configured to determine, based on an identification result of each target client object, an alternative trusted client object from among the target client objects;
the second acquisition subunit is used for acquiring the client information of each alternative credit client object;
and the fourth determining subunit is used for determining the alternative credit granting client object with the client information meeting the preset credit granting condition as the credit granting client object.
Compared with the prior art, the invention has the following advantages:
the invention provides a method and a device for determining a trusted client object, wherein the method comprises the following steps: acquiring a set of client objects to be processed; the set of customer objects comprises target customer objects; acquiring first characteristic information of each target client object; the first characteristic information of each target client object is obtained based on the service data of the target client object in a target time interval in a preset service time range; applying a credit authorization identification model to identify the first characteristic information of each target customer object to obtain an identification result of each target customer object; the credit granting identification model is obtained by training based on a training sample set, and the training sample set comprises all training samples; each training sample comprises second characteristic information and a label of a client object to which the training sample belongs, wherein the second characteristic information is obtained based on business data of an observation period of the client object in the business time range, and the label is obtained based on a performance result of the client object in a performance period of the business time range; and determining a credit client object in each target client object based on the identification result of each target client object. By applying the method for determining the credit granting customer object provided by the invention, the credit granting identification is carried out by applying the credit granting identification model, the accuracy rate of the client credit granting can be improved, the fraud risk caused by artificial rule leakage can be avoided by extracting the characteristics, and the security of the credit granting is improved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the provided drawings without creative efforts.
FIG. 1 is a flowchart of a method for determining a trusted client object according to the present invention;
FIG. 2 is a flow chart of a process for obtaining a set of customer objects to be processed according to the present invention;
FIG. 3 is a flowchart of a process for obtaining first characteristic information of each target client object according to the present invention;
FIG. 4 is a schematic structural diagram of a device for determining a trusted client object according to the present invention;
fig. 5 is a schematic structural diagram of an electronic device provided in the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
In this application, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
The embodiment of the invention provides a method for determining a trusted client object, which can be applied to electronic equipment, and the method flow chart of the method is shown in fig. 1, and specifically comprises the following steps:
s101: acquiring a set of client objects to be processed; the set of customer objects includes individual target customer objects.
In the method provided by the embodiment of the invention, the target customer object can be a customer object which does not meet the set initial credit granting rule, and the initial credit granting rule can be set according to income, financial assets and the initial credit granting amount of the customer object.
S102: acquiring first characteristic information of each target client object; the first characteristic information of each target client object is obtained based on the service data of the target client object in the target time period in the preset service time range.
In the method provided by the embodiment of the present invention, the service time range may include an observation period, a presentation period, and a target period, where the observation period may be a period before the presentation period, and the target period may be a period after the presentation period.
Optionally, the first feature information includes feature data of each preset credit granting feature dimension, the each preset credit granting feature dimension may include any combination of a plurality of items of a client age feature dimension, a transfer frequency feature dimension, a client consumption feature dimension, a repayment feature dimension, a transaction running feature dimension, an external data feature dimension, a risk portrait dimension, a bill dimension, an asset/debt dimension, an escort feature dimension, a credit granting feature dimension, and the like, and the first feature information may be in a form of a wide form.
The business data can include any combination of age, nationality, customer consumption data, repayment data, transaction flow data, external data, risk representation data, bill data, asset liability data, security data, credit investigation data and the like.
Specifically, feature engineering processing can be performed on the service data of the client object in the target time period to obtain first feature data of each preset credit granting feature dimension.
S103: applying a credit authorization identification model to identify the first characteristic information of each target customer object to obtain an identification result of each target customer object; the credit granting identification model is obtained by training based on a training sample set, and the training sample set comprises all training samples; each training sample comprises second characteristic information of a client object to which the training sample belongs and a label, wherein the second characteristic information is obtained based on business data of an observation period of the client object in the business time range, and the label is obtained based on a performance result of the client object in a performance period of the business time range.
Specifically, in a feasible manner, the identification result may represent a default probability of the target customer object to which the identification result belongs within a preset prediction period, where the prediction period may be a period next to the target period; in another possible approach, the identification result may characterize whether the target customer object to which it belongs violates the forecast period.
The second characteristic information comprises second characteristic data of each preset credit granting characteristic dimension, characteristic engineering processing can be performed on business data of the client object in the observation period, the second characteristic data of each preset credit granting characteristic dimension is obtained, and the second characteristic information can be in a form of a wide table.
Optionally, the training process of the trust identification model may be: the method comprises the steps of obtaining a training sample set and an initial credit granting identification model, training the initial credit granting identification model by applying each training sample in the training sample set until the initial credit granting identification model meets preset training completion conditions, determining the trained initial credit granting identification model as the credit granting identification model, wherein the initial credit granting identification model can be a machine learning model constructed according to algorithms such as random forest, GBDT or XGBOOST, and the like, and the training conditions can be that model scores of the initial credit granting identification model meet preset scoring conditions or training times of the initial credit granting identification model meet preset training time threshold values.
S104: and determining a credit client object in each target client object based on the identification result of each target client object.
In the method provided by the embodiment of the invention, after the credit client object is determined, the credit client object can be added to the preset credit white list, and further the business service corresponding to the credit white list can be provided for the credit client object in the credit white list.
By applying the method for determining the credit granting customer object provided by the invention, the credit granting identification is carried out by applying the credit granting identification model, the accuracy rate of the client credit granting can be improved, the fraud risk caused by artificial rule leakage can be avoided by extracting the characteristics, and the security of the credit granting is improved.
In the method provided in the embodiment of the present invention, based on the foregoing implementation process, specifically, the target client object may be a client object whose trust score is not greater than a set score threshold, where a feasible manner of acquiring a set of client objects to be processed is shown in fig. 2, and includes:
s201: an initial client object list is obtained, the initial client object list comprising a plurality of client objects.
S202: determining a credit rating for each of the customer subjects.
The credit rating can be calculated according to the income of the client, the financial assets, the initial credit line and other information.
Optionally, the income score of the client object may be calculated according to the income of the client object, the asset score of the client object may be calculated according to the financial asset of the client object, the initial credit line score of the client object may be calculated according to the initial credit line, and the income score, the asset score and the initial credit line are subjected to weighted summation to obtain the credit score of the client object.
S203: and determining the client object with the credit rating not greater than a preset rating threshold as a target client object.
The scoring threshold value can be set according to actual requirements.
S204: and forming the target client objects into a client object set to be processed.
By applying the method provided by the embodiment of the invention, the clients with stable consumption and good credit can be excavated from the client objects with poor credit scores for credit, so that the ceiling of the existing client group is broken, and the expansion of the client group is completed.
In the method provided in the embodiment of the present invention, based on the implementation process, specifically, the process of acquiring the first feature information of each target client object, as shown in fig. 3, includes:
s301: and performing characteristic engineering processing on the service data of the target client object in a target time interval in a preset service time range to obtain first characteristic data of preset credit feature dimensions.
In the method provided by the embodiment of the invention, characteristic engineering processing can be carried out on the client information data, the client consumption data, the payment data, the transaction flow data, the external data, the card opening data, the security data, the credit investigation data and the like of the target client object in the target time period, and first characteristic data such as age characteristic, consumption characteristic, transfer frequency characteristic, card opening frequency characteristic, credit investigation grade characteristic and the like can be obtained.
S302: and forming a first characteristic width table of the target customer object by using the first characteristic data of each credit granting characteristic dimension.
In the method provided by the embodiment of the present invention, the first characteristic width table may be a large table in which the client number is used as a main key and the client characteristics are displayed in multiple columns.
S303: and taking the first characteristic width table as first characteristic information of the target client object.
By applying the method provided by the embodiment of the invention, the first characteristic information of the target client object is obtained by performing characteristic engineering processing on the service data of the target client object in the target time interval, and then whether the target client object is a credit granting client object can be determined according to various characteristics of the target client object, so that the accuracy of credit granting identification is improved.
In the method provided by the embodiment of the present invention, based on the implementation process, specifically, the identification result represents a default probability of a target customer object to which the identification result belongs within a preset prediction period, where the prediction period is a period next to the target period;
accordingly, one possible way of determining the trusted client object in each target client object based on the identification result of each target client object in S104 is as follows:
and determining the target customer object with the default probability smaller than a preset probability threshold value in the target customer objects as a credit customer object.
In the method provided by the embodiment of the invention, the default probability of each target customer object in the prediction period can be compared with a preset default probability threshold, and if the default probability of the target customer object is smaller than the probability threshold, the target customer object can be determined as a credit customer object; if the default probability of the target customer object is greater than or equal to the probability threshold, risk marking may be performed on the target customer object.
By applying the method provided by the embodiment of the invention, the client object with lower default probability can be selected from all target client objects as the credit client object.
In the method provided in the embodiment of the present invention, based on the implementation process, specifically, based on the recognition result of each target client object in S104, another feasible way of determining the trusted client object in each target client object is as follows:
determining alternative credit client objects in the target client objects based on the identification result of each target client object;
acquiring the customer information of each alternative credit customer object;
and determining the alternative credit granting client object with client information meeting the preset credit granting conditions as the credit granting client object.
In the method provided by the embodiment of the present invention, the identification result may represent a default probability of the target customer object to which the identification result belongs in the prediction period, and then the target customer object whose default probability is smaller than a preset threshold may be determined as the alternative trusted customer object, or, in a case that the identification result represents whether the target customer object to which the identification result belongs in the prediction period is default, the target customer object which is not default in the prediction period may be determined as the alternative trusted customer object.
The client information may include information such as the age and nationality of the user, and the alternative trusted client object may be determined as the trusted client object when the alternative trusted client object satisfies the trust condition.
By applying the method provided by the embodiment of the invention, the credit granting condition can be flexibly added according to the business requirement or the regulation supervision requirement and the like, so that the credit granting client object is determined.
In the method provided in the embodiment of the present invention, based on the implementation process, specifically, the method further includes:
and forming a trusted client object list by the trusted client objects, and outputting the trusted client object list.
The trusted client object list can be displayed on a preset display interface, and the trusted client objects in the trusted client object list can be added to a set trusted white list.
Optionally, the trusted object list may be stored, and may also be transmitted.
By applying the method provided by the embodiment of the invention, the trusted client object list can be obtained, and a client data basis can be provided for business service.
Corresponding to the method described in fig. 1, an embodiment of the present invention further provides a device for determining a trusted client object, which is used to implement the method in fig. 1 specifically, and the device for determining a trusted client object provided in the embodiment of the present invention may be applied to an electronic device, and a schematic structural diagram of the device is shown in fig. 4, and specifically includes:
a first obtaining unit 401, configured to obtain a set of client objects to be processed; the set of customer objects comprises target customer objects;
a second obtaining unit 402, configured to obtain first feature information of each of the target client objects; the first characteristic information of each target client object is obtained based on the service data of the target client object in a target time interval in a preset service time range;
the identifying unit 403 is configured to identify, by using a trust identification model, first feature information of each target client object to obtain an identification result of each target client object; the credit granting identification model is obtained by training based on a training sample set, and the training sample set comprises all training samples; each training sample comprises second characteristic information and a label of a client object to which the training sample belongs, wherein the second characteristic information is obtained based on business data of an observation period of the client object in the business time range, and the label is obtained based on a performance result of the client object in a performance period of the business time range;
a determining unit 404, configured to determine, based on the recognition result of each target client object, a trusted client object among the target client objects.
In an embodiment provided by the present invention, based on the above scheme, specifically, the first obtaining unit 401 includes:
a first obtaining subunit, configured to obtain an initial client object list, where the initial client object list includes a plurality of client objects;
the first determining subunit is used for determining the credit granting score of each client object;
the second determining subunit is used for determining the client object with the credit granting score not greater than a preset score threshold as a target client object;
and the first execution subunit is used for forming the target client objects into a client object set to be processed.
In an embodiment of the present invention, based on the above scheme, specifically, the second obtaining unit 402 includes:
the characteristic processing subunit is used for performing characteristic engineering processing on the service data of the target client object in a target time interval in a preset service time range to obtain first characteristic data of preset credit granting characteristic dimensions;
the second execution subunit is configured to compose the first feature data of each trust feature dimension into a first feature width table of the target client object;
and the third execution subunit is used for taking the first feature width table as the first feature information of the target client object.
In an embodiment provided by the present invention, based on the above scheme, specifically, the determining unit 404 includes:
a third determining subunit, configured to determine, based on an identification result of each target client object, an alternative trusted client object from among the target client objects;
the second acquisition subunit is used for acquiring the client information of each alternative credit client object;
and the fourth confirming subunit is used for confirming the alternative credit granting client object with the client information meeting the preset credit granting condition as the credit granting client object.
The specific principle and the implementation process of each unit and module in the apparatus for determining a trusted client object disclosed in the embodiment of the present invention are the same as the method for determining a trusted client object disclosed in the embodiment of the present invention, and reference may be made to the corresponding parts in the method for determining a trusted client object provided in the embodiment of the present invention, which are not described herein again.
The embodiment of the invention also provides a storage medium, which comprises a stored instruction, wherein when the instruction runs, the device where the storage medium is located is controlled to execute the method for determining the trusted client object.
An electronic device is provided in an embodiment of the present invention, and the structural diagram of the electronic device is shown in fig. 5, which specifically includes a memory 501 and one or more instructions 502, where the one or more instructions 502 are stored in the memory 501, and are configured to be executed by one or more processors 503 to perform the following operations according to the one or more instructions 502:
acquiring a set of client objects to be processed; the set of customer objects comprises target customer objects;
acquiring first characteristic information of each target client object; the first characteristic information of each target client object is obtained based on the service data of the target client object in a target time interval in a preset service time range;
applying a credit authorization identification model to identify the first characteristic information of each target customer object to obtain an identification result of each target customer object; the credit granting identification model is obtained by training based on a training sample set, and the training sample set comprises all training samples; each training sample comprises second characteristic information and a label of a client object to which the training sample belongs, wherein the second characteristic information is obtained based on business data of an observation period of the client object in the business time range, and the label is obtained based on a performance result of the client object in a performance period of the business time range;
and determining a credit client object in each target client object based on the identification result of each target client object.
Based on the above implementation scheme, taking the determined scenario of the credit client object of the commercial bank as an example, the technical scheme of the present application is exemplified:
firstly, most of the current bank credit is for white list customers of the bank, such as a proxy payroll customer, a social security public accumulation fund customer, or a high-equity customer evaluated according to properties such as houses and deposits, and such a customer group is a high-quality customer group and has low loan risk. However, the limitation of the customer group is caused, along with the great popularization of consumption finance and general finance, how to more accurately evaluate customers and how to mine a batch of customers with stable consumption and good credit from stock customers of banks for credit service on the premise of meeting the wind control requirement, so that the ceiling of the existing customer group is broken, and the problem of expanding the customer group is urgently needed to be solved. Secondly, at present, the commercial bank mostly considers the factors with strong financial attributes, such as income, financial assets, existing credit and the like, to the credit of the client, multi-dimensional data such as client consumption, payment, transaction running, external data, security, credit investigation and the like are not fully involved, the user portrait can not be comprehensively described, and the credit service client group can not be effectively expanded. Moreover, at present, the credit logic of the commercial bank to the client is mostly calculated based on rules, namely according to the rule logic set by some business experiences, so that the human intervention is obvious, on one hand, the human intervention is not accurate, on the other hand, the fraud risk is easily identified by a fraud group.
In order to solve the above problems, in a specific implementation, the technical scheme of the application can be functionally divided into the following modules, namely a data preprocessing module, a feature engineering module, a model training module and a trust client release module;
the data preprocessing module mainly realizes the functions of data source loading, data exploration and data preprocessing.
The characteristic engineering module mainly realizes the construction of characteristic engineering, carries out statistical analysis on the characteristics, determines the label and forms a wide table.
A model training module: based on the broad table and the label, algorithm selection, model training and tuning are performed, and the overdue probability of the inventory client is output.
1. A data preprocessing module: the module mainly realizes the functions of data source loading, data exploration and data preprocessing.
The data sources include multi-dimensional data including multiple dimensions such as debit card transactions, transfers, credit card consumption, payments, bills, risk representation data, assets and liabilities, collateral data, credit investigation, and external data (e.g., blacklists, credit learning network data).
And data exploration is carried out after data are loaded, and data distribution condition, missing value statistics, correlation analysis and the like are included.
2. A characteristic engineering module: the module mainly realizes the construction of feature engineering, performs statistical analysis on features, determines labels and forms a broad table.
The construction of the characteristic function is the basis of modeling, the module can screen data with the time span of 6 months to process the characteristics, and whether overdue behaviors exist in 3 months after selection to mark or not is judged, namely, the label is designed.
Assuming that 12 months in 2020 is the present time, the business goal is to push out a part of the list of trusted customers for marketing, i.e. it needs to predict whether the customer will expire in 1 month and later in 2021 year; in this case, the observation period can be selected from 10 months in 2019 to 3 months in 2020 (credit data, x data); 4/2020-6/2020 (credit risk performance data, y data) as the performance period; and performing characteristic engineering processing on the observation period to form a wide table, counting overdue behaviors of customers in the expression period, and marking to form a label.
And a characteristic broad table with the time span of 7 months in 2020 to 12 months in 2020 needs to be constructed, the broad table is an input of a subsequent production credit list, and the characteristic engineering of the latest time span needs to be constructed because the overdue risk of a future period of time (1 month in 2021 to 4 months in 2021) at the current time point (12 months in 2020) is predicted.
3. A model training module: machine learning model training is carried out based on the wide list of the observation period (10 months-3 months in 2019) and the label of the presentation period (4 months-6 months in 2020), and the model algorithm can be selected from algorithms such as random forest, GBDT, XGBOST and the like to construct a model.
After the model training is finished, inputting a broad table of 7-2020-12-2020-1 of feature engineering processing into the model, and predicting and outputting the probability of overdue of 4-2021 of a client.
4. The credit client authorization module: because the model training module outputs the probability of future overdue of the client, the standard of the client can be flexibly planned according to the bearing capacity of the wind control risk of the client in business. Furthermore, the service can be combined with its own requirements to define rule screening, such as requiring the client's nationality to be China, age 25-50, etc. And finally outputting a trusted client list.
It should be noted that, in the present specification, the embodiments are all described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments may be referred to each other. For the device-like embodiment, since it is basically similar to the method embodiment, the description is simple, and for the relevant points, reference may be made to the partial description of the method embodiment.
Finally, it should also be noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
For convenience of description, the above devices are described as being divided into various units by function, and are described separately. Of course, the functions of the units may be implemented in the same software and/or hardware or in a plurality of software and/or hardware when implementing the invention.
From the above description of the embodiments, it is clear to those skilled in the art that the present invention can be implemented by software plus necessary general hardware platform. Based on such understanding, the technical solutions of the present invention may be embodied in the form of a software product, which may be stored in a storage medium, such as ROM/RAM, magnetic disk, optical disk, etc., and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method according to the embodiments or some parts of the embodiments.
The method for determining the trusted client object provided by the present invention is described in detail above, and the principle and the implementation of the present invention are explained in detail herein by applying specific examples, and the description of the above examples is only used to help understanding the method of the present invention and the core idea thereof; meanwhile, for a person skilled in the art, according to the idea of the present invention, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present invention.
Claims (10)
1. A method for determining a trusted client object, comprising:
acquiring a set of client objects to be processed; the set of customer objects comprises target customer objects;
acquiring first characteristic information of each target client object; the first characteristic information of each target client object is obtained based on the service data of the target client object in a target time interval in a preset service time range;
applying a credit authorization identification model to identify the first characteristic information of each target customer object to obtain an identification result of each target customer object; the credit granting identification model is obtained by training based on a training sample set, and the training sample set comprises all training samples; each training sample comprises second characteristic information and a label of a client object to which the training sample belongs, wherein the second characteristic information is obtained based on business data of an observation period of the client object in the business time range, and the label is obtained based on a performance result of the client object in a performance period of the business time range;
and determining a credit client object in each target client object based on the identification result of each target client object.
2. The method of claim 1, wherein obtaining the set of customer objects to be processed comprises:
acquiring an initial client object list, wherein the initial client object list comprises a plurality of client objects;
determining a credit rating for each of the customer subjects;
determining the client object with the credit rating not greater than a preset rating threshold as a target client object;
and forming the target client objects into a client object set to be processed.
3. The method of claim 1, wherein the obtaining first characteristic information of each of the target client objects comprises:
performing characteristic engineering processing on the service data of the target client object in a target time interval in a preset service time range to obtain first characteristic data of preset credit feature dimensions;
forming a first characteristic width table of the target customer object by using the first characteristic data of each credit granting characteristic dimension;
and taking the first characteristic width table as first characteristic information of the target client object.
4. The method according to claim 1, wherein the identification result represents a default probability of a target customer object to which the identification result belongs within a preset prediction period, and the prediction period is a period next to the target period;
the determining a credit customer object in each target customer object based on the identification result of each target customer object includes:
and determining the target customer object with the default probability smaller than a preset probability threshold value in the target customer objects as a credit customer object.
5. The method of claim 1, wherein determining a trusted client object among the target client objects based on the identification of each target client object comprises:
determining alternative credit client objects in the target client objects based on the identification result of each target client object;
acquiring the customer information of each alternative credit customer object;
and determining the alternative credit granting client object with client information meeting the preset credit granting conditions as the credit granting client object.
6. The method of claim 1, further comprising:
and forming a trusted client object list by the trusted client objects, and outputting the trusted client object list.
7. An apparatus for determining a trusted client object, comprising:
the system comprises a first acquisition unit, a second acquisition unit and a processing unit, wherein the first acquisition unit is used for acquiring a client object set to be processed; the set of customer objects comprises target customer objects;
a second acquisition unit configured to acquire first feature information of each of the target client objects; the first characteristic information of each target client object is obtained based on the service data of the target client object in a target time interval in a preset service time range;
the identification unit is used for identifying the first characteristic information of each target client object by applying a credit authorization identification model to obtain an identification result of each target client object; the credit granting identification model is obtained by training based on a training sample set, and the training sample set comprises all training samples; each training sample comprises second characteristic information and a label of a client object to which the training sample belongs, wherein the second characteristic information is obtained based on business data of an observation period of the client object in the business time range, and the label is obtained based on a performance result of the client object in a performance period of the business time range;
and the determining unit is used for determining the credit client object in each target client object based on the identification result of each target client object.
8. The apparatus of claim 7, wherein the first obtaining unit comprises:
a first obtaining subunit, configured to obtain an initial client object list, where the initial client object list includes a plurality of client objects;
the first determining subunit is used for determining the credit granting score of each client object;
the second determining subunit is used for determining the client object with the credit granting score not greater than a preset score threshold as a target client object;
and the first execution subunit is used for forming the target client objects into a client object set to be processed.
9. The apparatus of claim 7, wherein the second obtaining unit comprises:
the characteristic processing subunit is used for performing characteristic engineering processing on the service data of the target client object in a target time interval in a preset service time range to obtain first characteristic data of preset credit granting characteristic dimensions;
the second execution subunit is configured to compose the first feature data of each trust feature dimension into a first feature width table of the target client object;
and the third execution subunit is used for taking the first feature width table as the first feature information of the target client object.
10. The apparatus of claim 7, wherein the determining unit comprises:
a third determining subunit, configured to determine, based on an identification result of each target client object, an alternative trusted client object from among the target client objects;
the second acquisition subunit is used for acquiring the client information of each alternative credit client object;
and the fourth determining subunit is used for determining the alternative credit granting client object with the client information meeting the preset credit granting condition as the credit granting client object.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110484551.XA CN113159924A (en) | 2021-04-30 | 2021-04-30 | Method and device for determining trusted client object |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110484551.XA CN113159924A (en) | 2021-04-30 | 2021-04-30 | Method and device for determining trusted client object |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113159924A true CN113159924A (en) | 2021-07-23 |
Family
ID=76872994
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110484551.XA Pending CN113159924A (en) | 2021-04-30 | 2021-04-30 | Method and device for determining trusted client object |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113159924A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117709906A (en) * | 2024-02-04 | 2024-03-15 | 杭银消费金融股份有限公司 | External data source query decision method and device |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20180232805A1 (en) * | 2016-06-12 | 2018-08-16 | Tencent Technology (Shenzhen) Company Limited | User credit rating method and apparatus, and storage medium |
| CN111861729A (en) * | 2020-07-31 | 2020-10-30 | 重庆富民银行股份有限公司 | Behavior scoring system and method based on lstm |
| CN112561568A (en) * | 2020-12-04 | 2021-03-26 | 中国农业银行股份有限公司深圳市分行 | Target customer prediction method, device and storage medium |
| CN112598500A (en) * | 2020-12-21 | 2021-04-02 | 中国建设银行股份有限公司 | Credit processing method and system for non-limit client |
-
2021
- 2021-04-30 CN CN202110484551.XA patent/CN113159924A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20180232805A1 (en) * | 2016-06-12 | 2018-08-16 | Tencent Technology (Shenzhen) Company Limited | User credit rating method and apparatus, and storage medium |
| CN111861729A (en) * | 2020-07-31 | 2020-10-30 | 重庆富民银行股份有限公司 | Behavior scoring system and method based on lstm |
| CN112561568A (en) * | 2020-12-04 | 2021-03-26 | 中国农业银行股份有限公司深圳市分行 | Target customer prediction method, device and storage medium |
| CN112598500A (en) * | 2020-12-21 | 2021-04-02 | 中国建设银行股份有限公司 | Credit processing method and system for non-limit client |
Non-Patent Citations (1)
| Title |
|---|
| 蔡主希: "《智能风控与反欺诈 体系算法与实践》", 31 March 2021, 机械工业出版社, pages: 139 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117709906A (en) * | 2024-02-04 | 2024-03-15 | 杭银消费金融股份有限公司 | External data source query decision method and device |
| CN117709906B (en) * | 2024-02-04 | 2024-05-14 | 杭银消费金融股份有限公司 | External data source query decision method and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Wall | Some financial regulatory implications of artificial intelligence | |
| CN113034046A (en) | Data risk metering method and device, electronic equipment and storage medium | |
| CN112232950A (en) | Loan risk assessment method and device, equipment and computer-readable storage medium | |
| CN116563006A (en) | Service risk early warning method, device, storage medium and device | |
| CN113469818B (en) | Investment risk early warning method and device, electronic equipment and computer readable medium | |
| Kim et al. | Random effects model for credit rating transitions | |
| Zarutska | Structural-functional analysis of the Ukraine banking system | |
| Mustafin et al. | Evaluation of the choice of borrower rating groups | |
| CN113159924A (en) | Method and device for determining trusted client object | |
| CN110619564B (en) | Anti-fraud feature generation method and device | |
| JP6771513B2 (en) | Devices and methods for calculating default probability and programs for it | |
| CN115994819A (en) | Risk customer identification method, apparatus, device and medium | |
| CA3169417A1 (en) | Method of and system for appraising risk | |
| CN113421155A (en) | Steel trade enterprise portrait multidimensional model construction method, wind control management method and device | |
| Prakash et al. | ATM Card Fraud Detection System Using Machine Learning Techniques | |
| Lee et al. | Application of machine learning in credit risk scorecard | |
| Kang | Fraud Detection in Mobile Money Transactions Using Machine Learning | |
| CN116644372B (en) | Account type determining method and device, electronic equipment and storage medium | |
| Sinha et al. | Stock Prices Forecasting using Time Series Analysis with Arima | |
| US20220261575A1 (en) | Prediction device, prediction method, prediction program | |
| KR102136313B1 (en) | Operating method of providing investment performance forecasting information based on artificial intelligence and server providing the same | |
| CN117541884A (en) | Sample data processing method, device, storage medium and system | |
| CN116797343A (en) | Risk assessment method, model training method, device, medium and equipment | |
| Sadoghi | Attention to bitcoin | |
| Li et al. | Analysis of Textual Risk Disclosures in Financial Statements |
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 |