CN117974291A - Fraud intention processing method, device, computer equipment and medium for loan application - Google Patents
Fraud intention processing method, device, computer equipment and medium for loan application Download PDFInfo
- Publication number
- CN117974291A CN117974291A CN202410076278.0A CN202410076278A CN117974291A CN 117974291 A CN117974291 A CN 117974291A CN 202410076278 A CN202410076278 A CN 202410076278A CN 117974291 A CN117974291 A CN 117974291A
- Authority
- CN
- China
- Prior art keywords
- loan
- target user
- target
- user
- credit
- 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
- 238000003672 processing method Methods 0.000 title claims abstract description 13
- 238000013145 classification model Methods 0.000 claims abstract description 76
- 238000012549 training Methods 0.000 claims abstract description 49
- 238000012545 processing Methods 0.000 claims abstract description 41
- 230000006399 behavior Effects 0.000 claims abstract description 40
- 238000000034 method Methods 0.000 claims description 34
- 230000008569 process Effects 0.000 claims description 10
- 238000004590 computer program Methods 0.000 claims description 7
- 230000006870 function Effects 0.000 claims description 7
- 238000007477 logistic regression Methods 0.000 claims description 5
- 230000008901 benefit Effects 0.000 description 6
- 230000002159 abnormal effect Effects 0.000 description 5
- 238000012795 verification Methods 0.000 description 5
- 230000009471 action Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 238000004891 communication Methods 0.000 description 2
- 230000007774 longterm Effects 0.000 description 2
- 230000002093 peripheral effect Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000000802 evaporation-induced self-assembly Methods 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000011022 operating instruction Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Landscapes
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
Abstract
The present disclosure provides a fraud intent processing method, apparatus, computer device and medium for loan application, comprising: obtaining loan information of a target user, wherein the target user is a user who executes loan behaviors; training a target classification model, wherein the target classification model is used for carrying out credit classification on loan users, the target classification model is a multi-classification model, and each output result of the target classification model is used for describing the prediction probability of one credit classification; inputting loan information of the target user into a target classification model, and determining the prediction probability of each credit category to which the target user belongs based on the output result of the target classification model; identifying fraud intention of the target user when applying for loan based on the prediction probability of each credit category to which the target user belongs; and executing corresponding loan processing operation based on the identification result of the fraudulent intention when the target user applies for the loan. Therefore, whether the loan behavior belongs to fraud can be effectively identified, and corresponding loan processing operation is performed.
Description
Technical Field
Embodiments of the present disclosure relate to the field of computer processing technology, and in particular, to a fraud intent processing method, apparatus, computer device, and medium suitable for a loan application.
Background
Bank loans are an economic action in which individuals or businesses give funds lend to an individual or business in a fund demand at a certain interest rate to a certain bank according to the national policy of the bank, and settle the term returns, and more businesses or individuals borrow money to banks in the form of loans.
Among the numerous loan behaviors, there is a portion of malicious loan behaviors that can cause some loss to the lended bank, and therefore, identifying the fraudulent intent of the loan user during the loan process is of paramount importance. In the related art, it is mainly relied on whether the loan behavior of the loan user has a fraudulent intention by detecting the loan data of the loan user, such as whether to register in a large number, apply for various financial institution accounts, etc.
However, in the above implementation, the loan data has a certain delay, which affects the recognition efficiency, resulting in low recognition efficiency of the fraudulent intention.
Disclosure of Invention
Embodiments described herein provide a fraudulent intent processing method, apparatus, computer device, and medium for loan application that overcomes the above-described problems.
In a first aspect, according to the present disclosure, there is provided a fraud intention processing method of a loan application, including:
Obtaining loan information of a target user, wherein the target user is a user who executes loan behaviors;
Training a target classification model, wherein the target classification model is used for carrying out credit classification on loan users, the target classification model is a multi-classification model, and each output result of the target classification model is used for describing the prediction probability of one credit classification;
Inputting loan information of the target user into the target classification model, and determining the prediction probability of each credit category to which the target user belongs based on the output result of the target classification model;
Identifying fraud intention of the target user when applying for loan based on the prediction probability of each credit category to which the target user belongs, wherein the fraud intention of the target user when applying for loan is determined and obtained based on the credit score of the loan corresponding to the target user, and the credit score of the loan corresponding to the target user is determined and obtained based on the prediction probability of each credit category to which the target user belongs;
and executing corresponding loan processing operation based on the identification result of the fraudulent intention when applying for the loan to the target user.
In a second aspect, according to the present disclosure, there is provided a fraud intention processing apparatus of a loan application, characterized by comprising:
The acquisition module is used for acquiring loan information of a target user, wherein the target user is a user who executes loan behaviors;
The training module is used for training a target classification model, wherein the target classification model is used for carrying out credit classification on loan users, the target classification model is a multi-classification model, and each output result of the target classification model is used for describing the prediction probability of one credit classification;
The determining module is used for inputting loan information of the target user into the target classification model, and determining the prediction probability of each credit category to which the target user belongs based on the output result of the target classification model;
The identification module is used for identifying fraud intention of the target user when applying for loan based on the prediction probability of each credit category of the target user, wherein the fraud intention of the target user when applying for loan is determined and obtained based on the credit score of the loan corresponding to the target user, and the credit score of the loan corresponding to the target user is determined and obtained based on the prediction probability of each credit category of the target user;
And the processing module is used for executing corresponding loan processing operation based on the identification result of the fraudulent intention when the target user applies for the loan.
In a third aspect, there is provided a computer device comprising a memory in which a computer program is stored and a processor which, when executing the computer program, performs the steps of the fraud intention processing method of a loan application as in any of the above embodiments.
In a fourth aspect, there is provided a computer readable storage medium having stored thereon a computer program which, when executed by a processor, performs the steps of a fraud intention processing method of a loan application as in any of the above embodiments.
The fraud intention processing method for the loan application provided by the embodiment of the application obtains the loan information of the target user, wherein the target user is the user who executes the loan behavior; training a target classification model, wherein the target classification model is used for carrying out credit classification on loan users, the target classification model is a multi-classification model, and each output result of the target classification model is used for describing the prediction probability of one credit classification; inputting loan information of the target user into a target classification model, and determining the prediction probability of each credit category to which the target user belongs based on the output result of the target classification model; identifying fraud intention of the target user when applying for loan based on the prediction probability of each credit category to which the target user belongs, wherein the fraud intention of the target user when applying for loan is determined based on loan credit scores corresponding to the target user, and the loan credit scores corresponding to the target user are determined based on the prediction probability of each credit category to which the target user belongs; and executing corresponding loan processing operation based on the identification result of the fraudulent intention when the target user applies for the loan. In this way, in the process of executing the loan behavior by the loan user, the prediction probability of the credit category of the loan user is output through the pre-constructed target classification model, and then the fraud intention of the target user when applying for the loan is identified through the prediction probability, so that whether the loan behavior of the loan user belongs to fraud can be effectively identified, and the corresponding loan processing operation is performed.
The foregoing description is only an overview of the technical solutions of the embodiments of the present application, and may be implemented according to the content of the specification, so that the technical means of the embodiments of the present application can be more clearly understood, and the following specific embodiments of the present application are given for clarity and understanding.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the following brief description of the drawings of the embodiments will be given, it being understood that the drawings described below relate only to some embodiments of the present disclosure, not to limitations of the present disclosure, in which:
Fig. 1 is a flow chart of a fraud intent processing method of a loan application provided by the present disclosure.
Fig. 2 is a schematic structural view of a fraud intention processing apparatus of a loan application provided by the present disclosure.
Fig. 3 is a schematic structural diagram of a computer device provided in the present disclosure.
It is noted that the elements in the drawings are schematic and are not drawn to scale.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present disclosure more apparent, the technical solutions of the embodiments of the present disclosure will be clearly and completely described below with reference to the accompanying drawings. It will be apparent that the described embodiments are some, but not all, of the embodiments of the present disclosure. All other embodiments, which can be made by those skilled in the art based on the described embodiments of the present disclosure without the need for creative efforts, are also within the scope of the protection of the present disclosure.
Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the presently disclosed subject matter belongs. It will be further understood that terms, such as those defined in commonly used dictionaries, should be interpreted as having a meaning that is consistent with their meaning in the context of the specification and relevant art and will not be interpreted in an idealized or overly formal sense unless expressly so defined herein. As used herein, a statement that two or more parts are "connected" or "coupled" together shall mean that the parts are joined together either directly or joined through one or more intermediate parts.
Reference herein to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment may be included in at least one embodiment of the application. The appearances of the phrase "an embodiment" in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. Those of skill in the art will explicitly and implicitly appreciate that the embodiments described herein may be combined with other embodiments.
The term "and/or" is herein merely an association relationship describing an associated object, meaning that there may be three relationships, e.g., a and/or B, may represent: there are three cases, a, B, a and B simultaneously. In addition, the character "/" herein generally indicates that the front and rear associated objects are an "or" relationship. Terms such as "first" and "second" are used merely to distinguish one component (or portion of a component) from another component (or another portion of a component).
In the description of the present application, unless otherwise indicated, the meaning of "plurality" means two or more (including two), and similarly, "plural sets" means two or more (including two).
In order to make the person skilled in the art better understand the solution of the present application, the technical solution of the embodiment of the present application will be clearly and completely described below with reference to the accompanying drawings.
Fig. 1 is a flow chart of a fraud intent processing method of a loan application, provided by an embodiment of the disclosure. As shown in fig. 1, the specific procedure of the fraud intention processing method of the loan application includes:
S110, obtaining loan information of the target user.
The target user is a user who performs a loan, that is, a user who makes a loan application for a loan in a certain bank.
When a target user performs a bank loan, some pre-loan information is usually required to be filled, that is, obtaining the loan information of the target user may include: acquiring initial loan data filled by a target user, wherein the initial loan data comprises: income and property amount, real name authentication and the like; and performing behavior verification on the initial loan data to obtain the loan information of the target user.
Behavior verification may include: verifying whether the income property amount is not filled, such as being too high; judging whether the target user reads the account opening information or not according to the stay time of the target user on a reading interface of the account opening information and the operation behaviors (such as page turning, page turning downwards and the like); judging the input duration of the target user for filling the identification card number according to the preset duration range, if the filling duration of the target user is within the preset duration range, indicating that the filling duration of the target user is compliant, if the filling duration of the target user is less than the preset duration range, indicating that the filling duration of the target user is too fast, and not considering operation, and if the filling duration of the target user is greater than the preset duration range, indicating that the filling duration of the target user is too slow, and not considering operation; according to the verification mobile phone number used by the target user in real-name authentication, if the verification mobile phone number is one mobile phone number, the real-name authentication of the target user is correct, and if the verification mobile phone number is a plurality of different mobile phone numbers, the real-name authentication of the target user is abnormal.
Accordingly, the loan information for the target user may include, but is not limited to: the method comprises the steps of comparing the income property (such as overhigh), reading the account opening information (such as direct selection of reading or unread), identifying the identification time of the ID card number (such as abnormal or normal), and carrying out multiple real-name authentication on different equipment information (such as abnormal or normal).
S120, training a target classification model.
The target classification model is used for carrying out credit classification on loan users, the target classification model is a multi-classification model, and each output result of the target classification model is used for describing the prediction probability of one credit classification. The target classification model is a logistic regression model with multiple inputs and multiple outputs.
In some embodiments, training the object classification model includes:
Obtaining a model pre-training sample, the model pre-training sample comprising: the marked loan information corresponding to the first type of users and the marked loan information corresponding to the second type of users, wherein the first type of users are used for describing the users who have loaned and are default, and the second type of users are used for describing the users who have loan and are not default.
Wherein the first type of user may be a customer who violates more than three periods, and the second type of user may be a customer who has long-term (e.g., three years) credit activity but is not violated.
In some embodiments, the loan information comprises: the method comprises the steps of comparing income property, reading account opening information, identifying identification card number input duration, and carrying out multiple real-name authentication on different equipment information.
Accordingly, the loan information corresponding to the first type of user may include: the comparison result of income property corresponding to customers with more than three periods of default, the reading result of account opening information, the identification result of the identification card number input time length, and whether to carry out multiple real-name authentication of different equipment information. The loan information corresponding to the second type of user may include: there are long-term (e.g., three years) credit behavior, but not comparison results of incomes and properties corresponding to clients, reading results of account opening information, identification results of identification card number input duration, and whether to perform multiple real-name authentication of different device information.
Based on the model pre-training sample, training the logistic regression model by adopting a preset loss function to obtain a pre-training model.
Obtaining a model adjustment training sample; training the pre-training model by adopting a preset loss function based on the model adjustment training sample to obtain a target classification model.
The model adjustment training sample is used for describing loan information corresponding to a user with loan intention and without loan behaviors. The loan information may include: the method comprises the steps of comparing the income property (such as overhigh), reading the account opening information (such as direct selection of reading or unread), identifying the identification time of the ID card number (such as abnormal or normal), and carrying out multiple real-name authentication on different equipment information (such as abnormal or normal).
The logistic regression model is trained through the model pre-training sample to obtain a pre-training model, the training sample is adjusted through the model, the pre-training model is trained again through the preset loss function to obtain a target classification model, and the estimation accuracy of the target classification model is improved.
S130, inputting loan information of the target user into the target classification model, and determining the prediction probability of each credit category to which the target user belongs based on the output result of the target classification model.
For example, the target classification model may be a classification model, and the output result of the target classification model is the prediction probabilities of two credit categories, where the two credit categories are respectively: offending customers, non-offending customers.
S140, based on the prediction probability of each credit category to which the target user belongs, identifying the fraud intention of the target user when applying for the loan.
After determining the prediction probability of each credit category to which the target user belongs, the fraud intention of the target user when applying for the loan can be identified by combining the weight coefficient corresponding to each prediction probability.
The fraud intention of the target user when applying for the loan is determined based on the credit score of the loan corresponding to the target user, and the credit score of the loan corresponding to the target user is determined based on the prediction probability of each credit category to which the target user belongs.
In some embodiments, identifying fraudulent intent of the target user in making the loan application based on the predicted probability of each credit category to which the target user belongs includes:
determining a weight coefficient corresponding to the prediction probability of each credit category to which the target user belongs; and determining the credit score of the loan corresponding to the target user based on the prediction probability of each credit category to which the target user belongs and the weight coefficient corresponding to each prediction probability.
Wherein, the weight coefficient corresponding to each prediction probability can be determined based on the ratio between the prediction probabilities of each credit category to which the target user belongs, for example, the prediction probability of the offending customer is 0.8, the prediction probability of the non-offending customer is 0.4, and then the weight coefficient corresponding to the prediction probability of the offending customer can be determined to be 0.8/(0.8+0.4) =0.67, and the weight coefficient corresponding to the prediction probability of the non-offending customer is determined to be 1-0.67=0.33. Correspondingly, the credit score of the loan corresponding to the target user is as follows: (0.67 x 0.8+0.33 x 0.4) 100=66.8.
Alternatively, the same weight coefficient may be assigned to each prediction probability, for example, the weight coefficient corresponding to the prediction probability of the offending customer is 0.5, and the weight coefficient corresponding to the prediction probability of the non-offending customer is 0.5. Correspondingly, the credit score of the loan corresponding to the target user is as follows: (0.5 x 0.8+0.5 x 0.4) x 100=60.
When the credit rating of the loan corresponding to the target user is smaller than or equal to a first preset rating threshold, acquiring the user attribute of the target user, and identifying the fraud intention of the target user when applying for the loan based on the user attribute; and when the credit rating of the loan corresponding to the target user is larger than a first preset rating threshold, acquiring the loan application amount of the target user, and identifying the fraud intention of the target user when applying for the loan based on the loan application amount.
Therefore, based on the magnitude relation between the credit rating of the target user and the first preset rating threshold, the fraud intention of the target user when applying for the loan is determined according to the conditions, and the fraud intention of the user under different loan conditions is effectively and quantitatively identified.
S150, based on the identification result of the fraudulent intention when applying the loan to the target user, executing corresponding loan processing operation.
Wherein the loan processing operation may comprise; agreeing to the loan behaviors of the target user, refusing the loan behaviors of the target user, and reducing the loan application amount of the target user. The corresponding loan processing operation is convenient to be executed aiming at the identification results of different fraud intentions, and the appropriate loan processing can be performed on the premise of identifying the loan intentions of the user.
In this embodiment, by acquiring loan information of a target user, the target user is a user who performs a loan action; training a target classification model, wherein the target classification model is used for carrying out credit classification on loan users, the target classification model is a multi-classification model, and each output result of the target classification model is used for describing the prediction probability of one credit classification; inputting loan information of the target user into a target classification model, and determining the prediction probability of each credit category to which the target user belongs based on the output result of the target classification model; identifying fraud intention of the target user when applying for loan based on the prediction probability of each credit category to which the target user belongs, wherein the fraud intention of the target user when applying for loan is determined based on loan credit scores corresponding to the target user, and the loan credit scores corresponding to the target user are determined based on the prediction probability of each credit category to which the target user belongs; and executing corresponding loan processing operation based on the identification result of the fraudulent intention when the target user applies for the loan. In this way, in the process of executing the loan behavior by the loan user, the prediction probability of the credit category of the loan user is output through the pre-constructed target classification model, and then the fraud intention of the target user when applying for the loan is identified through the prediction probability, so that whether the loan behavior of the loan user belongs to fraud can be effectively identified, and the corresponding loan processing operation is performed.
In some embodiments, identifying fraudulent intent of the target user for the loan application based on the user attributes includes:
and when the user attribute indicates that the target user is the first loan, acquiring credit rating of the target user.
The credit rating of the target user can be obtained based on the credit rating data of the target user, for example, a certain weight proportion is set for different credit rating data respectively, and the credit rating of the target user is determined by combining the sum of products of all weight proportions and corresponding credit rating data.
When the credit rating is smaller than or equal to a second preset rating threshold, determining that the target user has fraud intention when applying for the loan; and when the credit rating is larger than a second preset rating threshold, determining that the target user has no fraud intention when applying for the loan.
Therefore, the fraud intention of the target user when the loan application is carried out is effectively and quantitatively distinguished based on the size judgment between the credit rating and the second preset rating threshold.
Based on the identification result of the fraudulent intention when applying for the loan to the target user, executing corresponding loan processing operation, including:
And rejecting the loan behavior of the target user when determining that the target user has the fraudulent intention when applying for the loan, thereby reducing the benefit loss of the loan bank by rejecting the loan behavior of the target user when determining that the target user for the first loan has the fraudulent intention.
And when the target user is determined to have no fraudulent intention in the process of applying for the loan, the loan application amount of the target user is reduced based on the credit score/credit score of the loan, so that when the credit score of the target user is larger than a second preset scoring threshold value but is the first loan user, the benefit loss of a loan bank caused by the condition that the first loan user has malicious illegal loan is avoided by reducing the loan application amount.
In addition, when the target user is determined to have a fraudulent intention in the loan application, after rejecting the loan behavior of the target user, the target user can be marked as a blacklist user, and the loan behavior of the target user is directly rejected when the target user loans next time, so that the recognition process is reduced, and the subsequent recognition efficiency is further improved.
When the target user is determined to have no fraudulent intention in the process of applying for the loan, a secondary loan application policy can be sent to the target user after the loan application amount of the target user is reduced based on the loan credit score/credit score, and when the secondary loan application of the target user is received and the current credit score of the target user is determined to be greater than a second preset scoring threshold value, the loan behavior of the target user is processed based on the secondary loan application policy. The secondary loan application policy may be to promote the loan amount, increase the loan amount, speed up the loan processing flow/progress, etc. Thereby increasing the viscosity of the target user.
In other embodiments, identifying fraudulent intent of the target user for the loan application based on the user attributes includes:
and when the user attribute indicates that the target user is not the first loan, acquiring historical credit average of the target user.
Wherein, the historical credit average is the average value of all credit scores of the target user in the historical loan behavior determination.
When the average score of the historical credit is smaller than or equal to a third preset scoring threshold value, determining that the target user has fraud intention when applying for loans; and when the average score of the historical credit is larger than a third preset scoring threshold value, determining that the target user has no fraudulent intention when applying for the loan.
Based on the identification result of the fraudulent intention when applying for the loan to the target user, executing corresponding loan processing operation, including:
and rejecting the loan behavior of the target user when determining that the target user has a fraudulent intention when applying for the loan, so that the benefit loss of the loan bank can be reduced by rejecting the loan behavior of the target user when the credit score of the loan corresponding to the target user is smaller than or equal to a first preset score threshold.
And when the target user is determined to have no fraudulent intention in the loan application, the loan application amount of the target user is reduced based on the historical credit average, so that when the loan credit score corresponding to the target user is smaller than or equal to a first preset score threshold value, the benefit loss of a loan bank caused by the malicious illegal loan condition of the non-first loan user is avoided by reducing the loan application amount.
In some embodiments, identifying a fraudulent intent of the target user for the loan application based on the loan application amount includes:
and obtaining the user grade of the target user.
The user grade of the target user can be obtained by grading and determining according to the historical loan situation/credit investigation situation of the target user.
Matching the grade loan amount corresponding to the user grade with the loan application amount of the target user; if the grade loan amount is matched with the loan application amount, determining that the target user does not have fraud intention when applying for the loan; if the level loan amount is not matched with the loan application amount, determining that the target user does not have fraud intention when applying for the loan.
Based on the identification result of the fraudulent intention when applying for the loan to the target user, executing corresponding loan processing operation, including:
When no fraudulent intention is determined when the target user applies for the loan based on the matching of the grade loan amount and the loan application amount, the loan behavior of the target user is agreed, so that when the target user is judged to have no fraudulent intention based on the matching of the grade loan amount and the loan application amount under the condition that the loan credit score corresponding to the target user is larger than the first preset scoring threshold value, the viscosity of the user is increased on the premise of ensuring that the bank benefit is not lost by agreeing to the loan behavior of the target user.
And when the fact that no fraudulent intention exists when the target user applies for the loan is determined based on the fact that the grade loan amount is not matched with the loan application amount, reducing the loan application amount of the target user based on the grade loan amount. Therefore, when the credit rating of the loan corresponding to the target user is larger than the first preset rating threshold value and the target user is judged to have no fraudulent intention based on the mismatching of the loan amount and the loan application amount, malicious loan behaviors are prevented from happening in a way of reducing the loan application amount, and bank losses are timely reduced.
Fig. 2 is a schematic structural diagram of a fraud intention processing device for loan application according to the present embodiment. The fraud intention processing device of the loan application may include: the system comprises an acquisition module 210, a training module 220, a determination module 230, an identification module 240 and a processing module 250.
The obtaining module 210 is configured to obtain loan information of a target user, where the target user is a user performing a loan activity.
The training module 220 is configured to train a target classification model, where the target classification model is used to classify credit of a loan user, and the target classification model is a multi-classification model, and each output result of the target classification model is used to describe a prediction probability of one credit classification.
A determining module 230, configured to input loan information of the target user into the target classification model, and determine a prediction probability of each credit category to which the target user belongs based on an output result of the target classification model.
The identifying module 240 is configured to identify, based on the predicted probability of each credit category to which the target user belongs, a fraud intention when the target user applies for a loan, where the fraud intention when the target user applies for a loan is determined based on a loan credit score corresponding to the target user, and the loan credit score corresponding to the target user is determined based on the predicted probability of each credit category to which the target user belongs.
And the processing module 250 is used for executing corresponding loan processing operation based on the identification result of the fraudulent intention when the target user applies for the loan.
In this embodiment, the training module 220 is optionally specifically configured to:
obtaining a model pre-training sample, the model pre-training sample comprising: the method comprises the steps of marking loan information corresponding to a first type of user and loan information corresponding to a second type of user, wherein the first type of user is used for describing the user who has loaned and is default, and the second type of user is used for describing the user who has loan and is not default; training a logistic regression model by adopting a preset loss function based on the model pre-training sample to obtain a pre-training model; obtaining a model adjustment training sample, wherein the model adjustment training sample is used for describing loan information corresponding to a user with loan intention and without loan behaviors; and adjusting a training sample based on the model, and training the pre-training model by adopting the preset loss function to obtain the target classification model.
In this embodiment, optionally, the identification module 240 includes: the device comprises a first determining unit, a second determining unit, a first identifying unit and a second identifying unit.
And the first determining unit is used for determining a weight coefficient corresponding to the prediction probability of each credit category to which the target user belongs.
And the second determining unit is used for determining the credit score of the loan corresponding to the target user based on the prediction probability of each credit category to which the target user belongs and the weight coefficient corresponding to each prediction probability.
And the first identification unit is used for acquiring the user attribute of the target user when the credit rating of the loan corresponding to the target user is smaller than or equal to a first preset rating threshold value, and identifying the fraud intention of the target user when applying for the loan based on the user attribute.
And the second identification unit is used for acquiring the loan application amount of the target user when the loan credit score corresponding to the target user is larger than the first preset score threshold value, and identifying the fraud intention of the target user when applying for the loan based on the loan application amount.
In this embodiment, optionally, the first identifying unit is specifically configured to:
When the user attribute indicates that the target user is a first loan, acquiring credit rating of the target user; when the credit rating is smaller than or equal to a second preset rating threshold, determining that the target user has fraud intention when applying for loan; and when the credit rating is larger than the second preset rating threshold, determining that the target user has no fraud intention when applying for the loan.
The processing module 250 is specifically configured to:
Rejecting the loan behavior of the target user when determining that the target user has a fraudulent intention when applying for the loan; and when the target user is determined to have no fraudulent intention in the process of applying for the loan, reducing the loan application amount of the target user based on the loan credit score/the credit score.
In this embodiment, optionally, the first identifying unit is specifically configured to:
When the user attribute indicates that the target user is not a first loan, acquiring historical credit average of the target user; when the average score of the historical credit is smaller than or equal to a third preset scoring threshold value, determining that the target user has fraud intention when applying for loans; and when the historical credit average score is larger than the third preset scoring threshold value, determining that the target user has no fraud intention when applying for the loan.
The processing module 250 is specifically configured to:
rejecting the loan behavior of the target user when determining that the target user has a fraudulent intention when applying for the loan; and when the target user is determined to have no fraudulent intention in the process of applying for the loan, the loan application amount of the target user is reduced based on the historical credit average.
In this embodiment, optionally, the second identifying unit is specifically configured to:
Acquiring the user grade of the target user; matching the loan amount of the grade corresponding to the user grade with the loan application amount of the target user; if the grade loan amount is matched with the loan application amount, determining that the target user does not have fraud intention when applying for the loan; if the grade loan amount is not matched with the loan application amount, determining that the target user does not have fraud intention when applying for the loan.
The processing module 250 is specifically configured to:
Agreeing to the loan behavior of the target user when determining that the target user does not have a fraudulent intention in applying for the loan based on the matching of the level loan amount and the loan application amount; and when no fraudulent intention is determined to be caused when the target user applies for the loan based on the mismatch of the grade loan amount and the loan application amount, reducing the loan application amount of the target user based on the grade loan amount.
In this embodiment, optionally, the loan information includes: the method comprises the steps of comparing income property, reading account opening information, identifying identification card number input duration, and carrying out multiple real-name authentication on different equipment information.
The fraud intention processing device for loan application provided in the present disclosure may execute the above method embodiment, and the specific implementation principle and technical effects of the fraud intention processing device may refer to the above method embodiment, which is not described herein again.
The embodiment of the application also provides computer equipment. Referring specifically to fig. 3, fig. 3 is a basic structural block diagram of a computer device according to the present embodiment.
The computer device includes a memory 310 and a processor 320 communicatively coupled to each other via a system bus. It should be noted that only computer devices having components 310-320 are shown in the figures, but it should be understood that not all of the illustrated components are required to be implemented and that more or fewer components may be implemented instead. It will be appreciated by those skilled in the art that the computer device herein is a device capable of automatically performing numerical calculation and/or information processing according to a preset or stored instruction, and its hardware includes, but is not limited to, a microprocessor, an Application SPECIFIC INTEGRATED Circuit (ASIC), a programmable gate array (Field-ProgrammableGate Array, FPGA), a digital Processor (DIGITAL SIGNAL Processor, DSP), an embedded device, and the like.
The computer device may be a desktop computer, a notebook computer, a palm computer, a cloud server, or the like. The computer device can perform man-machine interaction with a user through a keyboard, a mouse, a remote controller, a touch pad or voice control equipment and the like.
The memory 310 includes at least one type of readable storage medium including non-volatile memory (non-volatile memory) or volatile memory, such as flash memory (flash memory), hard disk, multimedia card, card memory (e.g., SD or DX memory, etc.), random access memory (random accessmemory, RAM), read-only memory (ROM), erasable programmable read-only memory (erasableprogrammable read-only memory, EPROM), electrically erasable programmable read-only memory (ELECTRICALLY ERASABLE PROGRAMMABLE READ-only memory, EEPROM), programmable read-only memory (programmable read-only memory, PROM), magnetic memory, RAM, optical disk, etc., which may include static or dynamic. In some embodiments, memory 310 may be an internal storage unit of a computer device, such as a hard disk or memory of the computer device. In other embodiments, the memory 310 may also be an external storage device of a computer device, such as a plug-in hard disk, a smart memory card (SMART MEDIA CARD, SMC), a Secure Digital (SD) card, or a flash memory card (FLASH CARD) or the like, which are provided on the computer device. Of course, memory 310 may also include both internal storage units for computer devices and external storage devices. In this embodiment, the memory 310 is typically used to store an operating system installed on a computer device and various types of application software, such as program codes of the above-described methods. In addition, the memory 310 may also be used to temporarily store various types of data that have been output or are to be output.
Processor 320 is typically used to perform the overall operations of the computer device. In this embodiment, the memory 310 is used for storing program codes or instructions, the program codes include computer operation instructions, and the processor 320 is used for executing the program codes or instructions stored in the memory 310 or processing data, such as the program codes for executing the above-mentioned method.
Herein, the bus may be an industry standard architecture (Industry Standard Architecture, ISA) bus, a peripheral component interconnect (PERIPHERAL COMPONENT INTERCONNECT, PCI) bus, or an extended industry standard architecture (Extended Industry Standard Architecture, EISA) bus, among others. The bus system may be classified as an address bus, a data bus, a control bus, etc. For ease of illustration, the figures are shown with only one bold line, but not with only one bus or one type of bus.
Still another embodiment of the present application provides a computer-readable medium, which may be a computer-readable signal medium or a computer-readable medium. A processor in a computer reads computer readable program code stored in a computer readable medium, such that the processor is capable of performing the functional actions specified in each step or combination of steps in the above-described method; a means for generating a functional action specified in each block of the block diagram or a combination of blocks.
The computer readable medium includes, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared memory or semiconductor system, apparatus or device, or any suitable combination of the foregoing, the memory storing program code or instructions, the program code including computer operating instructions, and the processor executing the program code or instructions of the above-described methods stored by the memory.
The definition of memory and processor may refer to the description of the embodiments of the computer device described above, and will not be repeated here.
In the several embodiments provided by the present application, it should be understood that the disclosed systems, devices, and methods may be implemented in other manners. For example, the apparatus embodiments described above are merely illustrative, e.g., the division of modules or units is merely a logical functional division, and there may be additional divisions when actually implemented, e.g., multiple units or components may be combined or integrated into another system, or some features may be omitted or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be an indirect coupling or communication connection via some interfaces, devices or units, which may be in electrical, mechanical or other form.
The functional units or modules in the embodiments of the present application may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units may be implemented in hardware or in software functional units.
The integrated units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present application may be embodied in essence or a part contributing to the prior art or all or part of the technical solution in the form of a software product stored in a storage medium, including several instructions to cause a computer device (which may be a personal computer, a server, or a network device, etc.) or a processor (processor) to execute all or part of the steps of the methods of the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (Random Access Memory, RAM), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
In the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The word "comprising" does not exclude the presence of elements or steps other than those listed in a claim. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The application may be implemented by means of hardware comprising several distinct elements, and by means of a suitably programmed computer. In the unit claims enumerating several means, several of these means may be embodied by one and the same item of hardware. The use of first, second, third, etc. does not denote any order, and the words are to be interpreted as names. The steps in the above embodiments should not be construed as limiting the order of execution unless specifically stated.
The above embodiments are only for illustrating the technical solution of the present application, and are not limiting; although the application has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present application.
Claims (10)
1. A method of fraud intent processing for a loan application, comprising:
Obtaining loan information of a target user, wherein the target user is a user who executes loan behaviors;
Training a target classification model, wherein the target classification model is used for carrying out credit classification on loan users, the target classification model is a multi-classification model, and each output result of the target classification model is used for describing the prediction probability of one credit classification;
Inputting loan information of the target user into the target classification model, and determining the prediction probability of each credit category to which the target user belongs based on the output result of the target classification model;
Identifying fraud intention of the target user when applying for loan based on the prediction probability of each credit category to which the target user belongs, wherein the fraud intention of the target user when applying for loan is determined and obtained based on the credit score of the loan corresponding to the target user, and the credit score of the loan corresponding to the target user is determined and obtained based on the prediction probability of each credit category to which the target user belongs;
and executing corresponding loan processing operation based on the identification result of the fraudulent intention when applying for the loan to the target user.
2. The method of claim 1, wherein the training a target classification model comprises:
obtaining a model pre-training sample, the model pre-training sample comprising: the method comprises the steps of marking loan information corresponding to a first type of user and loan information corresponding to a second type of user, wherein the first type of user is used for describing the user who has loaned and is default, and the second type of user is used for describing the user who has loan and is not default;
Training a logistic regression model by adopting a preset loss function based on the model pre-training sample to obtain a pre-training model;
Obtaining a model adjustment training sample, wherein the model adjustment training sample is used for describing loan information corresponding to a user with loan intention and without loan behaviors;
and adjusting a training sample based on the model, and training the pre-training model by adopting the preset loss function to obtain the target classification model.
3. The method of claim 1, wherein the identifying the fraud intent of the target user at the time of the loan application based on the predicted probability of each credit category to which the target user belongs comprises:
determining a weight coefficient corresponding to the prediction probability of each credit category to which the target user belongs;
Determining a credit rating corresponding to the target user based on the prediction probability of each credit category to which the target user belongs and the weight coefficient corresponding to each prediction probability;
When the credit rating of the loan corresponding to the target user is smaller than or equal to a first preset rating threshold, acquiring the user attribute of the target user, and identifying the fraud intention of the target user when applying for the loan based on the user attribute;
and when the credit rating of the loan corresponding to the target user is larger than the first preset rating threshold, acquiring the loan application amount of the target user, and identifying the fraud intention of the target user when applying for the loan based on the loan application amount.
4. A method according to claim 3, wherein said identifying fraudulent intent of said target user in making a loan application based on said user attributes comprises:
When the user attribute indicates that the target user is a first loan, acquiring credit rating of the target user;
when the credit rating is smaller than or equal to a second preset rating threshold, determining that the target user has fraud intention when applying for loan; when the credit rating is greater than the second preset rating threshold, determining that the target user has no fraud intention when applying for the loan;
the executing corresponding loan processing operation based on the identification result of the fraudulent intention when applying for the loan to the target user comprises the following steps:
rejecting the loan behavior of the target user when determining that the target user has a fraudulent intention when applying for the loan;
And when the target user is determined to have no fraudulent intention in the process of applying for the loan, reducing the loan application amount of the target user based on the loan credit score/the credit score.
5. A method according to claim 3, wherein said identifying fraudulent intent of said target user in making a loan application based on said user attributes comprises:
When the user attribute indicates that the target user is not a first loan, acquiring historical credit average of the target user;
When the average score of the historical credit is smaller than or equal to a third preset scoring threshold value, determining that the target user has fraud intention when applying for loans; when the historical credit average score is larger than the third preset scoring threshold value, determining that the target user does not have fraud intention when applying for loans;
the executing corresponding loan processing operation based on the identification result of the fraudulent intention when applying for the loan to the target user comprises the following steps:
rejecting the loan behavior of the target user when determining that the target user has a fraudulent intention when applying for the loan;
And when the target user is determined to have no fraudulent intention in the process of applying for the loan, the loan application amount of the target user is reduced based on the historical credit average.
6. A method according to claim 3, wherein the identifying the intention of fraud by the target user at the time of the loan application based on the loan application amount comprises:
Acquiring the user grade of the target user;
matching the loan amount of the grade corresponding to the user grade with the loan application amount of the target user;
if the grade loan amount is matched with the loan application amount, determining that the target user does not have fraud intention when applying for the loan; if the grade loan amount is not matched with the loan application amount, determining that the target user does not have fraud intention when applying for the loan;
the executing corresponding loan processing operation based on the identification result of the fraudulent intention when applying for the loan to the target user comprises the following steps:
agreeing to the loan behavior of the target user when determining that the target user does not have a fraudulent intention in applying for the loan based on the matching of the level loan amount and the loan application amount;
and when no fraudulent intention is determined to be caused when the target user applies for the loan based on the mismatch of the grade loan amount and the loan application amount, reducing the loan application amount of the target user based on the grade loan amount.
7. The method of claim 2, wherein the loan information comprises: the method comprises the steps of comparing income property, reading account opening information, identifying identification card number input duration, and carrying out multiple real-name authentication on different equipment information.
8. A fraud intent processing device for a loan application, comprising:
The acquisition module is used for acquiring loan information of a target user, wherein the target user is a user who executes loan behaviors;
The training module is used for training a target classification model, wherein the target classification model is used for carrying out credit classification on loan users, the target classification model is a multi-classification model, and each output result of the target classification model is used for describing the prediction probability of one credit classification;
The determining module is used for inputting loan information of the target user into the target classification model, and determining the prediction probability of each credit category to which the target user belongs based on the output result of the target classification model;
The identification module is used for identifying fraud intention of the target user when applying for loan based on the prediction probability of each credit category of the target user, wherein the fraud intention of the target user when applying for loan is determined and obtained based on the credit score of the loan corresponding to the target user, and the credit score of the loan corresponding to the target user is determined and obtained based on the prediction probability of each credit category of the target user;
And the processing module is used for executing corresponding loan processing operation based on the identification result of the fraudulent intention when the target user applies for the loan.
9. A computer device comprising a memory and a processor, the memory having stored therein a computer program, the processor when executing the computer program implementing a method of fraud intent processing for a loan application as claimed in any of claims 1 to 7.
10. A computer-readable storage medium, on which a computer program is stored, characterized in that the computer program, when executed by a processor, implements a fraud intention processing method of a loan application as claimed in any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410076278.0A CN117974291A (en) | 2024-01-18 | 2024-01-18 | Fraud intention processing method, device, computer equipment and medium for loan application |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410076278.0A CN117974291A (en) | 2024-01-18 | 2024-01-18 | Fraud intention processing method, device, computer equipment and medium for loan application |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117974291A true CN117974291A (en) | 2024-05-03 |
Family
ID=90860966
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410076278.0A Pending CN117974291A (en) | 2024-01-18 | 2024-01-18 | Fraud intention processing method, device, computer equipment and medium for loan application |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117974291A (en) |
-
2024
- 2024-01-18 CN CN202410076278.0A patent/CN117974291A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107025596B (en) | Risk assessment method and system | |
| CN110297912A (en) | Cheat recognition methods, device, equipment and computer readable storage medium | |
| CN111275546B (en) | Financial customer fraud risk identification method and device | |
| CN111340616B (en) | Method, device, equipment and medium for approving online loan | |
| CN110930218B (en) | Method and device for identifying fraudulent clients and electronic equipment | |
| WO2020177478A1 (en) | Credit-based qualification information auditing method, apparatus and device | |
| US11250368B1 (en) | Business prediction method and apparatus | |
| CN111260189B (en) | Risk control method, risk control device, computer system and readable storage medium | |
| CN107807941A (en) | Information processing method and device | |
| CN112150014A (en) | Enterprise risk early warning method, device, equipment and readable storage medium | |
| CN110046648B (en) | Method and device for classifying business based on at least one business classification model | |
| CN111598122B (en) | Data verification method and device, electronic equipment and storage medium | |
| CN111046184B (en) | Text risk identification method, device, server and storage medium | |
| Barua et al. | Swindle: Predicting the probability of loan defaults using catboost algorithm | |
| CN113887202A (en) | Text error correction method and device, computer equipment and storage medium | |
| CN112750038B (en) | Transaction risk determination method, device and server | |
| CN117974291A (en) | Fraud intention processing method, device, computer equipment and medium for loan application | |
| CN116245630A (en) | Anti-fraud detection method and device, electronic equipment and medium | |
| KR102271346B1 (en) | Credit loan platform using atypical-data and its operating method | |
| CN110570301B (en) | Risk identification method, device, equipment and medium | |
| CN111914542A (en) | Suspected illegal investment market subject identification method, device, terminal and storage medium | |
| CN114565030B (en) | Feature screening method and device, electronic equipment and storage medium | |
| Kang | Fraud Detection in Mobile Money Transactions Using Machine Learning | |
| US11915313B2 (en) | Using email history to estimate creditworthiness for applicants having insufficient credit history | |
| CN117196806A (en) | Transaction risk control method and device applied to banking outlets |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination |