CN111275338A - Method, device, equipment and storage medium for judging enterprise fraud behaviors - Google Patents
Method, device, equipment and storage medium for judging enterprise fraud behaviors Download PDFInfo
- Publication number
- CN111275338A CN111275338A CN202010067142.5A CN202010067142A CN111275338A CN 111275338 A CN111275338 A CN 111275338A CN 202010067142 A CN202010067142 A CN 202010067142A CN 111275338 A CN111275338 A CN 111275338A
- Authority
- CN
- China
- Prior art keywords
- enterprise
- target
- value
- fraud
- target enterprise
- 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
- 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
- G06Q10/0635—Risk analysis of enterprise or organisation activities
-
- 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
- G06Q30/00—Commerce
- G06Q30/018—Certifying business or products
- G06Q30/0185—Product, service or business identity fraud
-
- 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
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0609—Buyer or seller confidence or verification
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Finance (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Development Economics (AREA)
- Accounting & Taxation (AREA)
- Marketing (AREA)
- Entrepreneurship & Innovation (AREA)
- Educational Administration (AREA)
- Game Theory and Decision Science (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application discloses a method for judging enterprise fraud, which comprises the following steps: acquiring target characteristic data of a target enterprise; inputting the target characteristic data into a pre-trained enterprise rating card model to obtain a target enterprise rating value of a target enterprise; and determining the fraud risk degree of the target enterprise according to the preset grading segment and the grading value of the target enterprise. Therefore, the method avoids manual offline depth investigation, can save a large amount of human resources, and improves the judgment efficiency; and the enterprise scoring card model trained through machine learning obtains the target enterprise scoring value of the target enterprise, so that the accuracy of enterprise fraud judgment is improved. The application also discloses a device and equipment for judging enterprise fraud behaviors and a computer readable storage medium, which have the beneficial effects.
Description
Technical Field
The present invention relates to the field of information determination, and in particular, to a method, an apparatus, a device, and a computer-readable storage medium for determining enterprise fraud.
Background
With the rapid development of the economic society, the number of small and micro enterprises is rapidly increased; meanwhile, when small and micro enterprises operate, the demand of applying for loan from financial institutions exists. For the financial institution, in order to ensure the economic safety of the financial institution, the credit condition of the small and micro enterprise needs to be checked, that is, the small and micro enterprise is determined to be fraudulent to determine whether to credit the small and micro enterprise.
In the prior art, the judgment of the enterprise fraud is generally carried out through deep investigation and analysis under a manual line, so that the efficiency is low, a large amount of human resources are consumed, and the judgment process is doped with artificial subjective factors, so that the judgment result is inaccurate.
Therefore, how to efficiently and accurately determine enterprise fraud is a technical problem that needs to be solved by those skilled in the art.
Disclosure of Invention
In view of the above, the present invention provides a method for determining enterprise fraud, which can efficiently and accurately determine enterprise fraud; another object of the present invention is to provide an apparatus, a device and a computer-readable storage medium for determining enterprise fraud, all of which have the above advantages.
In order to solve the above technical problem, the present invention provides a method for determining enterprise fraud, including:
acquiring target characteristic data of a target enterprise;
inputting the target characteristic data into a pre-trained enterprise rating card model to obtain a target enterprise rating value of the target enterprise;
and determining the fraud risk degree of the target enterprise according to the score value of the target enterprise according to preset score segments.
Preferably, the process of training out the enterprise rating card model specifically includes:
acquiring sample characteristic data of a plurality of enterprises; wherein the sample feature data comprises sample feature variables and corresponding sample variable values;
setting FP-Tree for the sample characteristic variable according to the support degree, the reliability and the action degree, and determining a mode entering variable;
setting corresponding mold entering indexes according to the mold entering variables, and calculating the weight of each mold entering index according to the variable value of each sample;
and setting a calculation mode of calculating enterprise scoring values corresponding to the enterprises according to the modeling indexes, the corresponding weights and the box dividing coefficients of the modeling indexes to obtain the enterprise scoring card model.
Preferably, the setting of the FP-Tree for the sample feature variable according to the support degree, the reliability degree, and the action degree, and determining the process of the model entry variable specifically include:
setting the FP-Tree for the sample characteristic variable according to the support degree, the credibility and the action degree, and determining a mode entering variable;
and screening the module-entering variables by using the KS value, the AR value, the IV value and the VIF value, and updating the module-entering variables by using the screened module-entering variables.
Preferably, after the obtaining sample characteristic data of a plurality of enterprises, the method further comprises:
and performing data cleaning and exception handling on the sample characteristic data.
Preferably, after the determining the fraud risk degree of the target enterprise according to the target enterprise scoring value according to the preset scoring segment index, the method further includes:
and if the fraud risk degree of the target enterprise is higher than the early warning threshold value, sending out corresponding prompt information.
Preferably, further comprising:
and recording the enterprise information of the target enterprise of which the fraud risk degree is higher than the early warning threshold value.
Preferably, further comprising:
and displaying the target characteristic data and/or the target enterprise scoring value and/or the fraud risk degree of the target enterprise through a webpage.
In order to solve the above technical problem, the present invention further provides an apparatus for determining enterprise fraud, including:
the acquisition module is used for acquiring target characteristic data of a target enterprise;
the input module is used for inputting the target characteristic data into a pre-trained enterprise rating card model to obtain a target enterprise rating value of the target enterprise;
and the determining module is used for determining the fraud risk degree of the target enterprise according to the target enterprise scoring value according to preset scoring segments.
In order to solve the above technical problem, the present invention further provides an enterprise fraud determination apparatus, including:
a memory for storing a computer program;
and the processor is used for implementing the steps of any enterprise fraud behavior judging method when executing the computer program.
In order to solve the above technical problem, the present invention further provides a computer-readable storage medium, where a computer program is stored, and the computer program, when executed by a processor, implements the steps of any one of the above methods for determining enterprise fraud.
According to the enterprise fraud behavior judging method provided by the invention, the target enterprise grade value of the target enterprise is obtained by inputting the acquired target characteristic data of the target enterprise into the enterprise grade card model through the enterprise grade card model trained in advance; and then, according to preset grading segmentation, determining the fraud risk degree of the target enterprise according to the grading value of the target enterprise. Therefore, the method avoids manual offline depth investigation, can save a large amount of human resources, and improves the judgment efficiency; and the enterprise scoring card model trained through machine learning obtains the target enterprise scoring value of the target enterprise, so that the accuracy of enterprise fraud judgment is improved.
In order to solve the technical problem, the invention also provides a device, equipment and a computer readable storage medium for determining enterprise fraud, which have the beneficial effects.
Drawings
In order to more clearly illustrate the embodiments or technical solutions of the present invention, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the provided drawings without creative efforts.
Fig. 1 is a flowchart of a method for determining enterprise fraud according to an embodiment of the present invention;
FIG. 2 is a flowchart of a method for training an enterprise rating card model according to an embodiment of the present invention;
fig. 3 is a block diagram of an apparatus for determining enterprise fraud according to an embodiment of the present invention;
fig. 4 is a block diagram of an enterprise fraud determination apparatus according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The core of the embodiment of the invention is to provide a method for judging enterprise fraud, which can efficiently and accurately judge the enterprise fraud; another core of the present invention is to provide an apparatus, a device and a computer-readable storage medium for determining enterprise fraud, all of which have the above advantages.
In order that those skilled in the art will better understand the disclosure, the invention will be described in further detail with reference to the accompanying drawings and specific embodiments.
Fig. 1 is a flowchart of a method for determining enterprise fraud according to an embodiment of the present invention. As shown in fig. 1, a method for determining enterprise fraud includes:
s10: and acquiring target characteristic data of the target enterprise.
Specifically, in this embodiment, first, target feature data of a target enterprise, where fraud of the enterprise needs to be determined, is obtained, where the target feature data refers to obtaining a corresponding variable value according to a predetermined feature variable, that is, the target feature data includes a target feature variable of the target enterprise and a corresponding target variable value. It should be noted that, in actual operation, after legal authorization of a target enterprise, the target characteristic data may be acquired by API private line transmission, so that the acquired target characteristic data is more comprehensive and real.
In addition, in actual operation, when enterprise fraud determination needs to be performed on a plurality of target enterprises, target feature data of each target enterprise may be obtained in advance and stored in a database table, and then when enterprise fraud determination needs to be performed, an enterprise ID is input to obtain target feature data of a target enterprise corresponding to the enterprise ID from a preset database table.
S20: and inputting the target characteristic data into a pre-trained enterprise rating card model to obtain a target enterprise rating value of the target enterprise.
Specifically, in this step, after the target feature data of the target enterprise is obtained, the target feature data is input into a pre-trained enterprise rating card model, and the target enterprise rating value of the target enterprise is obtained by calculating through the enterprise rating card model by using the target feature data. It is understood that the specific calculation manner is determined according to the trained enterprise rating card model, which is not limited in this embodiment, and after the enterprise rating card model is trained, the enterprise rating card model is stored so as to input the target feature data into the enterprise rating card model.
S30: and determining the fraud risk degree of the target enterprise according to the preset grading segment and the grading value of the target enterprise.
Specifically, in the step, scoring segments are set for enterprise scoring values in advance, namely scoring value ranges for the enterprise scoring values are set, and then corresponding fraud risk degrees are set according to the scoring segments; in actual operation, a corresponding score segment may be set according to an actual situation, and the score range in the score segment is not limited in this embodiment. After the grading segment is set, the fraud risk degree of the target enterprise can be determined according to the grading value of the target enterprise. For example, assume that the target business score value is between score segments 70-80, indicating that the target business is at a higher risk of fraud; if the score value of the target enterprise is between score segments 80-90, it indicates that the fraud risk of the target enterprise is high, and the above is only a specific example and is not limited in any way.
According to the enterprise fraud behavior judging method provided by the embodiment of the invention, the target enterprise grade value of the target enterprise is obtained by inputting the acquired target characteristic data of the target enterprise into the enterprise grade card model through the enterprise grade card model trained in advance; and then, according to preset grading segmentation, determining the fraud risk degree of the target enterprise according to the grading value of the target enterprise. Therefore, the method avoids manual offline depth investigation, can save a large amount of human resources, and improves the judgment efficiency; and the enterprise scoring card model trained through machine learning obtains the target enterprise scoring value of the target enterprise, so that the accuracy of enterprise fraud judgment is improved.
Fig. 2 is a flowchart of a method for training an enterprise rating card model according to an embodiment of the present invention. On the basis of the above embodiments, this embodiment further describes and optimizes the technical solution, and specifically, in this embodiment, the process of training the enterprise rating card model specifically includes:
s21: acquiring sample characteristic data of a plurality of enterprises;
wherein the sample characteristic data comprises sample characteristic variables and corresponding sample variable values.
Specifically, when an enterprise rating card model is trained, firstly, sample characteristic data of a plurality of enterprises need to be acquired, wherein the sample characteristic data are preliminarily screened sample characteristic variables which can be used for judging enterprise fraud and sample variable values corresponding to the sample characteristic variables; the sample characteristic variables specifically comprise enterprise tax related information, industrial and commercial information, judicial information and the like, wherein the tax related information comprises enterprise tax registration information, stockholder information, income declaration information, overdue tax information and the like; the business information covers the business registration information, the share right information, the administrative punishment and the like of the enterprise; the judicial information covers enterprise complaint judgment documents, information of losing trust execution, information of high-limit consumption and entry and exit of enterprise owners and the like.
As a preferred embodiment, after obtaining sample feature data of a plurality of enterprises, the method further includes:
and performing data cleaning and exception handling on the sample characteristic data.
It should be noted that, in this embodiment, the data cleaning refers to finding and correcting recognizable errors in the sample feature data in combination with the tax-related data service logic, including checking consistency of the sample feature data, and processing invalid values, abnormal values, missing values, and the like in the sample feature data to improve training speed and accuracy of training the enterprise rating card model, and the specific data cleaning mode is not limited in this embodiment; the exception handling refers to deleting the sample characteristic variables with more missing sample variable values in the sample characteristic variables so as to avoid the influence of a large amount of missing data on the accuracy of the training enterprise scoring card model.
S22: and setting FP-Tree for the sample characteristic variables according to the support degree, the reliability and the action degree, and determining the mode entering variables.
Specifically, the Support (Support) is the ratio of the number of occurrences of several associated sample feature variables in the sample feature variable set to the total sample feature variable set, or the probability of the occurrence of several sample feature variables in association, that is,
wherein N represents the total category number of the sample characteristic variables;
in particular toSuch as, for example,
the support degree of the correlation of the X and Y sample characteristic variables is represented;
in another example of this application, a,
and the support degree of the correlation of the X, Y and Z sample characteristic variables is shown.
Specifically, the Confidence (Confidence) is used to measure the accuracy of the association rule, that is, the Confidence represents the probability that one sample feature variable appears and another sample feature variable appears after the other sample feature variable appears, or the conditional probability of the sample feature variable, that is, the conditional probability of the sample feature variable is used to measure the accuracy of the association rule
Specifically, for example,
representing the confidence of X to Y in two sample characteristic variables X and Y with relevance;
in another example of this application, a,
and representing the confidence of X to Y and Z in three sample characteristic variables X, Y and Z with relevance.
Specifically, the degree of action (or Lift) represents the correlation between the sample characteristic variables,
for example, for sample feature variables X and Y, a degree of action greater than 1, then
Are valid strong association rules; when the degree of action is less than or equal to 1, then
Is an invalid strong association rule. Need to make sure thatNote that, if the sample characteristic variables X and Y are independent, since P (X | Y) ═ P (X) at this time, therefore,
it should be noted that, the specific implementation manner of setting the FP-Tree according to the feature variable by using the support degree, the reliability and the action degree is common knowledge of those skilled in the art, and details are not described here. It should be noted that after the FP-Tree is set, the mode entry variable is determined according to the FP-Tree. Specifically, the in-mold variables refer to the characteristic variables which have relatively large influence on enterprise fraud in the sample characteristic variables.
As a preferred embodiment, setting FP-Tree for the sample feature variable according to the support degree, the reliability degree, and the action degree, and determining the process of the model entry variable specifically includes:
setting FP-Tree for the sample characteristic variable according to the support degree, the reliability and the action degree, and determining a mode entering variable;
and screening the module-entering variables by using the KS value, the AR value, the IV value and the VIF value, and updating the module-entering variables by using the screened module-entering variables.
Specifically, in this embodiment, after setting FP-Tree for the sample feature variable according to the support degree, the reliability degree, and the action degree, preliminarily determining the mode entry variable; and then, further screening the preliminarily determined model entering variables by utilizing the KS value, the AR value, the IV value and the VIF value, continuously and gradually regressing through a Logistic Regression (Logistic Regression) algorithm, continuously screening the model entering variables which have large influence on enterprise fraudulent behaviors, and taking the screened model entering variables as final model entering variables.
Wherein, KS (Kolmogorov-Smirnov) value is used for evaluating the distinguishing capability of the characteristic variable, and the index measures the difference between the accumulation subsections of the good and bad samples; the greater the cumulative difference of good and bad samples, the greater the KS index, and the stronger the risk discrimination ability of the characteristic variable. The ar (Accuracy ratio) value is a commonly used index in the evaluation of the financial wind control model, and is a means for analyzing the single-factor risk differentiation capability of the CAP curve (temporal Accuracy Profile). The larger the AR value, the closer the CAP curve is to the "perfect curve", the steeper the low score fraction the better the discrimination. The IV (information value) value is a quantity used to represent how much "information" each feature variable has to the target variable. Higher IV values indicate higher effective information content of the feature variables to the enterprise rating card model. The VIF (variance inflationfactors) value is a variance Inflation factor that detects multiple collinearity by examining the extent to which a given explanatory variable can be explained by all other explanatory variables in the regression equation. The higher the VIF value is, the more serious the influence of multiple collinearity is, and the comparison and elimination are carried out when the characteristic variables are selected.
Therefore, the model entering variables are further screened by the KS value, the AR value, the IV value and the VIF value, the screened model entering variables are used for updating the model entering variables, the number of the model entering variables can be further reduced, the complexity of the enterprise rating card model is reduced on the basis of ensuring the accuracy of the trained enterprise rating card model, and the enterprise rating card model is lighter.
S23: and setting corresponding mold entering indexes according to the mold entering variables, and calculating the weight of each mold entering index according to the variable value of each sample.
Specifically, after the mold entering variable is determined, a corresponding mold entering index is set according to the mold entering variable; the mold entering indexes are the combination of one or more mold entering variables, and the weight of each mold entering variable is determined according to the sample variable value corresponding to the mold entering variable in each mold entering index. Specifically, the determined mode entering indexes are subjected to box separation, then, the mode entering indexes are subjected to WOE conversion, and the weight corresponding to each mode entering index is obtained according to the WOE conversion.
S24: and setting a calculation mode for calculating enterprise rating values corresponding to enterprises according to the modeling indexes, the corresponding weights and the box dividing coefficients of the modeling indexes to obtain an enterprise rating card model.
In this step, each model entering index, the corresponding weight and the box dividing coefficient of each model entering index are set for weighted multiplication, and an enterprise rating value corresponding to each enterprise, namely, an enterprise rating card model is obtained. Therefore, in the subsequent operation, the target characteristic data of the target enterprise, including the target characteristic variable and the corresponding variable value, is input into the enterprise rating card model, so that the enterprise rating card model can determine the corresponding target characteristic index according to the target characteristic variable, and determine the corresponding enterprise rating value according to the weight corresponding to the target characteristic index and the box dividing coefficient of the target characteristic index.
Therefore, the enterprise rating card model is trained according to the method of the embodiment, and the operation mode is fast and convenient.
On the basis of the foregoing embodiment, the embodiment further describes and optimizes the technical solution, and specifically, after determining the fraud risk degree of the target enterprise according to the target enterprise scoring value and according to the preset scoring segment index, the embodiment further includes:
and if the fraud risk degree of the target enterprise is higher than the early warning threshold value, sending out corresponding prompt information.
Specifically, the early warning threshold is a preset limit value of the fraud risk degree, after the fraud risk degree corresponding to the target enterprise is determined according to the score value of the target enterprise, whether the fraud risk degree is higher than the early warning threshold is further judged, and if the fraud risk degree is higher than the early warning threshold, the prompt device is further triggered to send out corresponding prompt information.
It should be noted that the prompting device may specifically be a buzzer and/or an indicator light and/or a display, and the prompting device triggers the buzzer/the indicator light/the display to send out corresponding prompting information, such as a buzzer sound/a flashing light/displayed characters or images, so as to intuitively prompt the user that the fraud risk degree of the target enterprise is higher, and the target enterprise has a greater fraud risk, so that the use experience of the user can be further improved.
On the basis of the above embodiments, the present embodiment further describes and optimizes the technical solution, and specifically, the present embodiment further includes:
and recording the enterprise information of the target enterprise of which the fraud risk degree is higher than the early warning threshold value.
In this embodiment, after it is determined that the fraud risk level of the target enterprise is higher than the early warning threshold, enterprise information of the target enterprise is further recorded.
Specifically, in actual operation, the enterprise information may be recorded in a text document, or information such as the enterprise information of the target enterprise, the determination time, and the corresponding fraud risk degree may be recorded in a form of a statistical table, and a specific recording manner is not limited in this embodiment. More specifically, the storage may be performed in a Memory bank, a hard disk, a TF (Trans-flash card) card, an sd (secure Digital Memory card), or the like, and the selection is specifically performed according to actual requirements, which is not limited in this embodiment.
In this embodiment, through further recording the enterprise information of the target enterprise of which the fraud risk degree is higher than the early warning threshold, the credit condition of each target enterprise can be conveniently checked according to the recorded enterprise information in the follow-up further process, and the use experience is further improved.
On the basis of the above embodiments, the present embodiment further describes and optimizes the technical solution, and specifically, the present embodiment further includes:
and displaying the target characteristic data and/or the target enterprise scoring value and/or the fraud risk degree of the target enterprise through the webpage.
In this embodiment, after the target characteristic data of the target enterprise is acquired, or after the score value of the target enterprise is obtained by using the enterprise scoring card model, or after the fraud course degree of the target enterprise is determined, the target characteristic data of the target enterprise and/or the score value of the target enterprise and/or the fraud risk degree are further displayed through a webpage. It should be noted that, in actual operation, the display may be in the form of characters or icons, which is not limited in this embodiment.
Therefore, the target characteristic data and/or the target enterprise score value and/or the fraud risk degree of the target enterprise are further displayed through the webpage, so that the user can more conveniently check the fraud behavior of the target enterprise, and the use experience of the client is further improved.
The above detailed description is given to the embodiment of the method for determining enterprise fraud, and the present invention further provides a device, an apparatus, and a computer-readable storage medium for determining enterprise fraud corresponding to the method.
Fig. 3 is a structural diagram of an apparatus for determining enterprise fraud according to an embodiment of the present invention, and as shown in fig. 3, an apparatus for determining enterprise fraud includes:
an obtaining module 31, configured to obtain target feature data of a target enterprise;
the input module 32 is used for inputting the target characteristic data into a pre-trained enterprise rating card model to obtain a target enterprise rating value of a target enterprise;
and the determining module 33 is configured to determine the fraud risk degree of the target enterprise according to the preset scoring segment and the target enterprise scoring value.
The enterprise fraud judgment device provided by the embodiment of the invention has the beneficial effect of the enterprise fraud judgment method.
As a preselected implementation, the input module specifically includes:
the system comprises an acquisition unit, a processing unit and a processing unit, wherein the acquisition unit is used for acquiring sample characteristic data of a plurality of enterprises; the sample characteristic data comprises sample characteristic variables and corresponding sample variable values;
the setting unit is used for setting FP-Tree for the sample characteristic variable according to the support degree, the credibility and the action degree and determining the mode entering variable;
the calculating unit is used for setting corresponding mold entering indexes according to the mold entering variables and calculating the weight of each mold entering index according to the variable value of each sample;
and the combination unit is used for setting a calculation mode of calculating the enterprise rating value corresponding to each enterprise according to each module entering index, the corresponding weight and the box dividing coefficient of each module entering index, so as to obtain the enterprise rating card model.
As a preselected embodiment, the setting unit specifically includes:
the first setting subunit is used for setting FP-Tree for the sample characteristic variable according to the support degree, the credibility and the action degree, and determining the mode entering variable;
and the second setting subunit is used for screening the module-entering variables by using the KS value, the AR value, the IV value and the VIF value and updating the module-entering variables by using the screened module-entering variables.
As a pre-selected embodiment, further comprising:
and the data processing module is used for carrying out data cleaning and exception handling on the sample characteristic data.
As a pre-selected embodiment, further comprising:
and the early warning module is used for sending out corresponding prompt information if the fraud risk degree of the target enterprise is higher than the early warning threshold value.
As a pre-selected embodiment, further comprising:
and the recording module is used for recording the enterprise information of the target enterprise of which the fraud risk degree is higher than the early warning threshold value.
As a pre-selected embodiment, further comprising:
and the display module is used for displaying the target characteristic data and/or the target enterprise scoring value and/or the fraud risk degree of the target enterprise through the webpage.
Fig. 4 is a structural diagram of an enterprise fraud determination apparatus according to an embodiment of the present invention, and as shown in fig. 4, an enterprise fraud determination apparatus includes:
a memory 41 for storing a computer program;
a processor 42, adapted to implement the steps of the above-mentioned enterprise fraud determination method when executing the computer program.
The enterprise fraud judgment device provided by the embodiment of the invention has the beneficial effect of the enterprise fraud judgment method.
In order to solve the above technical problem, the present invention further provides a computer-readable storage medium, on which a computer program is stored, and the computer program, when executed by a processor, implements the steps of the above method for determining enterprise fraud.
The computer-readable storage medium provided by the embodiment of the invention has the beneficial effect of the method for judging the enterprise fraud behavior.
The method, the device, the equipment and the computer readable storage medium for determining enterprise fraud provided by the invention are described in detail above. The principles and embodiments of the present invention are explained herein using specific examples, which are set forth only to help understand the method and its core ideas of the present invention. It should be noted that, for those skilled in the art, it is possible to make various improvements and modifications to the present invention without departing from the principle of the present invention, and those improvements and modifications also fall within the scope of the claims of the present invention.
The embodiments are described in a progressive manner in the specification, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other. The device disclosed by the embodiment corresponds to the method disclosed by the embodiment, so that the description is simple, and the relevant points can be referred to the method part for description.
Those of skill would further appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both, and that the various illustrative components and steps have been described above generally in terms of their functionality in order to clearly illustrate this interchangeability of hardware and software. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
Claims (10)
1. A method for determining enterprise fraud, comprising:
acquiring target characteristic data of a target enterprise;
inputting the target characteristic data into a pre-trained enterprise rating card model to obtain a target enterprise rating value of the target enterprise;
and determining the fraud risk degree of the target enterprise according to the score value of the target enterprise according to preset score segments.
2. The method according to claim 1, wherein the process of training out the enterprise rating card model specifically comprises:
acquiring sample characteristic data of a plurality of enterprises; wherein the sample feature data comprises sample feature variables and corresponding sample variable values;
setting FP-Tree for the sample characteristic variable according to the support degree, the reliability and the action degree, and determining a mode entering variable;
setting corresponding mold entering indexes according to the mold entering variables, and calculating the weight of each mold entering index according to the variable value of each sample;
and setting a calculation mode of calculating enterprise scoring values corresponding to the enterprises according to the modeling indexes, the corresponding weights and the box dividing coefficients of the modeling indexes to obtain the enterprise scoring card model.
3. The method according to claim 2, wherein the step of setting the FP-Tree for the sample feature variable according to the support degree, the reliability degree, and the action degree to determine the mode-entering variable specifically comprises:
setting the FP-Tree for the sample characteristic variable according to the support degree, the credibility and the action degree, and determining a mode entering variable;
and screening the module-entering variables by using the KS value, the AR value, the IV value and the VIF value, and updating the module-entering variables by using the screened module-entering variables.
4. The method of claim 2, wherein after said obtaining sample profile data for a plurality of businesses, further comprising:
and performing data cleaning and exception handling on the sample characteristic data.
5. The method of claim 1, wherein after determining the fraud risk level of the target enterprise according to the target enterprise rating value according to the preset rating segment index, further comprising:
and if the fraud risk degree of the target enterprise is higher than the early warning threshold value, sending out corresponding prompt information.
6. The method of claim 5, further comprising:
and recording the enterprise information of the target enterprise of which the fraud risk degree is higher than the early warning threshold value.
7. The method of any one of claims 1 to 6, further comprising:
and displaying the target characteristic data and/or the target enterprise scoring value and/or the fraud risk degree of the target enterprise through a webpage.
8. An apparatus for determining fraud in an enterprise, comprising:
the acquisition module is used for acquiring target characteristic data of a target enterprise;
the input module is used for inputting the target characteristic data into a pre-trained enterprise rating card model to obtain a target enterprise rating value of the target enterprise;
and the determining module is used for determining the fraud risk degree of the target enterprise according to the target enterprise scoring value according to preset scoring segments.
9. An apparatus for determining fraud in an enterprise, comprising:
a memory for storing a computer program;
a processor for implementing the steps of the enterprise fraud determination method of any of claims 1 to 7 when executing said computer program.
10. A computer-readable storage medium, characterized in that the computer-readable storage medium has stored thereon a computer program which, when being executed by a processor, carries out the steps of the method for determining enterprise fraud according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010067142.5A CN111275338A (en) | 2020-01-20 | 2020-01-20 | Method, device, equipment and storage medium for judging enterprise fraud behaviors |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010067142.5A CN111275338A (en) | 2020-01-20 | 2020-01-20 | Method, device, equipment and storage medium for judging enterprise fraud behaviors |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111275338A true CN111275338A (en) | 2020-06-12 |
Family
ID=71003338
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010067142.5A Pending CN111275338A (en) | 2020-01-20 | 2020-01-20 | Method, device, equipment and storage medium for judging enterprise fraud behaviors |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111275338A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111798151A (en) * | 2020-07-10 | 2020-10-20 | 深圳前海微众银行股份有限公司 | Enterprise fraud risk assessment method, device, equipment and readable storage medium |
| CN111951027A (en) * | 2020-08-14 | 2020-11-17 | 上海冰鉴信息科技有限公司 | Enterprise identification method and device with fraud risk |
| CN112001785A (en) * | 2020-07-21 | 2020-11-27 | 小花网络科技(深圳)有限公司 | Network credit fraud identification method and system based on image identification |
| CN113724061A (en) * | 2021-08-18 | 2021-11-30 | 杭州信雅达泛泰科技有限公司 | Consumer financial product credit scoring method and device based on customer grouping |
-
2020
- 2020-01-20 CN CN202010067142.5A patent/CN111275338A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111798151A (en) * | 2020-07-10 | 2020-10-20 | 深圳前海微众银行股份有限公司 | Enterprise fraud risk assessment method, device, equipment and readable storage medium |
| CN112001785A (en) * | 2020-07-21 | 2020-11-27 | 小花网络科技(深圳)有限公司 | Network credit fraud identification method and system based on image identification |
| CN111951027A (en) * | 2020-08-14 | 2020-11-17 | 上海冰鉴信息科技有限公司 | Enterprise identification method and device with fraud risk |
| CN113724061A (en) * | 2021-08-18 | 2021-11-30 | 杭州信雅达泛泰科技有限公司 | Consumer financial product credit scoring method and device based on customer grouping |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107025596B (en) | Risk assessment method and system | |
| CN111275338A (en) | Method, device, equipment and storage medium for judging enterprise fraud behaviors | |
| CN111507831A (en) | Credit risk automatic assessment method and device | |
| CN110689438A (en) | Enterprise financial risk scoring method and device, computer equipment and storage medium | |
| CN111507822A (en) | Enterprise risk assessment method based on feature engineering | |
| CN110246031A (en) | Appraisal procedure, system, equipment and the storage medium of business standing | |
| CN110738564A (en) | Post-loan risk assessment method and device and storage medium | |
| CN112132233A (en) | Criminal personnel dangerous behavior prediction method and system based on effective influence factors | |
| CN112102073A (en) | Credit risk control method and system, electronic device and readable storage medium | |
| CN112215702A (en) | Credit risk assessment method, mobile terminal and computer storage medium | |
| CN111090833A (en) | Data processing method, system and related equipment | |
| US11250368B1 (en) | Business prediction method and apparatus | |
| CN110634060A (en) | User credit risk assessment method, system, device and storage medium | |
| CN110866832A (en) | Risk control method, system, storage medium and computing device | |
| CN113919432A (en) | Classification model construction method, data classification method and device | |
| CN112434862B (en) | Method and device for predicting financial dilemma of marketing enterprises | |
| CN109308660B (en) | Credit assessment scoring model evaluation method, apparatus, device and storage medium | |
| CN117132383A (en) | Credit data processing method, device, equipment and readable storage medium | |
| CN112131354B (en) | Answer screening method and device, terminal equipment and computer readable storage medium | |
| CN114387089A (en) | Client credit risk assessment method, device, equipment and storage medium | |
| US11915313B2 (en) | Using email history to estimate creditworthiness for applicants having insufficient credit history | |
| CN118037430A (en) | Prediction method and device for user credit default, electronic equipment and storage medium | |
| CN113011971B (en) | Risk measure method, apparatus, system and computer storage medium | |
| CN117349728A (en) | Quality evaluation method and device for intelligent model | |
| CN114662903A (en) | Automated credit granting method, apparatus, device, medium, and program product |
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 | ||
| CB02 | Change of applicant information |
Address after: Guangdong Nanshan District street of Shenzhen city in Guangdong province 518000 Shenzhen international software park a 8 Building 202 room Applicant after: Shenzhen Weizhong credit Technology Co.,Ltd. Address before: Guangdong Nanshan District street of Shenzhen city in Guangdong province 518000 Shenzhen international software park a 8 Building 202 room Applicant before: SHENZHEN VZOOM CREDIT INFORMATION SERVICE Co.,Ltd. |
|
| CB02 | Change of applicant information |