CN112907356A - Overdue collection method, device and system and computer readable storage medium - Google Patents
Overdue collection method, device and system and computer readable storage medium Download PDFInfo
- Publication number
- CN112907356A CN112907356A CN202110243561.4A CN202110243561A CN112907356A CN 112907356 A CN112907356 A CN 112907356A CN 202110243561 A CN202110243561 A CN 202110243561A CN 112907356 A CN112907356 A CN 112907356A
- Authority
- CN
- China
- Prior art keywords
- overdue
- information
- preset
- migration
- risk
- 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 65
- 238000013508 migration Methods 0.000 claims abstract description 129
- 230000005012 migration Effects 0.000 claims abstract description 129
- 238000012545 processing Methods 0.000 claims abstract description 39
- 238000012549 training Methods 0.000 claims description 13
- 239000003054 catalyst Substances 0.000 claims description 11
- 238000011084 recovery Methods 0.000 claims description 11
- 230000010365 information processing Effects 0.000 claims description 9
- 230000003197 catalytic effect Effects 0.000 claims description 6
- 238000004140 cleaning Methods 0.000 claims description 6
- 238000007781 pre-processing Methods 0.000 claims description 5
- 238000012216 screening Methods 0.000 claims description 5
- 238000000926 separation method Methods 0.000 claims description 5
- 238000003306 harvesting Methods 0.000 claims 1
- 230000006870 function Effects 0.000 description 12
- 230000008569 process Effects 0.000 description 11
- 238000004891 communication Methods 0.000 description 7
- 238000010586 diagram Methods 0.000 description 7
- 238000005516 engineering process Methods 0.000 description 7
- 230000006399 behavior Effects 0.000 description 4
- 238000007477 logistic regression Methods 0.000 description 4
- 230000008859 change Effects 0.000 description 3
- 238000012417 linear regression Methods 0.000 description 3
- 230000007774 longterm Effects 0.000 description 3
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000007405 data analysis Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 238000011835 investigation Methods 0.000 description 2
- 238000013077 scoring method Methods 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 125000004122 cyclic group Chemical group 0.000 description 1
- 238000009795 derivation Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000011144 upstream manufacturing 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
- 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
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
- G06Q10/06311—Scheduling, planning or task assignment for a person or group
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Development Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Physics & Mathematics (AREA)
- Finance (AREA)
- General Physics & Mathematics (AREA)
- Marketing (AREA)
- Accounting & Taxation (AREA)
- Educational Administration (AREA)
- Technology Law (AREA)
- Game Theory and Decision Science (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
Abstract
The invention discloses an overdue collection method, a device and a system thereof and a computer readable storage medium, wherein the method comprises the following steps: when detecting an overdue client, acquiring data information of the overdue client in each preset data field, and processing the data information to obtain corresponding index information; inputting the index information into a preset risk scoring model to output migration risk scoring of the overdue client; and determining a collection urging strategy aiming at overdue clients according to the migration risk score, and executing the collection urging strategy. According to the invention, when the overdue client is detected, the risk condition of the overdue client is more accurately judged by acquiring the data information of the overdue client from a plurality of data fields, so that the collection urging success rate of relevant financial institutions for the overdue client is improved.
Description
Technical Field
The invention relates to the technical field of financial technology (Fintech), in particular to an overdue collection method, device and system and a computer readable storage medium.
Background
In recent years, with the development of financial technology (Fintech), particularly internet financial technology, data analysis technology has been introduced into daily services of financial institutions such as banks. In the daily service process of financial institutions such as banks, in order to receive back the loan amount and interest, the overdue customers need to be urged to receive the loan amount and the interest. The traditional overdue hastening method is to perform polling dialing on all overdue clients by means of manual dialing and the like, and a large amount of manpower is consumed, so that how to perform intelligent hastening on the overdue clients is an important problem to be solved by financial institutions such as banks and the like.
An intelligent overdue collection method mainly carries out risk judgment on overdue clients according to overdue days and overdue money of the overdue clients, and therefore a corresponding collection strategy is adopted according to risk conditions of the overdue clients. However, when the method is used for risk judgment of overdue customers, the used customer information is less, so that the risk condition of the overdue customers cannot be accurately judged, and the collection urging success rate of relevant financial institutions for the overdue customers is lower.
Disclosure of Invention
The invention mainly aims to provide an overdue collection method, device and system and a computer readable storage medium, aiming at more accurately judging the risk condition of overdue customers and improving the collection success rate of the overdue customers.
To achieve the above object, the present invention provides an overdue catalyst comprising the steps of:
when detecting an overdue client, acquiring data information of the overdue client in each preset data field, and processing the data information to obtain corresponding index information;
inputting the index information into a preset risk scoring model to output migration risk scoring of the overdue customer;
and determining a collection urging strategy aiming at the overdue client according to the migration risk score, and executing the collection urging strategy.
Preferably, the step of processing the data information to obtain corresponding index information includes:
acquiring a preset storage format, and formatting each data message according to the preset storage format to obtain corresponding formatted information;
and carrying out secondary processing on the formatted information to obtain corresponding index information.
Preferably, the step of performing secondary processing on the formatted information to obtain corresponding index information includes:
carrying out data cleaning and box separation processing on the formatted information to obtain a processed index variable;
and carrying out derivative treatment on the index variable, and screening the derivative treated index variable to obtain index information.
Preferably, the step of determining a hastening strategy for the overdue customer according to the migration risk score includes:
comparing the migration risk score with a preset migration risk threshold value to determine the migration risk level of the overdue customer;
and determining a collection urging strategy aiming at the overdue client according to the migration risk level.
Preferably, the step of comparing the migration risk score with a preset migration risk threshold to determine the migration risk level of the overdue customer includes:
when the migration risk score is smaller than a preset risk threshold value, determining the migration risk grade of the overdue customer as a first grade;
and when the migration risk score is larger than a preset risk threshold value, determining the migration risk grade of the overdue customer as a second grade.
Preferably, the step of determining a hastening policy for the overdue customer according to the migration risk level includes:
when the migration risk level is a first level, determining that the collection urging strategy for the overdue customer is machine collection urging;
and when the migration risk level is a second level, determining that the collection urging strategy for the overdue client is manual collection urging.
Preferably, before the step of inputting the index information into a preset risk scoring model, the method further includes:
acquiring historical data information of historical overdue clients in each preset data field, and preprocessing the historical data information based on preset processing rules to obtain processed historical index information;
acquiring actual migration conditions corresponding to the historical overdue clients, and marking the historical index information according to the actual migration conditions to obtain corresponding sample data information;
and training the sample data information to obtain a preset risk scoring model.
In addition, to achieve the above object, the present invention also provides an overdue catalyst device, including:
the system comprises an information processing module, a data processing module and a data processing module, wherein the information processing module is used for acquiring data information of overdue clients in each preset data field when the overdue clients are detected, and processing the data information to obtain corresponding index information;
the risk scoring module is used for inputting the index information into a preset risk scoring model so as to output migration risk scoring of the overdue client;
and the strategy execution module is used for determining a collection urging strategy aiming at the overdue client according to the migration risk score and executing the collection urging strategy.
Preferably, the information processing module is further configured to:
acquiring a preset storage format, and formatting each data message according to the preset storage format to obtain corresponding formatted information;
and carrying out secondary processing on the formatted information to obtain corresponding index information.
Preferably, the information processing module is further configured to:
carrying out data cleaning and box separation processing on the formatted information to obtain a processed index variable;
and carrying out derivative treatment on the index variable, and screening the derivative treated index variable to obtain index information.
Preferably, the policy enforcement module is further configured to:
comparing the migration risk score with a preset migration risk threshold value to determine the migration risk level of the overdue customer;
and determining a collection urging strategy aiming at the overdue client according to the migration risk level.
Preferably, the policy enforcement module is further configured to:
when the migration risk score is smaller than a preset risk threshold value, determining the migration risk grade of the overdue customer as a first grade;
and when the migration risk score is larger than a preset risk threshold value, determining the migration risk grade of the overdue customer as a second grade.
Preferably, the policy enforcement module is further configured to:
when the migration risk level is a first level, determining that the collection urging strategy for the overdue customer is machine collection urging;
and when the migration risk level is a second level, determining that the collection urging strategy for the overdue client is manual collection urging.
Preferably, the overdue hastening device further comprises a model training module, and the model training module is configured to:
acquiring historical data information of historical overdue clients in each preset data field, and preprocessing the historical data information based on preset processing rules to obtain processed historical index information;
acquiring actual migration conditions corresponding to the historical overdue clients, and marking the historical index information according to the actual migration conditions to obtain corresponding sample data information;
and training the sample data information to obtain a preset risk scoring model.
In addition, to achieve the above object, the present invention also provides an overdue catalyst system, including: the overdue collection system comprises a memory, a processor and an overdue collection program which is stored on the memory and can run on the processor, wherein the overdue collection program realizes the steps of the overdue collection method when being executed by the processor.
In addition, to achieve the above object, the present invention further provides a computer readable storage medium, which stores an overdue collection program, and when the overdue collection program is executed by a processor, the computer readable storage medium implements the steps of the overdue collection method as described above.
According to the overdue collection method, when an overdue client is detected, data information of the overdue client in each preset data field is obtained, and the data information is processed to obtain corresponding index information; inputting the index information into a preset risk scoring model to output migration risk scoring of the overdue client; and determining a collection urging strategy aiming at overdue clients according to the migration risk score, and executing the collection urging strategy. According to the invention, when the overdue client is detected, the risk condition of the overdue client is more accurately judged by acquiring the data information of the overdue client from a plurality of data fields, so that the collection urging success rate of relevant financial institutions for the overdue client is improved.
Drawings
FIG. 1 is a system diagram of a hardware operating environment according to an embodiment of the present invention;
FIG. 2 is a schematic flow chart of a first embodiment of the overdue catalytic recovery method of the present invention;
FIG. 3 is a table diagram illustrating data information of overdue clients and their related descriptions in a preferred embodiment of the overdue hastening method of the present invention;
FIG. 4 is a table diagram of an index information base in accordance with a preferred embodiment of the overdue hastening method of the present invention;
FIG. 5 is a functional image of a logistic function in the late catalyst recovery method of the present invention;
FIG. 6 is a functional block diagram of a preferred embodiment of the overdue catalyst recovery method of the present invention.
The implementation, functional features and advantages of the objects of the present invention will be further explained with reference to the accompanying drawings.
Detailed Description
It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
As shown in fig. 1, fig. 1 is a system structural diagram of a hardware operating environment according to an embodiment of the present invention.
The system of the embodiment of the invention can be a mobile terminal, a robot or PC equipment and the like.
As shown in fig. 1, the system may include: a processor 1001, such as a CPU, a network interface 1004, a user interface 1003, a memory 1005, a communication bus 1002. Wherein a communication bus 1002 is used to enable connective communication between these components. The user interface 1003 may include a Display screen (Display), an input unit such as a Keyboard (Keyboard), and the optional user interface 1003 may also include a standard wired interface, a wireless interface. The network interface 1004 may optionally include a standard wired interface, a wireless interface (e.g., WI-FI interface). The memory 1005 may be a high-speed RAM memory or a non-volatile memory (e.g., a magnetic disk memory). The memory 1005 may alternatively be a storage device separate from the processor 1001.
Those skilled in the art will appreciate that the system architecture shown in FIG. 1 is not intended to be limiting of the system, and may include more or fewer components than those shown, or some components may be combined, or a different arrangement of components.
As shown in fig. 1, the memory 1005, which is a kind of computer storage medium, may include therein an operating system, a network communication module, a user interface module, and a overdue hastening program.
The operating system is a program for managing and controlling overdue hastening systems and software resources, and supports the operation of a network communication module, a user interface module, an overdue hastening program and other programs or software; the network communication module is used for managing and controlling the network interface 1002; the user interface module is used to manage and control the user interface 1003.
In the overdue catalyst system shown in fig. 1, the overdue catalyst system calls an overdue catalyst program stored in the memory 1005 through the processor 1001 and performs operations in various embodiments of the overdue catalyst method described below.
Based on the hardware structure, the embodiment of the overdue catalyst recovery method is provided.
Referring to fig. 2, fig. 2 is a schematic flow chart of a first embodiment of the overdue catalytic recovery method of the present invention, which comprises:
step S10, when detecting the overdue client, acquiring the data information of the overdue client in each preset data field, and processing the data information to obtain corresponding index information;
the overdue collection method is applied to overdue collection systems of financial institutions such as financial institutions or banks. In recent years, with the development of financial technology (Fintech), particularly internet financial technology, data analysis technology has been introduced into daily services of financial institutions such as banks. In the daily service process of financial institutions such as banks, in order to receive back the loan amount and interest, the overdue customers need to be urged to receive the loan amount and the interest. The traditional overdue hastening method is to carry out polling dialing on all overdue customers by means of manual dialing and the like, obviously, the hastening cost is high, and therefore, how to carry out intelligent hastening on the overdue customers is an important problem to be solved by financial institutions such as banks and the like.
The existing overdue hastening method mainly judges the risk of overdue customers according to the overdue days and the overdue amount of the overdue customers, the used customer information is less, so that the risk condition of the overdue customers cannot be accurately judged, a related financial institution cannot make a refined hastening strategy according to the risk condition of the overdue customers, and the success rate of hastening the overdue customers is low.
In this embodiment, an overdue customer is a customer who has not fully paid for a due loan beyond the bill day or grace period. When an overdue client is newly added, namely when the overdue client is detected by an overdue hastening system, on the premise of meeting national laws and regulations and rules, the data information of the newly added overdue client in each preset field is obtained, wherein the preset fields can comprise the fields of enterprise basic attributes, industrial and commercial information, tax information, client historical hastening conditions and the like. Referring to fig. 3, fig. 3 is a list diagram illustrating data information of overdue clients and descriptions related to the data information in a preferred embodiment of the overdue clients in the present invention, wherein the data information of the overdue clients includes enterprise basic information, information of a convergent legal person and a company, historical collection of the clients, historical behavior of the clients, information of credit investigation of people, tax information, information of multi-headed loan, information of industrial and commercial enterprises, and the like. Specifically, the enterprise basic information includes information such as the area where the enterprise is located, the industry where the enterprise belongs, the type of taxpayer of the enterprise, the type of the enterprise, and the like; the information of the confluent legal people and the company comprises information of enterprise people credit investigation, individuals of enterprise legal representatives and the like; the historical collection urging situation of the client comprises information such as collection urging feedback situation when the history is overdue and the like existing in a financial institution (hereinafter referred to as a 'home financial institution') where the overdue collection urging system is located by the overdue client; the historical behavior of the client comprises the information of historical overdue, repayment, limit use and the like of the overdue client in the financial institution; the information of the people's bank credit includes the information of the people's bank credit of enterprises, the individual people's bank credit of legal representatives, etc.; the tax information comprises enterprise tax report information; the multi-head loan information comprises information of loan application, use expression and the like of the overdue client in other financial institutions; enterprise business information enterprise business registration and change. And then processing the data information of the overdue client in each preset data field to obtain corresponding index information, and outputting a migration risk score of the overdue client according to the index information.
Further, the step of processing the data information to obtain corresponding index information includes:
a1, acquiring a preset storage format, and formatting each data message according to the preset storage format to obtain corresponding formatted information;
and a2, performing secondary processing on the formatted information to obtain corresponding index information.
In this embodiment, when processing the data information of the overdue client, the data information may be formatted according to a preset storage format, for example, the formatted information may be obtained after the data information is formatted, the formatted information is stored in the preset database in the form of a database table, and then the formatted information is processed again, that is, the formatted information is processed again, so as to obtain the index information.
It should be noted that the preset storage formats may be the same or different for various data information of many overdue clients. Different preset storage formats can be designed according to different overdue business types of overdue customers, for example, aiming at data information of overdue customers with the overdue business type being bank business, the data information can be stored in a preset database in the form of a database table; for data information of overdue customers with overdue business types as leasing business, the data information can be stored in a preset database in a picture format, and the like. The invention does not specifically limit the preset storage format of the data information, and the specifically selected storage format can be set according to the actual requirement.
Further, step a2 further includes:
step a21, carrying out data cleaning and box separation processing on the formatted information to obtain processed index variables;
step a22, deriving the index variables, and screening the derived index variables to obtain index information.
In this embodiment, Data cleansing (Data cleansing) is a process of reviewing and verifying Data, and aims to delete duplicate information, correct existing errors, such as format errors, and maintain consistency of Data information; the data binning processing is to divide a section of continuous data information into a plurality of sections, regard the data information of each section as a classification, that is, convert the continuous data information into discrete data information to obtain a processed index variable, further perform derivation processing on the index variable to obtain a corresponding derived variable, and then screen the derived variable and the index variable to obtain corresponding index information. In the process of carrying out statistical analysis on the data information, secondary processing is carried out on the formatted information, so that the accuracy of model training can be further improved.
Referring to fig. 4, fig. 4 is a table diagram of an index information base in a preferred embodiment of the overdue collection method of the present invention, wherein the basic information of an enterprise belongs to the information of the basic attribute data field of the enterprise, including the information of company status, tax payment credit rating, industry, operation range, region, corporate basic information, etc.; the information of the college legal person and the company comprises the field of the college legal person and the field of the college legal company, the field of the college legal person also comprises information such as executed information, network loan information, height limit and entry and exit limit information, and the field of the college legal company also comprises information such as company folk loan, small amount loan, guarantee, contract dispute, execution, property guarantee and the like; the historical customer collection urging condition belongs to the historical customer collection urging condition and comprises information such as historical manual and robot collection urging times of customers, manual and robot collection urging reaching rate, promised repayment rate and the like; the client historical behaviors comprise information in the fields of limit information, borrowing information, withdrawal information, settlement information, early compensation information, overdue information and the like, specifically, the limit information field further comprises information of limit usage rate, full days and the like, and derivative variables corresponding to the full days comprise information of the proportion of the full days and the like; the borrowing information field comprises information such as the number of borrowing strokes in the latest 1/3/6/12/24 months, and the derivative variables corresponding to the borrowing information can be the same ratio and the ring ratio of the number of borrowing strokes in the latest 1/3/6/9/12 months; the withdrawal information field comprises information such as the number of withdrawals in nearly 1/3/6/12/24 months, the amount and the like, and the corresponding derivative variables can be the number of withdrawals in the latest 1/3/6/9/12 months, the same proportion of the amount, the ring proportion and the like; the clearing information field comprises information such as the number of clear strokes, the amount of money and the shortest clearing time of nearly 1/3/6/12/24 months, and corresponding derivative variables can be the number of clear strokes, the amount of money and the shortest clearing time of the latest 1/3/6/9/12 months, the same ratio, the cyclic ratio and the like; the early-compensation information field comprises information such as early-compensation stroke number of the latest 1/3/6/12/24 months, and corresponding derivative variables can be the same ratio and ring ratio of the early-compensation stroke number of the latest 1/3/6/9/12 months; the overdue information field comprises the current overdue state, the historical maximum overdue days, whether the current overdue days are overdue, short/medium/long-term overdue times, the same ring ratio of the overdue times and the like, and also comprises the curing conditions of the short/medium/long-term overdue, such as the repayment conditions of short/medium/long-term clients after the clients are overdue, and the personal credit information comprises the personal credit information of an enterprise legal person and the credit information of an enterprise end and the like; the tax information belongs to the information of the tax information field, including enterprise tax basic information, tax payment records, obtained tax and income conditions, enterprise upstream and downstream transaction information, tax collection conditions, tax change conditions, tax violation conditions, asset liability and profit information and the like; the multi-head loan information belongs to information in the field of multi-head loan, and comprises multi-head loan information of overdue clients; the business information belongs to the information in the business information field, and comprises the change times of related information of legal persons and enterprises and related punishment information of the business and the industry.
Step S20, inputting the index information into a preset risk scoring model to output migration risk scoring of the overdue client;
in this embodiment, index information of an overdue client is used as an input of the risk scoring model, so that a migration risk score of the overdue client is output, and the larger the score of the overdue risk score is, the higher the possibility that the overdue client of the current overdue stage migrates to the next overdue stage is. And when the client enters overdue, the risk scoring is carried out on the overdue client, so that reference can be timely provided for the arrangement of the collection urging strategy.
In addition, regarding model training, the index historical sample information may be used to estimate model parameters through a predetermined algorithm, such as Logistic Regression. Wherein the Logistic regression scoring method is to introduce a conversion function such as Logistic function based on linear regression, and the migration risk score obtained by linear regression can be converted into a score in [0,1]]The value of the interval. Specifically, if the index information of the overdue client includes (x)0，x1，x2，...，xd) Calculating by using a linear regression scoring method to obtain a migration risk score s of the overdue client as follows:
wherein d is the index information number of overdue clients;
xithe ith index information;
wiis xiCorresponding riskAnd (4) scoring parameters.
The higher the calculated score is, the greater the risk that the overdue customer migrates to the next stage; the lower the score, the less risk of the overdue customer migrating to the next stage. Since the value of the migration risk score S is [ - ∞, + ∞ ] and if the migration risk score is to be converted to a value between [0 and 1], a conversion function, such as a Logistic function, is introduced to convert the migration risk score S to a value between [0 and 1], and the Logistic function is an "S" -shaped function, the shape of which is shown in fig. 5. Furthermore, the Logistic function, also called sigmoid function, can map the migration risk score s between [0,1 ]. The Logistic function θ(s) is calculated by the formula:
where e is a natural constant.
In summary, the function calculation formula of the entire Logistic Regression is as follows:
wherein, wTIs a set of weights that have been trained.
The migration risk score s is converted into a value between [0,1] through a logistic regression function, so that the prediction effect of the model is improved, the probability of migration of an overdue client to the next overdue stage can be known more intuitively, and different collection strategies can be deployed by related financial institutions according to the overdue clients with different migration probabilities.
It should be noted that a rule fit method can also be combined to participate in the modeling process of the migration risk model. Because RuleFit automatically adds feature interaction to the linear model, the method solves the problem that the linear model of interaction items must be manually added, is very helpful for modeling, can also process classification and regression tasks through the RuleFit method, and can further improve the prediction effect of the model.
Further, before the step of inputting the index information into a preset risk scoring model, the method further includes:
step b1, acquiring historical data information of the historical overdue clients in each preset data field, and preprocessing the historical data information based on preset processing rules to obtain processed historical index information;
step b2, acquiring actual migration conditions corresponding to the history overdue clients, and marking the history index information according to the actual migration conditions to obtain corresponding sample data information;
and b3, training the sample data information to obtain a preset risk scoring model.
In this embodiment, before the preset risk scoring model is applied to the actually deployed application, the initial risk scoring model needs to be trained to obtain the trained preset risk scoring model. Specifically, historical data information of known historical overdue clients in each preset data field is obtained, the processed historical index information is obtained according to preset processing rules, such as format processing, data cleaning, box dividing processing and the like on the historical data information, actual migration conditions corresponding to the historical overdue clients are obtained, overdue grades corresponding to the historical overdue clients are determined according to the actual migration conditions, and therefore the historical index information of the historical overdue clients with different overdue grades is marked. For example, the historical overdue clients include M1 overdue clients (clients known to have overdue days within 1 to 30 days) and M2 overdue clients (clients known to have overdue days within 31 to 60 days), and within a preset statistical period, the M2 overdue clients include part of original M1 overdue clients, that is, part of original M1 overdue clients have migrated to M2 for accounting to become M2 overdue clients, and clients migrating to M2 overdue stages after M1 overdue for 30 days are negative samples (which may be marked as 1), and clients not migrating to M2 overdue stages after M1 overdue for 30 days are positive samples (which may be marked as 0), so as to mark the historical index information of each historical overdue client, obtain corresponding sample data information, and input the sample data information into the initial risk scoring model for training, so as to obtain the preset risk scoring model after training is completed.
And step S30, determining a collection urging strategy aiming at the overdue customer according to the migration risk score, and executing the collection urging strategy.
In this embodiment, assume that the overdue risk score of the M1 overdue client is between [0,1000], where the overdue risk score is close to 0, which indicates that the pre-set risk score model has a very low possibility of predicting the overdue client to migrate to the next overdue stage; when the overdue risk score approaches 1000, it indicates that the overdue client has a very high probability of migrating to the next overdue stage. Specifically, if the preset risk score model outputs a migration risk score of 360 points for the M1 overdue client a, it indicates that the M1 overdue client a migrates to the M2 stage, and the possibility of becoming the M2 overdue client is high; the preset risk score model outputs a migration risk score of 960 for the M1 overdue client B, which indicates that the M1 overdue client a has a high possibility of becoming an M2 overdue client when migrating to the M2 stage. The method comprises the steps of obtaining data information of newly-added overdue clients in each preset data field, carrying out data processing on the data information to obtain index information, and inputting the index information into a preset risk scoring model to output risk scores of the overdue clients. On the premise of meeting national laws and regulations, data information of overdue clients, such as enterprise basic information, information of a convergent legal person and a company, historical collection conditions of the clients, historical behaviors of the clients, credit information of a person enterprise terminal, credit information of a legal person representative personal terminal, tax related information, multi-head loan information, industrial and commercial information and the like, is acquired from a plurality of data fields, and the data information is processed and then input into a risk scoring model, so that the overdue clients can be more accurately output for migration risk scoring, and relevant financial institutions can be facilitated to deploy corresponding collection urging strategies for the overdue clients in time.
According to the overdue hastening method, when an overdue client is detected, data information of the overdue client in each preset data field is acquired, and the data information is processed to obtain corresponding index information; inputting the index information into a preset risk scoring model to output migration risk scoring of the overdue client; and determining a collection urging strategy aiming at overdue clients according to the migration risk score, and executing the collection urging strategy. According to the invention, when the overdue client is detected, the risk condition of the overdue client is more accurately judged by acquiring the data information of the overdue client from a plurality of data fields, so that the collection urging success rate of relevant financial institutions for the overdue client is improved.
Further, based on the first embodiment of the overdue catalytic recovery method of the present invention, a second embodiment of the overdue catalytic recovery method of the present invention is provided.
The second embodiment of the overdue promotion method differs from the first embodiment of the overdue promotion method in that the step of determining a promotion policy for the overdue customer according to the migration risk score comprises:
step c, comparing the migration risk score with a preset migration risk threshold value to determine the migration risk level of the overdue customer;
in this embodiment, after the preset risk scoring model is deployed online, the migration risk score of the newly added overdue client can be automatically calculated by inputting index information of the overdue client without manual intervention. Assuming that the overdue risk score of the overdue client is in the range of [0,1000], a reasonable score can be selected from the score (0,1000) as a preset migration risk threshold n, namely 0< n <1000, and the migration risk score of the overdue client is compared with the preset migration risk threshold to determine the migration risk level of the overdue client, so that the financial institution can conveniently adopt different hastening strategies for the overdue clients with different migration risk levels.
Further, step c further comprises:
step c1, when the migration risk score is smaller than a preset risk threshold, determining the migration risk grade of the overdue customer as a first grade;
and c2, when the migration risk score is larger than a preset risk threshold, determining the migration risk grade of the overdue customer as a second grade.
In this embodiment, if the migration risk score of an overdue client is output to be smaller than a preset migration risk threshold n, the overdue client is determined to be a low-risk client, and the migration risk grade of the overdue client is determined to be a first grade; and if the output migration risk score is larger than a preset migration risk threshold n, determining that the overdue client is a high-risk client, determining that the migration risk grade of the overdue client is a second grade, and determining the migration risk grade of the overdue client according to the actual business requirement for the overdue client with the migration risk score equal to the preset migration risk threshold n.
And d, determining a collection urging strategy aiming at the overdue client according to the migration risk level.
In the embodiment, differential collection urging strategies and means can be formulated according to migration risk levels of overdue clients, the overdue clients can be purposefully collected, and the collection urging success rate can be improved to a certain extent.
Further, step d further comprises:
step d1, when the migration risk level is a first level, determining that the collection urging strategy for the overdue customer is machine collection urging;
and d2, when the migration risk level is a second level, determining that the collection urging strategy for the overdue client is manual collection urging.
In this embodiment, when an overdue customer is a low-risk customer, that is, the migration risk level of the overdue customer is a first level, it may be determined that the collection policy for the low-risk customer is machine collection, where machine collection means that the overdue collection system automatically notifies the overdue customer, for example, a short message service module in the overdue collection system sends a short message to notify the relevant overdue customer that the customer has paid off for the past bill day or the unlimited term without full payment; when the overdue client is a high-risk client, namely the migration risk grade of the overdue client is the second grade, the overdue collection system can send the relevant information of the overdue client to a worker with authority on the premise of not revealing client information so as to inform the worker of needing to manually collect the overdue client, for example, the corresponding worker can be arranged to carry out telephone communication on the overdue client so as to inform the client that the loan is not paid back for the overdue bill day or the unlimited term, and inform the client of measures and the like if the loan is not paid back within the specified date. The overdue promotion system outputs migration risk scores of all overdue clients according to a preset risk score model when the clients just enter an overdue stage, automatically judges whether a machine promotion strategy with low promotion intensity is adopted for the overdue clients or manual promotion is needed to be intervened in advance to remind the overdue clients to pay back loans within a specified date, and adopts promotion strategies with different intensities for the overdue clients with different migration risk grades, so that the promotion success rate of promotion for the highly overdue clients to promote the receipts can be further improved.
It should be noted that the migration risk level of the overdue customer can be determined comprehensively according to the actual business requirements of the financial institution, the customer group and the like, so that the number of the migration risk levels of the overdue customer is not limited by the application.
The overdue hastening method of the embodiment adopts different hastening strategies aiming at overdue clients with different migration risk levels, and purposefully hastens the overdue clients, so that the success rate of hastening and receiving can be further improved.
The invention also provides an overdue recovery device. Referring to fig. 6, the overdue catalyst device of the present invention includes:
the information processing module 10 is configured to, when an overdue client is detected, acquire data information of the overdue client in each preset data field, and process the data information to obtain corresponding index information;
the risk scoring module 20 is used for inputting the index information into a preset risk scoring model so as to output migration risk scoring of the overdue client;
and the policy execution module 30 is configured to determine a revenue promotion policy for the overdue customer according to the migration risk score, and execute the revenue promotion policy.
Preferably, the information processing module is further configured to:
acquiring a preset storage format, and formatting each data message according to the preset storage format to obtain corresponding formatted information;
and carrying out secondary processing on the formatted information to obtain corresponding index information.
Preferably, the information processing module is further configured to:
carrying out data cleaning and box separation processing on the formatted information to obtain a processed index variable;
and carrying out derivative treatment on the index variable, and screening the derivative treated index variable to obtain index information.
Preferably, the policy enforcement module is further configured to:
comparing the migration risk score with a preset migration risk threshold value to determine the migration risk level of the overdue customer;
and determining a collection urging strategy aiming at the overdue client according to the migration risk level.
Preferably, the policy enforcement module is further configured to:
when the migration risk score is smaller than a preset risk threshold value, determining the migration risk grade of the overdue customer as a first grade;
and when the migration risk score is larger than a preset risk threshold value, determining the migration risk grade of the overdue customer as a second grade.
Preferably, the policy enforcement module is further configured to:
when the migration risk level is a first level, determining that the collection urging strategy for the overdue customer is machine collection urging;
and when the migration risk level is a second level, determining that the collection urging strategy for the overdue client is manual collection urging.
Preferably, the overdue hastening device further comprises a model training module, and the model training module is configured to:
acquiring historical data information of historical overdue clients in each preset data field, and preprocessing the historical data information based on preset processing rules to obtain processed historical index information;
acquiring actual migration conditions corresponding to the historical overdue clients, and marking the historical index information according to the actual migration conditions to obtain corresponding sample data information;
and training the sample data information to obtain a preset risk scoring model.
The invention also provides a computer readable storage medium.
The computer readable storage medium of the present invention stores an overdue collection program, and the overdue collection program, when executed by a processor, implements the steps of the overdue collection method as described above.
The method implemented when the overdue solicitation program running on the processor is executed may refer to various embodiments of the overdue solicitation method of the present invention, and will not be described herein again.
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or system 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 system. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or system that comprises the element.
The above-mentioned serial numbers of the embodiments of the present invention are merely for description and do not represent the merits of the embodiments.
Through the above description of the embodiments, those skilled in the art will clearly understand that the method of the above embodiments can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware, but in many cases, the former is a better implementation manner. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium (e.g., ROM/RAM, magnetic disk, optical disk) as described above and includes instructions for enabling a terminal system (e.g., a mobile phone, a computer, a server, an air conditioner, or a network system) to execute the method according to the embodiments of the present invention.
The above description is only a preferred embodiment of the present invention, and not intended to limit the scope of the present invention, and all modifications of equivalent structures and equivalent processes, which are made by using the contents of the present specification and the accompanying drawings, or directly or indirectly applied to other related technical fields, are included in the scope of the present invention.
Claims (10)
1. An overdue catalytic recovery method, characterized in that the overdue catalytic recovery method comprises the following steps:
when detecting an overdue client, acquiring data information of the overdue client in each preset data field, and processing the data information to obtain corresponding index information;
inputting the index information into a preset risk scoring model to output migration risk scoring of the overdue customer;
and determining a collection urging strategy aiming at the overdue client according to the migration risk score, and executing the collection urging strategy.
2. The overdue hastening method according to claim 1, wherein the step of processing the data information to obtain corresponding index information comprises:
acquiring a preset storage format, and formatting each data message according to the preset storage format to obtain corresponding formatted information;
and carrying out secondary processing on the formatted information to obtain corresponding index information.
3. The overdue hastening method according to claim 2, wherein the step of performing secondary processing on the formatted information to obtain corresponding index information comprises:
carrying out data cleaning and box separation processing on the formatted information to obtain a processed index variable;
and carrying out derivative treatment on the index variable, and screening the derivative treated index variable to obtain index information.
4. The overdue hastening method of claim 1, wherein the determining a hastening strategy for the overdue customer based on the migration risk score comprises:
comparing the migration risk score with a preset migration risk threshold value to determine the migration risk level of the overdue customer;
and determining a collection urging strategy aiming at the overdue client according to the migration risk level.
5. The overdue harvesting method of claim 4, wherein the step of comparing the migration risk score to a preset migration risk threshold to determine a migration risk level of the overdue customer comprises:
when the migration risk score is smaller than a preset risk threshold value, determining the migration risk grade of the overdue customer as a first grade;
and when the migration risk score is larger than a preset risk threshold value, determining the migration risk grade of the overdue customer as a second grade.
6. The overdue hastening method of claim 5, wherein the determining a hastening strategy for the overdue customer according to the migration risk level comprises:
when the migration risk level is a first level, determining that the collection urging strategy for the overdue customer is machine collection urging;
and when the migration risk level is a second level, determining that the collection urging strategy for the overdue client is manual collection urging.
7. The overdue hastening method according to any one of claims 1 to 6, wherein the step of inputting the index information into a preset risk scoring model is preceded by the steps of:
acquiring historical data information of historical overdue clients in each preset data field, and preprocessing the historical data information based on preset processing rules to obtain processed historical index information;
acquiring actual migration conditions corresponding to the historical overdue clients, and marking the historical index information according to the actual migration conditions to obtain corresponding sample data information;
and training the sample data information to obtain a preset risk scoring model.
8. An over-term recovery device, comprising:
the system comprises an information processing module, a data processing module and a data processing module, wherein the information processing module is used for acquiring data information of overdue clients in each preset data field when the overdue clients are detected, and processing the data information to obtain corresponding index information;
the risk scoring module is used for inputting the index information into a preset risk scoring model so as to output migration risk scoring of the overdue client;
and the strategy execution module is used for determining a collection urging strategy aiming at the overdue client according to the migration risk score and executing the collection urging strategy.
9. An overdue catalyst system, comprising: a memory, a processor, and an overdue collection program stored on the memory and executable on the processor, the overdue collection program when executed by the processor implementing the steps of the overdue collection method of any of claims 1 to 7.
10. A computer readable storage medium, characterized in that the computer readable storage medium has stored thereon an overdue collection program, which when executed by a processor implements the steps of the overdue collection method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110243561.4A CN112907356A (en) | 2021-03-04 | 2021-03-04 | Overdue collection method, device and system and computer readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110243561.4A CN112907356A (en) | 2021-03-04 | 2021-03-04 | Overdue collection method, device and system and computer readable storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112907356A true CN112907356A (en) | 2021-06-04 |
Family
ID=76108231
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110243561.4A Pending CN112907356A (en) | 2021-03-04 | 2021-03-04 | Overdue collection method, device and system and computer readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112907356A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113643118A (en) * | 2021-07-02 | 2021-11-12 | 重庆度小满优扬科技有限公司 | Method and device for client layering |
| CN114066620A (en) * | 2021-11-29 | 2022-02-18 | 中国工商银行股份有限公司 | Client information processing method and device based on client portrait |
| CN114357525A (en) * | 2022-03-10 | 2022-04-15 | 杭银消费金融股份有限公司 | Data security processing method, equipment and medium based on financial business |
| CN117372153A (en) * | 2023-10-31 | 2024-01-09 | 金扁担(北京)数字科技有限公司 | Fraud risk model and credit risk model based collection promoting method |
| CN117788189A (en) * | 2023-12-22 | 2024-03-29 | 广州智租信息科技有限责任公司 | Intelligent collection promoting system and method for enterprise service business based on big data analysis |
-
2021
- 2021-03-04 CN CN202110243561.4A patent/CN112907356A/en active Pending
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113643118A (en) * | 2021-07-02 | 2021-11-12 | 重庆度小满优扬科技有限公司 | Method and device for client layering |
| CN113643118B (en) * | 2021-07-02 | 2023-08-25 | 度小满科技(北京)有限公司 | Method and device for client layering |
| CN114066620A (en) * | 2021-11-29 | 2022-02-18 | 中国工商银行股份有限公司 | Client information processing method and device based on client portrait |
| CN114357525A (en) * | 2022-03-10 | 2022-04-15 | 杭银消费金融股份有限公司 | Data security processing method, equipment and medium based on financial business |
| CN114357525B (en) * | 2022-03-10 | 2022-06-14 | 杭银消费金融股份有限公司 | Data security processing method, equipment and medium based on financial business |
| CN117372153A (en) * | 2023-10-31 | 2024-01-09 | 金扁担(北京)数字科技有限公司 | Fraud risk model and credit risk model based collection promoting method |
| CN117788189A (en) * | 2023-12-22 | 2024-03-29 | 广州智租信息科技有限责任公司 | Intelligent collection promoting system and method for enterprise service business based on big data analysis |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112907356A (en) | Overdue collection method, device and system and computer readable storage medium | |
| CN110400215B (en) | Method and system for constructing enterprise family-oriented small micro enterprise credit assessment model | |
| US7933878B2 (en) | Assessing and managing operational risk in organizational operations | |
| US8751436B2 (en) | Analyzing data quality | |
| US20130179314A1 (en) | Risk Based Data Assessment | |
| CN109102145B (en) | Process orchestration | |
| WO2010037030A1 (en) | Evaluating loan access using online business transaction data | |
| US20100332899A1 (en) | Quality Management in a Data-Processing Environment | |
| CN107103548A (en) | The monitoring method and system and risk monitoring and control method and system of network behavior data | |
| CN108346036A (en) | Insurance policy concentrates vouching method, electronic device and readable storage medium storing program for executing | |
| WO2022155740A1 (en) | Systems and methods for outlier detection of transactions | |
| CN111709826A (en) | Target information determination method and device | |
| CN113609193A (en) | Method and device for training prediction model for predicting customer transaction behavior | |
| CN106408325A (en) | User consumption behavior prediction analysis method based on user payment information and system | |
| CN112950359B (en) | User identification method and device | |
| CN114493686A (en) | Operation content generation and pushing method and device | |
| CN112712270B (en) | Information processing method, device, equipment and storage medium | |
| CN114092230A (en) | Data processing method and device, electronic equipment and computer readable medium | |
| CN117172612A (en) | Enterprise object evaluation method and device | |
| CN117196630A (en) | Transaction risk prediction method, device, terminal equipment and storage medium | |
| CN116934131A (en) | Enterprise operation condition assessment method, device and equipment | |
| AU2012201419A1 (en) | Risk based data assessment | |
| JP7231967B1 (en) | Credit risk judgment system and credit risk judgment program | |
| KR20230094936A (en) | Activist alternative credit scoring system model using work behavior data and method for providing the same | |
| CN116485213A (en) | User access method, device, electronic equipment and computer readable storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210604 |
|
| RJ01 | Rejection of invention patent application after publication |