CN113159917A - Information prediction method, device and storage medium - Google Patents

Information prediction method, device and storage medium Download PDF

Info

Publication number
CN113159917A
CN113159917A CN202110371166.4A CN202110371166A CN113159917A CN 113159917 A CN113159917 A CN 113159917A CN 202110371166 A CN202110371166 A CN 202110371166A CN 113159917 A CN113159917 A CN 113159917A
Authority
CN
China
Prior art keywords
predicted
credit
attribute information
information
determining
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202110371166.4A
Other languages
Chinese (zh)
Inventor
郭豪
蔡准
孙悦
郭晓鹏
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beijing Trusfort Technology Co ltd
Original Assignee
Beijing Trusfort Technology Co ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Beijing Trusfort Technology Co ltd filed Critical Beijing Trusfort Technology Co ltd
Priority to CN202110371166.4A priority Critical patent/CN113159917A/en
Publication of CN113159917A publication Critical patent/CN113159917A/en
Pending legal-status Critical Current

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
    • 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
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
    • G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/067—Enterprise or organisation modelling

Landscapes

  • Business, Economics & Management (AREA)
  • Engineering & Computer Science (AREA)
  • Human Resources & Organizations (AREA)
  • Strategic Management (AREA)
  • Economics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Development Economics (AREA)
  • Theoretical Computer Science (AREA)
  • Marketing (AREA)
  • Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Educational Administration (AREA)
  • Game Theory and Decision Science (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • Finance (AREA)
  • Accounting & Taxation (AREA)
  • Technology Law (AREA)
  • Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)

Abstract

The invention discloses an information prediction method, an information prediction device and a storage medium, wherein the method comprises the following steps: acquiring attribute information of an object to be predicted, and determining credit information of the object to be predicted in each object set and matching probability of the object to be predicted and each object set according to the attribute information so as to predict the credit level of the object to be predicted according to the matching probability and the credit information. Therefore, the matching probability of the object to be predicted and the object set is fully considered, the object to be predicted is effectively distinguished, the credit level of the object to be predicted is predicted more accurately according to the credit information of the object to be predicted in each object set, and the accuracy of the prediction result is obviously improved.

Description

Information prediction method, device and storage medium
Technical Field
The present invention relates to the field of internet technologies, and in particular, to an information prediction method, an information prediction apparatus, and a computer-readable storage medium.
Background
With the development of internet technology and the expansion of financial service business, credit borrowing is greatly promoted. However, credit loan brings convenience to people, and meanwhile, certain loopholes exist, which brings opportunities to some people with no special interest, and the cases of loan are frequently rare. Therefore, how to effectively process the credit loan information so as to effectively improve the risk control level of the financial industry becomes an urgent problem to be solved.
Disclosure of Invention
In order to solve the above problems in the process of predicting credit loan information, embodiments of the present invention creatively provide an information prediction method, an information prediction apparatus, and a computer-readable storage medium.
According to a first aspect of the present invention, there is provided an information prediction method, the method comprising: acquiring attribute information of an object to be predicted; according to the attribute information, determining credit information of the object to be predicted in each object set; determining the matching probability of the object to be predicted and each object set according to the attribute information; and predicting the credit level of the object to be predicted according to the matching probability and the credit information.
According to an embodiment of the present invention, before determining, according to the attribute information, a matching probability of the object to be predicted with each object set and a credit score probability of the object to be predicted in each object set, the method further includes: and converting continuous variables in the attribute information into discrete variables.
According to an embodiment of the present invention, the converting the continuous variable in the attribute information into the discrete variable includes: and performing box separation treatment on the continuous variable.
According to an embodiment of the present invention, before determining the probability of credit score of the object to be predicted in each of the object sets according to the attribute information, the method further includes: acquiring sample attribute information of a plurality of sample objects; performing feature extraction on the sample attribute information to obtain a feature extraction result; training a basic model according to the feature extraction result to determine the credit score probability of the object to be predicted in each object set according to the attribute information; wherein, according to the feature extraction result, training a basic model comprises: and determining a logistic regression parameter of the basic model by using a logistic regression algorithm according to the feature extraction result so as to obtain the basic model.
According to an embodiment of the present invention, the determining credit information of the object to be predicted in each object set according to the attribute information includes: determining a prediction ratio logarithm according to the logistic regression parameter; determining a base credit and a credit coefficient of the object to be predicted in each object set according to the prediction ratio logarithm; and determining the credit information according to the basic credit and the credit coefficient.
According to an embodiment of the invention, the method further comprises: and training an information prediction model by using a logistic regression algorithm and a loss function according to the set number of the object sets, the feature extraction result and the logistic regression parameters so as to determine the matching probability of the object to be predicted and each object set according to the attribute information.
According to a second aspect of the present invention, there is provided an information prediction apparatus, the apparatus comprising: the acquisition module is used for acquiring attribute information of an object to be predicted; the credit determining module is used for determining the credit information of the object to be predicted in each object set according to the attribute information; the matching module is used for determining the matching probability of the object to be predicted and each object set according to the attribute information; and the prediction module is used for predicting the credit level of the object to be predicted according to the matching probability and the credit information.
According to an embodiment of the invention, the apparatus further comprises: the sample acquisition module is used for acquiring sample attribute information of a plurality of sample objects before determining the credit score probability of the object to be predicted in each object set according to the attribute information; the characteristic extraction module is used for extracting the characteristics of the sample attribute information to obtain a characteristic extraction result; and the first training module is used for training a basic model according to the feature extraction result so as to determine the credit score probability of the object to be predicted in each object set according to the attribute information.
According to an embodiment of the invention, the apparatus further comprises: and the second training module is used for training an information prediction model by using a logistic regression algorithm and a loss function according to the set number of the object sets, the feature extraction result and the logistic regression parameters so as to determine the matching probability of the object to be predicted and each object set according to the attribute information.
According to a third aspect of the present invention, there is also provided a computer-readable storage medium comprising a set of computer-executable instructions which, when executed, are operable to perform any of the information prediction methods described above.
The information prediction method, the device and the storage medium of the embodiment of the invention acquire the attribute information of the object to be predicted, and determine the credit information of the object to be predicted in each object set and the matching probability of the object to be predicted and each object set according to the attribute information, so as to predict the credit level of the object to be predicted according to the matching probability and the credit information. Therefore, the matching probability of the object to be predicted and the object set is fully considered, the object to be predicted is effectively distinguished, the credit level of the object to be predicted is predicted more accurately according to the credit information of the object to be predicted in each object set, and the accuracy of the prediction result is obviously improved.
It is to be understood that the teachings of the present invention need not achieve all of the above-described benefits, but rather that specific embodiments may achieve specific technical results, and that other embodiments of the present invention may achieve benefits not mentioned above.
Drawings
The above and other objects, features and advantages of exemplary embodiments of the present invention will become readily apparent from the following detailed description read in conjunction with the accompanying drawings. Several embodiments of the invention are illustrated by way of example, and not by way of limitation, in the figures of the accompanying drawings and in which:
in the drawings, the same or corresponding reference numerals indicate the same or corresponding parts.
FIG. 1 is a flow chart illustrating an implementation of an information prediction method according to an embodiment of the present invention;
FIG. 2 is a diagram illustrating a comparison between a single card mode and a multi-card mode in an application example of the information prediction method according to the embodiment of the present invention;
fig. 3 is a schematic diagram showing the configuration of the information prediction apparatus according to the present invention.
Detailed Description
The principles and spirit of the present invention will be described with reference to a number of exemplary embodiments. It is understood that these embodiments are given only to enable those skilled in the art to better understand and to implement the present invention, and do not limit the scope of the present invention in any way. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art.
The technical solution of the present invention is further elaborated below with reference to the drawings and the specific embodiments.
Fig. 1 shows a schematic flow chart of an implementation of an information prediction method according to an embodiment of the present invention.
Referring to fig. 1, an information prediction method according to an embodiment of the present invention at least includes the following operation flows: operation 101, acquiring attribute information of an object to be predicted; operation 102, determining credit information of the object to be predicted in each object set according to the attribute information; operation 103, determining a matching probability between the object to be predicted and each object set according to the attribute information; and operation 104, predicting the credit level of the object to be predicted according to the matching probability and the credit information.
In operation 101, attribute information of an object to be predicted is acquired.
In this embodiment of the present invention, the object to be predicted may be a user, and the user may be an individual user, an enterprise user, or another suitable object to be predicted. The attribute information is information for characterizing a user. For example: for the age, gender, academic calendar, bank flow information, housing, credit amount, number of credit cards, age, nature of work, etc. of the individual user.
In operation 102, credit information of the object to be predicted in each object set is determined according to the attribute information.
In this embodiment of the present invention, the credit information of the object to be predicted in each object set may be determined by using a general scoring card, where the scoring card includes a plurality of object attributes and a weight of each object attribute. For example, the score card of the first object set includes an object attribute a, an object attribute B, an object attribute C, an object attribute D, and an object attribute E, and each object attribute is assigned with a weight, which may be shown in the form of a score or a percentage, or the like. For example: the weights may be shown in the form of scores, with scores for object property a, object property B, object property C, object property D, and object property E being 20, 35, 25, 15, and 5, respectively, and a total score for each object property in the first set of objects being 100. And the corresponding scores of the object to be predicted under different attributes in the first object set are respectively 18, 30, 15, 13 and 5, and the credit information of the object to be predicted in the first object set is 81.
Here, each set of objects can characterize a subspace having a common specific object population, for example: in the financial field, different scoring cards can be respectively set for student groups, urban white-collar workers, new-generation agronomy workers, entrepreneurs and the like in the process of predicting the credit level of a user. The larger the value of the credit information of the object to be predicted under the score of the object set is, the higher the credit level of the object to be predicted obtained from the evaluation criterion of the object set is.
In this embodiment of the present invention, before determining the matching probability of the object to be predicted and each object set and the credit score probability of the object to be predicted in each object set according to the attribute information, the continuous variable in the attribute information is also converted into a discrete variable. For example, the continuous variable in the attribute information may be converted into a discrete variable by performing binning processing on the continuous variable. The strategy of the box separation processing comprises equal frequency box separation, equal width box separation and the like, and continuous variables are converted into discrete variables.
In this embodiment of the present invention, before determining the probability of the credit score of the object to be predicted in each object set according to the attribute information, the base model is first trained by adopting the following operation steps: acquiring sample attribute information of a plurality of sample objects; carrying out feature extraction on the sample attribute information to obtain a feature extraction result; and training a basic model according to the feature extraction result so as to determine the credit score probability of the object to be predicted in each object set according to the attribute information. For training the basic model according to the feature extraction result, a logistic regression algorithm can be used to determine logistic regression parameters of the basic model according to the feature extraction result to obtain the basic model.
For example, the following training procedure may be used to configure scoring cards for each set of subjects.
First, a continuous variable in sample attribute information of a sample object is subjected to binning processing. For example: bank running information of the credit applicant, and the like. Then, the discrete variables generated after the binning process and the discrete variables in the original sample attribute information are subjected to WOE (Weight of Evidence) encoding process using the following formula (1).
Figure RE-GDA0003089678060000051
Wherein p is1Represents the proportion of samples with label of 1 in the samples with corresponding attribute discrete values; p is a radical of0Represents the proportion of samples with label of 0 in the sample objects corresponding to the discrete values of the sample attributes; label y represents whether the sample object is overdue, y value of 1 represents that the sample object is overdue, and y value of 0 represents that the sample object is normal.
Next, the feature vector x after WOE encoding and the corresponding label y are input into a logistic regression formula of the following formula (2) to train a basic model, and a model parameter w in the following formula (2) is determined.
Figure RE-GDA0003089678060000061
Where w is a model parameter of LR (logistic regression algorithm), which may also be referred to herein as logistic regression parameter.
In this embodiment of the present invention, based on the basic model, the following operation steps may be adopted to determine the credit information of the object to be predicted in each object set according to the attribute information of the object to be predicted: and determining a prediction ratio logarithm according to the logistic regression parameters, then determining a basic credit and a credit coefficient of the object to be predicted in each object set according to the prediction ratio logarithm, and finally determining credit information according to the basic credit and the credit coefficient.
For example, after the basic model training is completed and the logistic regression parameter w is determined, the logarithm of the prediction ratio odd of the intermediate variable is introduced, and the logarithm of the prediction ratio odd is shown in the following formula (3).
Figure RE-GDA0003089678060000062
Thus, for a sample object, the predicted score of the scoring card: scoretotal ═ a + B odd, given the incremental score PDO when odd doubles: scoretotal + PDO ═ a + B × 2 odd. Then the basic score A and the score coefficient B can be obtained by solving the binary equation.
Finally, a scoring card final score for the object to be predicted is obtained according to the logistic regression result as shown in the following formula (4).
Scoretotal＝A+B(β0+β1WOE1+…+βnWOEn)＝A+B*β0+B*β1WOE1+...+B*βnWOEn
(4)
Wherein, B is beta0、B*β1WOE1……B*βnWOEnRepresenting the corresponding scores of the sample object under different object attributes.
In operation 103, according to the attribute information, a matching probability of the object to be predicted and each object set is determined.
In this embodiment of the present invention, the information prediction model is trained by using a logistic regression algorithm and a loss function according to the set number, the feature extraction result, and the logistic regression parameter of the object set, so as to determine the matching probability between the object to be predicted and each object set according to the attribute information.
For example, the transformation formula of the information prediction model is determined based on the basic model as shown in the following formula (5), where the information prediction model is a model that uses scoring cards of a plurality of object sets to comprehensively predict information of objects to be predicted, and thus may be referred to as a multi-card model:
Figure RE-GDA0003089678060000071
the variable K in the formula represents the number of the scoring cards and belongs to the hyper-parameters of the information prediction model, wherein one scoring card is configured for each object set, so that the number of the scoring cards is the same as the number of the object sets. In the practical application process, the setting can be carried out according to the practical situation, and the value range can be between [3 and 6 ].
In the above formula (5), LRMulti-card(x) Is composed of two items, wherein
Figure RE-GDA0003089678060000072
Representing the matching probability of a sample object belonging to the ith object set in the total K object sets; while
Figure RE-GDA0003089678060000073
It represents the credit information of the sample object in the ith set of objects. In view of the form of the formula (5),
Figure RE-GDA0003089678060000074
the logistic regression algorithm formula is consistent with the logistic regression algorithm formula depended by the scoring card of the basic model, and the logistic regression algorithm formula is referred to the formula (2), so that the logistic regression algorithm formula can be converted into the scoring card attribute information in different object sets. While
Figure RE-GDA0003089678060000075
The sample object is shown to depend on the weighting factor of the scoring card for each set of objects.
For the information prediction simulation, the solution of the model parameters can be performed by using a cross entropy loss function as shown in the following formula (6):
Figure RE-GDA0003089678060000076
and using yi as a label to identify whether the current sample object is a fraud sample, wherein parameters mu and w in the information prediction model can be trained and updated by using a gradient descent algorithm.
And operation 104, predicting the credit level of the object to be predicted according to the matching probability and the credit information.
For example, the object set includes an object set 1, an object set 2, and an object set 3, the matching probabilities of the object DX1 to be predicted with the object set 1, the object set 2, and the object set 3 are 60%, 30%, and 10%, respectively, and the credit levels of the object DX1 in the object set 1, the object set 2, and the object set 3 are 80, 75, and 90, respectively, so that the credit level of the finally determined object to be predicted is 60% × 80+ 30% × 75+ 10% × 90 — 79.5.
Here, a preset credit threshold may be set, and in the case that the credit level of the object to be predicted is less than the set threshold, it is determined that the object to be predicted is an object that is easy to expire or an object that is easy to generate fraud, etc.
Fig. 2 is a schematic diagram illustrating comparison between a single-card mode and a multi-card mode in an application example of the information prediction method according to the embodiment of the present invention.
Referring to fig. 2, a credit scoring method in a multi-card mode in an application example of the embodiment of the present application is described in a manner of comparing a single-card mode with a multi-card mode. As shown in fig. 2, the left side shows a scoring card diagram of a single card mode of a credit scoring card, and the right side shows a scoring mode diagram of a multi-card mode of credit scoring. Card corresponding to three different object sets1，Card2，Card3Three scoring cards are taken as examples, and a multi-card mode of credit scoring is explained.
α1、α2、α3And the weight coefficient represents the probability that the object to be predicted belongs to each object set, namely the weight coefficient of the object to be predicted and the score card of the corresponding object set. The value of alpha can be used in the information prediction model
Figure RE-GDA0003089678060000081
Thus obtaining the product.
The matching probability of the object to be predicted and each object set in the multi-card mode is calculated according to the information prediction model. Thus, the matching probability α of each object to be predicted to each object set is different. Therefore, the information prediction model is adopted to carry out self-adaptation aiming at different objects to be predicted, and the matching probability is determined, so that the information prediction model can effectively distinguish user groups, and a better credit level identification effect is achieved.
The information prediction method, the information prediction device and the storage medium of the embodiment of the invention acquire the attribute information of the object to be predicted, and determine the credit information of the object to be predicted in each object set and the matching probability of the object to be predicted and each object set according to the attribute information so as to predict the credit level of the object to be predicted according to the matching probability and the credit information. Therefore, the matching probability of the object to be predicted and the object set is fully considered, the object to be predicted is effectively distinguished, the credit level of the object to be predicted is predicted more accurately according to the credit information of the object to be predicted in each object set, and the accuracy of the prediction result is obviously improved.
Similarly, based on the above information prediction method, an embodiment of the present invention further provides a computer-readable storage medium, in which a program is stored, and when the program is executed by a processor, the processor is caused to perform at least the following operation steps: operation 101, acquiring attribute information of an object to be predicted; operation 102, determining credit information of the object to be predicted in each object set according to the attribute information; operation 103, determining a matching probability between the object to be predicted and each object set according to the attribute information; and operation 104, predicting the credit level of the object to be predicted according to the matching probability and the credit information.
Further, based on the above information prediction method, an embodiment of the present invention further provides an information prediction apparatus, as shown in fig. 3, where the apparatus 30 includes: an obtaining module 301, configured to obtain attribute information of an object to be predicted; the credit determining submodule 302 is configured to determine credit information of the object to be predicted in each object set according to the attribute information; the matching submodule 303 is configured to determine, according to the attribute information, a matching probability between the object to be predicted and each object set; and the credit prediction module 304 is used for predicting the credit level of the object to be predicted according to the matching probability and the credit information.
According to an embodiment of the present invention, the apparatus 30 further comprises: the sample acquisition module is used for acquiring sample attribute information of a plurality of sample objects before determining the credit score probability of the object to be predicted in each object set according to the attribute information; the characteristic extraction module is used for extracting the characteristics of the sample attribute information to obtain a characteristic extraction result; and the first training module is used for training the basic model according to the feature extraction result so as to determine the credit scoring probability of the object to be predicted in each object set according to the attribute information.
According to an embodiment of the present invention, the apparatus 30 further comprises: and the second training module is used for training the information prediction model by using a logistic regression algorithm and a loss function according to the set number, the feature extraction result and the logistic regression parameters of the object set so as to determine the matching probability of the object to be predicted and each object set according to the attribute information.
Here, it should be noted that: the above description of the embodiment of the information prediction apparatus is similar to the description of the method embodiment shown in fig. 1 to 2, and has similar beneficial effects to the method embodiment shown in fig. 1 to 2, and therefore, the description is omitted. For technical details not disclosed in the embodiment of the information prediction apparatus of the present invention, please refer to the description of the method embodiment shown in fig. 1 to 2 of the present invention for understanding, and therefore, for brevity, will not be described 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 apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
In the several embodiments provided in the present application, it should be understood that the disclosed apparatus and method may be implemented in other ways. The above-described device embodiments are merely illustrative, for example, the division of a unit is only one logical function division, and there may be other division ways in actual implementation, such as: multiple units or components may be combined, or may be integrated into another system, or some features may be omitted, or not implemented. In addition, the coupling, direct coupling or communication connection between the components shown or discussed may be through some interfaces, and the indirect coupling or communication connection between the devices or units may be electrical, mechanical or other forms.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units; can be located in one place or distributed on a plurality of network units; some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, all the functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may be separately regarded as one unit, or two or more units may be integrated into one unit; the integrated unit can be realized in a form of hardware, or in a form of hardware plus a software functional unit.
Those of ordinary skill in the art will understand that: all or part of the steps for realizing the method embodiments can be completed by hardware related to program instructions, the program can be stored in a computer readable storage medium, and the program executes the steps comprising the method embodiments when executed; and the aforementioned storage medium includes: various media that can store program codes, such as a removable Memory device, a Read Only Memory (ROM), a magnetic disk, or an optical disk.
Alternatively, the integrated unit of the present invention may be stored in a computer-readable storage medium if it is implemented in the form of a software functional module and sold or used as a separate product. Based on such understanding, the technical solutions of the embodiments of the present invention may be essentially implemented or a part contributing to the prior art may be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for enabling a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the methods of the embodiments of the present invention. And the aforementioned storage medium includes: a removable storage device, a ROM, a magnetic or optical disk, or other various media that can store program code.
The above description is only for the specific embodiments of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present invention, and the changes or substitutions should be covered within the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims (10)

1. An information prediction method, the method comprising:
acquiring attribute information of an object to be predicted;
according to the attribute information, determining credit information of the object to be predicted in each object set;
determining the matching probability of the object to be predicted and each object set according to the attribute information;
and predicting the credit level of the object to be predicted according to the matching probability and the credit information.
2. The method of claim 1, wherein before determining the matching probability of the object to be predicted with each set of objects and the credit score probability of the object to be predicted in each set of objects according to the attribute information, the method further comprises:
and converting continuous variables in the attribute information into discrete variables.
3. The method of claim 2, wherein converting the continuous variable in the attribute information into a discrete variable comprises: and performing box separation treatment on the continuous variable.
4. The method of claim 1, wherein before determining the probability of the credit score of the object to be predicted in each of the object sets according to the attribute information, the method further comprises:
acquiring sample attribute information of a plurality of sample objects;
performing feature extraction on the sample attribute information to obtain a feature extraction result;
training a basic model according to the feature extraction result to determine the credit score probability of the object to be predicted in each object set according to the attribute information;
wherein, according to the feature extraction result, training a basic model comprises: and determining a logistic regression parameter of the basic model by using a logistic regression algorithm according to the feature extraction result so as to obtain the basic model.
5. The method according to claim 4, wherein the determining credit information of the object to be predicted in each object set according to the attribute information comprises:
determining a prediction ratio logarithm according to the logistic regression parameter;
determining a base credit and a credit coefficient of the object to be predicted in each object set according to the prediction ratio logarithm;
and determining the credit information according to the basic credit and the credit coefficient.
6. The method of claim 4, further comprising:
and training an information prediction model by using a logistic regression algorithm and a loss function according to the set number of the object sets, the feature extraction result and the logistic regression parameters so as to determine the matching probability of the object to be predicted and each object set according to the attribute information.
7. An information prediction apparatus, characterized in that the apparatus comprises:
the acquisition module is used for acquiring attribute information of an object to be predicted;
the credit determining module is used for determining the credit information of the object to be predicted in each object set according to the attribute information;
the matching module is used for determining the matching probability of the object to be predicted and each object set according to the attribute information;
and the prediction module is used for predicting the credit level of the object to be predicted according to the matching probability and the credit information.
8. The apparatus of claim 7, further comprising:
the sample acquisition module is used for acquiring sample attribute information of a plurality of sample objects before determining the credit score probability of the object to be predicted in each object set according to the attribute information;
the characteristic extraction module is used for extracting the characteristics of the sample attribute information to obtain a characteristic extraction result;
and the first training module is used for training a basic model according to the feature extraction result so as to determine the credit score probability of the object to be predicted in each object set according to the attribute information.
9. The apparatus of claim 8, further comprising:
and the second training module is used for training an information prediction model by using a logistic regression algorithm and a loss function according to the set number of the object sets, the feature extraction result and the logistic regression parameters so as to determine the matching probability of the object to be predicted and each object set according to the attribute information.
10. A computer-readable storage medium comprising a set of computer-executable instructions that, when executed, perform the information prediction method of any one of claims 1-7.
CN202110371166.4A 2021-04-07 2021-04-07 Information prediction method, device and storage medium Pending CN113159917A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110371166.4A CN113159917A (en) 2021-04-07 2021-04-07 Information prediction method, device and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110371166.4A CN113159917A (en) 2021-04-07 2021-04-07 Information prediction method, device and storage medium

Publications (1)

Publication Number Publication Date
CN113159917A true CN113159917A (en) 2021-07-23

Family

ID=76888537

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110371166.4A Pending CN113159917A (en) 2021-04-07 2021-04-07 Information prediction method, device and storage medium

Country Status (1)

Country Link
CN (1) CN113159917A (en)

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110544155A (en) * 2019-09-02 2019-12-06 中诚信征信有限公司 User credit score acquisition method, acquisition device, server and storage medium
CN110909970A (en) * 2018-09-17 2020-03-24 北京京东金融科技控股有限公司 Credit scoring method and device
CN111080397A (en) * 2019-11-18 2020-04-28 支付宝(杭州)信息技术有限公司 Credit evaluation method and device and electronic equipment
CN111222982A (en) * 2020-01-16 2020-06-02 随手(北京)信息技术有限公司 Internet credit overdue prediction method, device, server and storage medium
CN111951093A (en) * 2020-08-05 2020-11-17 苏州好懿春数据服务有限公司 Personal credit score scoring method
CN112017040A (en) * 2020-10-16 2020-12-01 银联商务股份有限公司 Credit scoring model training method, scoring system, equipment and medium
CN112330048A (en) * 2020-11-18 2021-02-05 中国光大银行股份有限公司 Scoring card model training method and device, storage medium and electronic device
CN112598294A (en) * 2020-12-25 2021-04-02 四川云从天府人工智能科技有限公司 Method, device, machine readable medium and equipment for establishing scoring card model on line

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110909970A (en) * 2018-09-17 2020-03-24 北京京东金融科技控股有限公司 Credit scoring method and device
CN110544155A (en) * 2019-09-02 2019-12-06 中诚信征信有限公司 User credit score acquisition method, acquisition device, server and storage medium
CN111080397A (en) * 2019-11-18 2020-04-28 支付宝(杭州)信息技术有限公司 Credit evaluation method and device and electronic equipment
CN111222982A (en) * 2020-01-16 2020-06-02 随手(北京)信息技术有限公司 Internet credit overdue prediction method, device, server and storage medium
CN111951093A (en) * 2020-08-05 2020-11-17 苏州好懿春数据服务有限公司 Personal credit score scoring method
CN112017040A (en) * 2020-10-16 2020-12-01 银联商务股份有限公司 Credit scoring model training method, scoring system, equipment and medium
CN112330048A (en) * 2020-11-18 2021-02-05 中国光大银行股份有限公司 Scoring card model training method and device, storage medium and electronic device
CN112598294A (en) * 2020-12-25 2021-04-02 四川云从天府人工智能科技有限公司 Method, device, machine readable medium and equipment for establishing scoring card model on line

Similar Documents

Publication Publication Date Title
CN111291816B (en) Method and device for carrying out feature processing aiming at user classification model
CN108090800B (en) Game prop pushing method and device based on player consumption potential
CN107133865B (en) Credit score obtaining and feature vector value output method and device
CN111597348B (en) User image drawing method, device, computer equipment and storage medium
CN109739844B (en) Data classification method based on attenuation weight
EP3608802A1 (en) Model variable candidate generation device and method
CN111428217A (en) Method and device for identifying cheat group, electronic equipment and computer readable storage medium
CN112990386A (en) User value clustering method and device, computer equipment and storage medium
CN113706291A (en) Fraud risk prediction method, device, equipment and storage medium
CN115204985A (en) Shopping behavior prediction method, device, equipment and storage medium
WO2022157973A1 (en) Information processing system, information processing method, and computer program
CN115730125A (en) Object identification method and device, computer equipment and storage medium
CN113327132A (en) Multimedia recommendation method, device, equipment and storage medium
CN115841345B (en) Cross-border big data intelligent analysis method, system and storage medium
CN110619564B (en) Anti-fraud feature generation method and device
CN113159917A (en) Information prediction method, device and storage medium
CN113065892B (en) Information pushing method, device, equipment and storage medium
CN115757973A (en) Sorting method, device and equipment of recommendation system and readable storage medium
CN111859057B (en) Data feature processing method and data feature processing device
CN113706258A (en) Product recommendation method, device, equipment and storage medium based on combined model
CN113886697A (en) Clustering algorithm based activity recommendation method, device, equipment and storage medium
CN113420789A (en) Method, device, storage medium and computer equipment for predicting risk account
CN113313582A (en) Guest refusing and reflashing model training method and device and electronic equipment
CN111160929A (en) Method and device for determining client type
CN112132690A (en) Foreign exchange product information pushing method and device, computer equipment and 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