CN113628026A - Method and device for predicting overdue risk ranking - Google Patents
Method and device for predicting overdue risk ranking Download PDFInfo
- Publication number
- CN113628026A CN113628026A CN202110731639.7A CN202110731639A CN113628026A CN 113628026 A CN113628026 A CN 113628026A CN 202110731639 A CN202110731639 A CN 202110731639A CN 113628026 A CN113628026 A CN 113628026A
- Authority
- CN
- China
- Prior art keywords
- risk
- ranking
- overdue
- user
- model
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/03—Credit; Loans; Processing thereof
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- 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
Abstract
The invention aims to provide a method and a device for predicting overdue risk ranking. The method comprises the following steps: acquiring overdue characteristic information of a user to be evaluated; inputting overdue characteristic information of the user to be evaluated into a preset risk sorting model; predicting a ranking of the overdue risk of the user to be evaluated based on an output of a risk ranking model, wherein the output of the risk ranking model is used for indicating the ranking of the overdue risk of the user to be evaluated. The embodiment of the application has the following advantages: the risk ranking model is established and trained on the basis of pairwise risk ranking relations of the training samples, so that overdue risk ranking of the users is predicted through the risk ranking model, the model can judge overdue risk ranking of different users on the basis of differences of overdue behaviors such as overdue time and the like, and compared with a traditional risk control model, the prediction accuracy is improved.
Description
Technical Field
The invention relates to the technical field of computers, in particular to a method and a device for predicting overdue risk ranking.
Background
In prior art solutions, the pre-or post-loan risk control model generally takes the following two approaches: 1) evaluating the card mode; 2) other machine models, such as decision trees, gbdt, deep learning, etc. In the two modes, the risk judgment problem is used as a classification problem to be fitted, an optimized estimation of the overdue probability under x is obtained based on input x, so that a model is finally obtained, x is input into the model, and the overdue probability is output.
However, the solutions based on the prior art, as above two approaches, have some drawbacks: because a fixed observation period, for example, 12 months, is required, the training and verification of the model only judges whether the observation period is overdue, i.e. the observation period is used as a 0 and 1 label of the classification task, and the factors such as the time length of overdue are not further distinguished. For example, the loan is also overdue, but the overdue time is earlier, and the loan is overdue by the 1 st month, namely the loan is generally a bad fraudulent client. In addition, in the method, the observation period is required to be kept long enough for training and testing the model, so that the time for training data is more advanced, and the modeling effect is influenced.
Disclosure of Invention
The invention aims to provide a method and a device for predicting overdue risk ranking.
According to an embodiment of the application, there is provided a method for predicting overdue risk ranking, wherein the method comprises the steps of:
acquiring overdue characteristic information of a user to be evaluated;
inputting overdue characteristic information of the user to be evaluated into a preset risk sorting model;
predicting a ranking of the overdue risk of the user to be evaluated based on an output of a risk ranking model, wherein the output of the risk ranking model is used for indicating the ranking of the overdue risk of the user to be evaluated.
According to an embodiment of the present application, there is provided an apparatus for predicting overdue risk ranking, wherein the apparatus includes:
means for obtaining overdue feature information of a user to be assessed;
means for inputting overdue feature information of the user to be assessed to a preset risk ranking model;
means for predicting a ranking of the overdue risk of the user to be assessed based on an output of a risk ranking model, wherein the output of the risk ranking model is indicative of the ranking of the overdue risk of the user to be assessed.
According to an embodiment of the present application, there is provided a computer device including a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the processor implements the method of the embodiment of the present application when executing the program.
According to an embodiment of the present application, there is provided a computer-readable storage medium on which a computer program is stored, characterized in that the program, when executed by a processor, implements the method of the embodiment of the present application.
Compared with the prior art, the embodiment of the application has the following advantages: the risk ranking model is established and trained on the basis of pairwise risk ranking relations of the training samples, so that overdue risk ranking of the users is predicted through the risk ranking model, the model can judge overdue risk ranking of different users on the basis of differences of overdue behaviors such as overdue time and the like, and compared with a traditional risk control model, the prediction accuracy is improved.
Drawings
Other features, objects and advantages of the invention will become more apparent upon reading of the detailed description of non-limiting embodiments made with reference to the following drawings:
FIG. 1 shows a flow diagram of a method for predicting overdue risk ranking according to an embodiment of the application;
fig. 2 shows a schematic structural diagram of an apparatus for predicting overdue risk ranking according to an embodiment of the present application.
The same or similar reference numbers in the drawings identify the same or similar elements.
Detailed Description
Before discussing exemplary embodiments in more detail, it should be noted that some exemplary embodiments are described as processes or methods depicted as flowcharts. Although a flowchart may describe the operations as a sequential process, many of the operations can be performed in parallel, concurrently, or simultaneously. In addition, the order of the operations may be re-arranged. The process may be terminated when its operations are completed, but may have additional steps not included in the figure. The processes may correspond to methods, functions, procedures, subroutines, and the like.
The term "computer device" or "computer" in this context refers to an intelligent electronic device that can execute predetermined processes such as numerical calculation and/or logic calculation by running predetermined programs or instructions, and may include a processor and a memory, wherein the processor executes a pre-stored instruction stored in the memory to execute the predetermined processes, or the predetermined processes are executed by hardware such as ASIC, FPGA, DSP, or a combination thereof. Computer devices include, but are not limited to, servers, personal computers, laptops, tablets, smart phones, and the like.
The computer equipment comprises user equipment and network equipment. Wherein the user equipment includes but is not limited to computers, smart phones, PDAs, etc.; the network device includes, but is not limited to, a single network server, a server group consisting of a plurality of network servers, or a Cloud computing (Cloud computing) -based Cloud consisting of a large number of computers or network servers, wherein Cloud computing is one of distributed computing, a super virtual computer consisting of a collection of loosely coupled computers. The computer equipment can be independently operated to realize the application, and can also be accessed into a network to realize the application through the interactive operation with other computer equipment in the network. The network in which the computer device is located includes, but is not limited to, the internet, a wide area network, a metropolitan area network, a local area network, a VPN network, and the like.
It should be noted that the user equipment, the network device, the network, etc. are only examples, and other existing or future computer devices or networks may also be included in the scope of the present application, if applicable, and are included by reference.
The methods discussed below, some of which are illustrated by flow diagrams, may be implemented by hardware, software, firmware, middleware, microcode, hardware description languages, or any combination thereof. When implemented in software, firmware, middleware or microcode, the program code or code segments to perform the necessary tasks may be stored in a machine or computer readable medium such as a storage medium. The processor(s) may perform the necessary tasks.
Specific structural and functional details disclosed herein are merely representative and are provided for purposes of describing example embodiments of the present application. This application may, however, be embodied in many alternate forms and should not be construed as limited to only the embodiments set forth herein.
It will be understood that, although the terms first, second, etc. may be used herein to describe various elements, these elements should not be limited by these terms. These terms are only used to distinguish one element from another. For example, a first element may be termed a second element, and, similarly, a second element may be termed a first element, without departing from the scope of example embodiments. As used herein, the term "and/or" includes any and all combinations of one or more of the associated listed items.
It will be understood that when an element is referred to as being "connected" or "coupled" to another element, it can be directly connected or coupled to the other element or intervening elements may be present. In contrast, when an element is referred to as being "directly connected" or "directly coupled" to another element, there are no intervening elements present. Other words used to describe the relationship between elements (e.g., "between" versus "directly between", "adjacent" versus "directly adjacent to", etc.) should be interpreted in a similar manner.
The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of example embodiments. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms "comprises" and/or "comprising," when used herein, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof.
It should also be noted that, in some alternative implementations, the functions/acts noted may occur out of the order noted in the figures. For example, two figures shown in succession may, in fact, be executed substantially concurrently, or the figures may sometimes be executed in the reverse order, depending upon the functionality/acts involved.
The present invention is described in further detail below with reference to the attached drawing figures.
FIG. 1 shows a flow diagram of a method for predicting overdue risk ranking according to an embodiment of the application. The method includes step S1, step S2, and step S3.
Referring to fig. 1, in step S1, overdue feature information of a user to be evaluated is acquired.
Wherein the overdue feature information includes, but is not limited to, at least any one of:
1) the time of overdue occurrence;
2) the time of expiration; e.g., the number of months or days over which the term persists, etc.
3) An overdue amount;
4) the number of overdue times;
5) hastening the information; such as whether an unlink occurred, the number of times an unlink condition occurred, etc.
According to one embodiment, for a user who is not overdue, the overdue feature information may further include various other information that may be used to indicate a potential overdue risk, such as a user's academic history, and the like.
In step S2, overdue feature information of the user to be evaluated is input to the risk ranking model.
In step S3, a ranking of overdue risks for the user to be assessed is predicted based on the output of the risk ranking model.
Wherein the output of the risk ranking model is used to indicate a ranking of overdue risks for the user to be assessed.
According to one embodiment, the method builds or trains the risk ranking model by performing steps S4 and S5.
In step S4, overdue feature information of a plurality of users is acquired as a training sample set.
In step S5, for any two samples in the training sample set, training is performed based on the risk ranking relationship between the two samples to build or train a risk ranking model.
Specifically, based on a predetermined ranking rule, the order of risk ranking between the two samples is determined. Then, training is performed based on the respective ranking order of the two samples to build or train the risk ranking model. And, comparing every two of all training samples in the same manner, thereby training the risk ranking model until a predetermined training end condition is satisfied.
Where the sort rules may be formulated based on one or more items of overdue feature information, e.g., specifying that the longer the duration of overdue, the higher its corresponding risk of overdue, the greater the amount of overdue, and so on.
For another example, for a user who does not have overdue or does not have serious overdue, other reference information is used as overdue feature information. For example, a user with a high degree of scholarness is given a lower degree of overdue risk relative to other users with a lower degree of scholarness. As another example, users with higher loan requirements are specified to have a lower corresponding risk of overdue, and so on.
It should be noted that those skilled in the art will be familiar with the training of various loss functions with ranking properties, and those skilled in the art can select an appropriate loss function for the training of the risk ranking model based on actual needs.
According to an embodiment, the output of the risk ranking model is a ranking value indicating an overdue risk order of the user compared to other users, the step S5 further includes the step S501 and the step S502, and the step S3 further includes the step S301.
After step S4 is performed, in step S501, the respective ranking values of the two samples are determined based on a predetermined ranking rule.
Then, in step S502, training is performed based on the respective ranking values of the two samples to build or train the risk ranking model.
For example, a numerical magnitude of the ordering value y, which defines the ordering of the samples, is used to measure the orderliness. If sample 1 is expected to be ordered before sample 2, then y for sample 1 is defined to be greater than y for sample 2. Also, assuming that a predetermined sort rule calculates a sort value based on an expiration time, the sort value is defined to be equal to 12 minus the number of expired months, e.g., 12-1 to 11 for 1-month-expired y-value and 12-2 to 10 for 2-month-expired y-value.
The input feature vectors corresponding to the two samples X1 and X2 are respectively represented as X1 and X2, the established risk ranking model is represented as f, and the minimization loss function used in the training process is represented as:
max(0,-sign(y1-y2)*(f(x1)-f(x2)+margin)
where y1 and y2 represent the rank values corresponding to the two samples, f (x1) and f (x2) represent the model outputs corresponding to the two samples, and margin represents the parameter for setting the minimum segmentation boundary. The loss function-s ign (y1-y2) ((x 1) -f (x2)) can be replaced by other loss functions with ordering properties, such as-s ign (y1-y2) log (f (x1) -f (x 2)).
It should be noted that the above examples are only for better illustrating the technical solutions of the present invention, and not for limiting the present invention, and those skilled in the art should understand that any implementation manner of training based on the respective ranking values of two samples to establish the risk ranking model is included in the scope of the present invention.
Continuing with the description of the foregoing embodiment, when the established risk ranking model needs to be used, the overdue feature information of the user to be evaluated is obtained by executing step S1 and step S2, and the overdue feature information of the user to be evaluated is input to the preset risk ranking model. Next, in step S301, the ranking of overdue risks of the user to be evaluated is predicted based on the output ranking values of the risk ranking model.
The method for predicting the ranking of overdue risks of the user to be evaluated based on the output ranking value of the risk ranking model includes, but is not limited to, any one of the following:
1) obtaining the probability that one user has a higher overdue risk than another user based on a risk ranking model;
specifically, for a first user and a second user to be evaluated, the risk ranking model is used to obtain their respective corresponding ranking values. And then, calculating the difference of the ranking values of the first user and the second user to obtain the probability that the first user has higher overdue risk relative to the second user.
For example, assuming that F represents the risk ranking model for user a and user B, the output of the input of the features of user a and user B to the risk ranking model is represented as F (xa) and F (xb), respectively. By calculating the difference f (xa) -f (xb), the probability that user a has a higher risk of overdue than user B can be obtained.
2) Based on the risk ranking model, obtaining the probability that a user has higher overdue risk than a preset risk benchmark;
specifically, the risk ranking model is used to obtain a ranking value corresponding to the user. Then, by calculating the difference between the user and the risk benchmark, the probability that the user has a higher overdue risk than the risk benchmark is obtained.
For example, for user a in the above example, assuming that the value of the predetermined criterion is expressed as threshold1, by calculating the difference f (xa) -threshold1, the probability that user a has a higher risk of overdue than the predetermined criterion can be obtained.
Preferably, the method according to the present embodiment further includes step S6 and step S7.
In step S6, the correspondence between the pre-stored ranking value ranges and the corresponding loan operations is acquired.
In step S7, a loan operation corresponding to the ranking value is determined according to the ranking value corresponding to the user to be assessed and the correspondence.
The loan operations include various operations related to the issuance of loans.
For example, if the rank value is lower than a predetermined threshold, the corresponding relationship may be set to refuse to offer a loan to the corresponding user. As another example, the correspondence may include suggested loan amounts corresponding to different ranked value ranges.
According to the method, the risk ranking model is established and trained on the basis of the pairwise risk ranking relation of the training samples, so that overdue risk ranking of the users is predicted through the risk ranking model, the model can judge overdue risk ranking of different users on the basis of differences of overdue behaviors such as overdue time and the like, and compared with a traditional risk control model, prediction accuracy is improved.
Fig. 2 shows a schematic structural diagram of an apparatus for predicting overdue risk ranking according to an embodiment of the present application. The device comprises: the system comprises a device for acquiring overdue characteristic information of a user to be evaluated (hereinafter referred to as an acquiring device 1), a device for inputting the overdue characteristic information of the user to be evaluated into a preset risk ranking model (hereinafter referred to as an input device 2), and a device for predicting the ranking of the overdue risk of the user to be evaluated based on the output of the risk ranking model (hereinafter referred to as a prediction device 3).
Referring to fig. 2, the acquiring apparatus 1 acquires overdue feature information of a user to be evaluated.
Wherein the overdue feature information includes, but is not limited to, at least any one of:
1) the time of overdue occurrence;
2) the time of expiration; e.g., the number of months or days over which the term persists, etc.
3) An overdue amount;
4) the number of overdue times;
5) hastening the information; such as whether an unlink occurred, the number of times an unlink condition occurred, etc.
According to one embodiment, for a user who is not overdue, the overdue feature information may further include various other information that may be used to indicate a potential overdue risk, such as a user's academic history, and the like.
The input device 2 inputs overdue characteristic information of the user to be evaluated to the risk ranking model.
The prediction means 3 predicts the ranking of overdue risks for the user to be assessed based on the output of the risk ranking model.
Wherein the output of the risk ranking model is used to indicate a ranking of overdue risks for the user to be assessed.
According to one embodiment, the apparatus comprises: the risk ranking model building method includes a means for acquiring overdue feature information of a plurality of users as a training sample set (hereinafter referred to as "sample acquiring means"), and a means for training any two samples in the training sample set based on a risk ranking relationship between the two samples to build or train a risk ranking model (hereinafter referred to as "modeling means").
The sample acquisition device acquires overdue feature information of a plurality of users as a training sample set.
The modeling device trains any two samples in the training sample set based on the risk ranking relation between the two samples so as to build or train a risk ranking model.
Specifically, the modeling means determines the order of risk ranking between the two samples based on a predetermined ranking rule. Then, training is performed based on the respective ranking order of the two samples to build or train the risk ranking model. And, comparing every two of all training samples in the same manner, thereby training the risk ranking model until a predetermined training end condition is satisfied.
Where the sort rules may be formulated based on one or more items of overdue feature information, e.g., specifying that the longer the duration of overdue, the higher its corresponding risk of overdue, the greater the amount of overdue, and so on.
For another example, for a user who does not have overdue or does not have serious overdue, other reference information is used as overdue feature information. For example, a user with a high degree of scholarness is given a lower degree of overdue risk relative to other users with a lower degree of scholarness. As another example, users with higher loan requirements are specified to have a lower corresponding risk of overdue, and so on.
It should be noted that those skilled in the art will be familiar with the training of various loss functions with ranking properties, and those skilled in the art can select an appropriate loss function for the training of the risk ranking model based on actual needs.
According to one embodiment, the output of the risk ranking model is a ranking value indicating an order of overdue risk for a user compared to other users.
After the sample acquiring device executes the operation, the modeling device determines respective sorting values of the two samples based on a predetermined sorting rule. Then, the modeling device trains based on the respective ranking values of the two samples to build or train the risk ranking model.
When the established risk ranking model needs to be used, the overdue feature information of the user to be evaluated is acquired by executing the operation of the acquisition device 1 and the input device 2, and the overdue feature information of the user to be evaluated is input into the preset risk ranking model. Then, the prediction means 3 predicts the ranking of the overdue risk of the user to be evaluated based on the output ranking value of the risk ranking model.
The method for predicting the ranking of overdue risks of the user to be evaluated based on the output ranking value of the risk ranking model includes, but is not limited to, any one of the following:
1) obtaining the probability that one user has a higher overdue risk than another user based on a risk ranking model;
specifically, for a first user and a second user to be evaluated, the risk ranking model is used to obtain their respective corresponding ranking values. And then, calculating the difference of the ranking values of the first user and the second user to obtain the probability that the first user has higher overdue risk relative to the second user.
For example, assuming that F represents the risk ranking model for user a and user B, the output of the input of the features of user a and user B to the risk ranking model is represented as F (xa) and F (xb), respectively. By calculating the difference f (xa) -f (xb), the probability that user a has a higher risk of overdue than user B can be obtained.
2) Based on the risk ranking model, obtaining the probability that a user has higher overdue risk than a preset risk benchmark;
specifically, the risk ranking model is used to obtain a ranking value corresponding to the user. Then, by calculating the difference between the user and the risk benchmark, the probability that the user has a higher overdue risk than the risk benchmark is obtained.
For example, for user a in the above example, assuming that the value of the predetermined criterion is expressed as threshold1, by calculating the difference f (xa) -threshold1, the probability that user a has a higher risk of overdue than the predetermined criterion can be obtained.
Preferably, the apparatus according to the present embodiment further includes means for acquiring a correspondence relationship between a pre-stored ranking value range and a corresponding loan operation (hereinafter referred to as "relationship acquisition means"), and means for determining a loan operation corresponding to a ranking value to be evaluated, based on the ranking value corresponding to the user and the correspondence relationship (hereinafter referred to as "operation determination means").
The relationship acquisition means acquires a correspondence relationship between a pre-stored ranking value range and a corresponding loan operation.
And the operation determining device determines loan operation corresponding to the ranking value according to the ranking value corresponding to the user to be evaluated and the corresponding relation.
The loan operations include various operations related to the issuance of loans.
For example, if the rank value is lower than a predetermined threshold, the corresponding relationship may be set to refuse to offer a loan to the corresponding user. As another example, the correspondence may include suggested loan amounts corresponding to different ranked value ranges.
According to the scheme of the embodiment of the application, the risk ranking model is established and trained on the basis of the pairwise risk ranking relation of the training samples, so that overdue risk ranking of the users is predicted through the risk ranking model, the model can judge overdue risk ranking of different users on the basis of differences of overdue behaviors such as overdue time and the like, and compared with a traditional risk control model, the prediction accuracy is improved.
The software program of the present invention can be executed by a processor to implement the steps or functions described above. Also, the software programs (including associated data structures) of the present invention can be stored in a computer readable recording medium, such as RAM memory, magnetic or optical drive or diskette and the like. Additionally, some of the steps or functionality of the present invention may be implemented in hardware, for example, as circuitry that cooperates with the processor to perform various functions or steps.
In addition, some of the present invention can be applied as a computer program product, such as computer program instructions, which when executed by a computer, can invoke or provide the method and/or technical solution according to the present invention through the operation of the computer. Program instructions which invoke the methods of the present invention may be stored on a fixed or removable recording medium and/or transmitted via a data stream on a broadcast or other signal-bearing medium and/or stored within a working memory of a computer device operating in accordance with the program instructions. An embodiment according to the invention herein comprises an apparatus comprising a memory for storing computer program instructions and a processor for executing the program instructions, wherein the computer program instructions, when executed by the processor, trigger the apparatus to perform a method and/or solution according to embodiments of the invention as described above.
It will be evident to those skilled in the art that the invention is not limited to the details of the foregoing illustrative embodiments, and that the present invention may be embodied in other specific forms without departing from the spirit or essential attributes thereof. The present embodiments are therefore to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference sign in a claim should not be construed as limiting the claim concerned. Furthermore, it is obvious that the word "comprising" does not exclude other elements or steps, and the singular does not exclude the plural. A plurality of units or means recited in the system claims may also be implemented by one unit or means in software or hardware. The terms first, second, etc. are used to denote names, but not any particular order.
Claims (13)
1. A method for predicting overdue risk ranking, wherein the method comprises the steps of:
acquiring overdue characteristic information of a user to be evaluated;
inputting overdue characteristic information of the user to be evaluated into a preset risk sorting model;
predicting a ranking of the overdue risk of the user to be evaluated based on an output of a risk ranking model, wherein the output of the risk ranking model is used for indicating the ranking of the overdue risk of the user to be evaluated.
2. The method of claim 1, wherein the method further comprises:
obtaining overdue feature information of a plurality of users as a training sample set;
for any two samples in the training sample set, training is performed based on the risk ranking relationship between the two samples to establish or train a risk ranking model.
3. The method of claim 1 or 2, wherein the output of the risk ranking model is a ranking value indicating an order of overdue risk for a user compared to other users, the training based on the risk ranking relationship between the two samples to build or train a risk ranking model further comprises:
determining respective sorting values of the two samples based on a predetermined sorting rule;
training based on respective ranking values of the two samples to establish or train the risk ranking model;
wherein the step of predicting a ranking of overdue risks for the user to be assessed based on the output of the risk ranking model comprises:
and predicting the ordering of the overdue risks of the user to be evaluated based on the output ordering value of the risk ordering model.
4. The method of claim 1, wherein the predicting a ranking of overdue risk for a user to be assessed based on the ranked values of the output of the risk ranking model comprises:
for a first user and a second user to be evaluated, obtaining respective corresponding ranking values by using the risk ranking model;
and calculating the difference of the ranking values of the first user and the second user to obtain the probability that the first user has higher overdue risk relative to the second user.
5. The method of claim 3, wherein the method comprises:
acquiring a corresponding relation between a pre-stored ranking value range and a corresponding loan operation;
and determining loan operation corresponding to the ranking value according to the ranking value corresponding to the user to be evaluated and the corresponding relation.
6. The method of claim 1, wherein the overdue feature information comprises:
the time of overdue occurrence;
time of overdue duration
An overdue amount;
the number of overdue times;
and (5) hastening information collection.
7. An apparatus for predicting overdue risk ranking, wherein the apparatus comprises:
means for obtaining overdue feature information of a user to be assessed;
means for inputting overdue feature information of the user to be assessed to a preset risk ranking model;
means for predicting a ranking of the overdue risk of the user to be assessed based on an output of a risk ranking model, wherein the output of the risk ranking model is indicative of the ranking of the overdue risk of the user to be assessed.
8. The apparatus of claim 7, wherein the apparatus further comprises:
the device is used for acquiring overdue characteristic information of a plurality of users as a training sample set;
and training any two samples in the training sample set based on the risk ranking relation between the two samples to establish or train a risk ranking model.
9. The apparatus of claim 7 or 8, wherein the output of the risk ranking model is a ranking value indicating an order of overdue risk of a user compared to other users, and the means for training any two samples in the training sample set based on a risk ranking relationship between the two samples to establish or train the risk ranking model is to:
determining respective sorting values of the two samples based on a predetermined sorting rule;
training based on respective ranking values of the two samples to establish or train the risk ranking model;
wherein the means for predicting a ranking of overdue risk for the user to be assessed based on an output of a risk ranking model is to:
and predicting the ordering of the overdue risks of the user to be evaluated based on the output ordering value of the risk ordering model.
10. The apparatus of claim 9, wherein predicting a ranking of overdue risk for a user to be assessed based on the ranked values of the output of the risk ranking model comprises:
for a first user and a second user to be evaluated, obtaining respective corresponding ranking values by using the risk ranking model;
and calculating the difference of the ranking values of the first user and the second user to obtain the probability that the first user has higher overdue risk relative to the second user.
11. The apparatus of claim 9, wherein the apparatus comprises:
means for obtaining a correspondence between a pre-stored rank value range and a corresponding loan operation;
and the device is used for determining loan operation corresponding to the ranking value according to the ranking value corresponding to the user to be evaluated and the corresponding relation.
12. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the method according to any of claims 1 to 6 when executing the program.
13. A computer-readable storage medium, on which a computer program is stored, which program, when being executed by a processor, carries out the method according to any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110731639.7A CN113628026A (en) | 2021-06-30 | 2021-06-30 | Method and device for predicting overdue risk ranking |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110731639.7A CN113628026A (en) | 2021-06-30 | 2021-06-30 | Method and device for predicting overdue risk ranking |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113628026A true CN113628026A (en) | 2021-11-09 |
Family
ID=78378631
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110731639.7A Pending CN113628026A (en) | 2021-06-30 | 2021-06-30 | Method and device for predicting overdue risk ranking |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113628026A (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108694673A (en) * | 2018-05-16 | 2018-10-23 | 阿里巴巴集团控股有限公司 | A kind of processing method, device and the processing equipment of insurance business risk profile |
| CN110070430A (en) * | 2019-03-12 | 2019-07-30 | 平安科技（深圳）有限公司 | Assess method and device, the storage medium, electronic equipment of refund risk |
| CN110533520A (en) * | 2019-06-06 | 2019-12-03 | 上海凯京信达科技集团有限公司 | A kind of ranking method of the individual customer overdue loan grade based on multi-model |
| CN112785086A (en) * | 2021-02-10 | 2021-05-11 | 中国工商银行股份有限公司 | Credit overdue risk prediction method and device |
-
2021
- 2021-06-30 CN CN202110731639.7A patent/CN113628026A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108694673A (en) * | 2018-05-16 | 2018-10-23 | 阿里巴巴集团控股有限公司 | A kind of processing method, device and the processing equipment of insurance business risk profile |
| CN110070430A (en) * | 2019-03-12 | 2019-07-30 | 平安科技（深圳）有限公司 | Assess method and device, the storage medium, electronic equipment of refund risk |
| CN110533520A (en) * | 2019-06-06 | 2019-12-03 | 上海凯京信达科技集团有限公司 | A kind of ranking method of the individual customer overdue loan grade based on multi-model |
| CN112785086A (en) * | 2021-02-10 | 2021-05-11 | 中国工商银行股份有限公司 | Credit overdue risk prediction method and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Schlegel et al. | Towards a rigorous evaluation of XAI methods on time series | |
| CN111124840B (en) | Method and device for predicting alarm in business operation and maintenance and electronic equipment | |
| CN110009171B (en) | User behavior simulation method, device, equipment and computer readable storage medium | |
| US20190251458A1 (en) | System and method for particle swarm optimization and quantile regression based rule mining for regression techniques | |
| CN111369299B (en) | Identification method, device, equipment and computer readable storage medium | |
| CN108536784B (en) | Comment information sentiment analysis method and device, computer storage medium and server | |
| Friederich et al. | Towards data-driven reliability modeling for cyber-physical production systems | |
| CN113010389A (en) | Training method, fault prediction method, related device and equipment | |
| CN111160959B (en) | User click conversion prediction method and device | |
| US20220214957A1 (en) | Machine learning models applied to interaction data for facilitating modifications to online environments | |
| CN111373406A (en) | Accelerated simulation setup procedure using a priori knowledge extraction for problem matching | |
| CN111143517B (en) | Human selection label prediction method, device, equipment and storage medium | |
| CN113268403A (en) | Time series analysis and prediction method, device, equipment and storage medium | |
| Caesarita et al. | Identifying bottlenecks and fraud of business process using alpha++ and heuristic miner algorithms (Case study: CV. Wicaksana Artha) | |
| CN115730947A (en) | Bank customer loss prediction method and device | |
| CN115796312A (en) | Multivariable time series prediction method and system | |
| CN111159241A (en) | Click conversion estimation method and device | |
| CN113159441A (en) | Prediction method and device for implementation condition of banking business project | |
| CN115393030A (en) | Model training method and device for risk prediction | |
| CN111967774B (en) | Software quality risk prediction method and device | |
| CN113628026A (en) | Method and device for predicting overdue risk ranking | |
| CN114742644A (en) | Method and device for training multi-scene wind control system and predicting business object risk | |
| Zouari et al. | Towards an adaptive curation services composition based on machine learning | |
| CN111767959A (en) | Method and device for classifying pile fibers | |
| CN111126694A (en) | Time series data prediction method, system, medium and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20211209 Address after: 100193 Room 606, 6 / F, building 4, West District, courtyard 10, northwest Wangdong Road, Haidian District, Beijing Applicant after: Du Xiaoman Technology (Beijing) Co.,Ltd. Address before: 401120 b7-7-2, Yuxing Plaza, No.5, Huangyang Road, Yubei District, Chongqing Applicant before: Chongqing duxiaoman Youyang Technology Co.,Ltd. |
|
| TA01 | Transfer of patent application right |