CN111369342B - Loan approval method, device, equipment and storage medium based on machine learning - Google Patents
Loan approval method, device, equipment and storage medium based on machine learning Download PDFInfo
- Publication number
- CN111369342B CN111369342B CN202010147886.8A CN202010147886A CN111369342B CN 111369342 B CN111369342 B CN 111369342B CN 202010147886 A CN202010147886 A CN 202010147886A CN 111369342 B CN111369342 B CN 111369342B
- Authority
- CN
- China
- Prior art keywords
- information
- user
- loan
- preset
- risk
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/03—Credit; Loans; Processing thereof
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- 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
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/16—Real estate
Abstract
The application relates to a loan approval method, a device, equipment and a storage medium based on machine learning. The method comprises the following steps: acquiring loan application information input by a user and equipment information of an input device currently used by the user, wherein the loan application information comprises loan information and borrower information; when the repayment probability of the user is determined to be larger than a preset first threshold value according to the borrower information, the equipment information, the historical transaction behavior information of the user and a preset risk early warning model, external data of the user are obtained, wherein the risk early warning model is a machine learning model, and the external data are non-prestored data; and determining the score of the user according to the loan application information, the external data and a preset scoring rule, and outputting an approval result according to the score. The method can improve the approval efficiency of the loan and simultaneously reduce the approval risk brought by manually approving the loan.
Description
Technical Field
The present application relates to the internet field, and in particular, to a loan approval method, apparatus, device and storage medium based on machine learning.
Background
With the rapid development of real estate, housing loan business has become an important expansion business for banks. How to provide housing loan services to customers more efficiently is a problem to be solved by those skilled in the art.
Traditional housing loan approval mainly adopts an offline application offline approval mode. The approval mode excessively depends on subjective judgment of a client manager, and has higher requirements on experience and personal quality of the client manager; meanwhile, for clients, the time cost of approval is high, and the timeliness is poor, so that the loan approval efficiency is low.
Disclosure of Invention
Based on this, it is necessary to provide a loan approval method, apparatus, device and storage medium based on machine learning, aiming at the technical problem of low loan approval efficiency in the conventional technology.
In a first aspect, an embodiment of the present application provides a loan approval method based on machine learning, including:
acquiring loan application information input by a user and equipment information of an input device currently used by the user, wherein the loan application information comprises loan information and borrower information;
When the repayment probability of the user is determined to be larger than a preset first threshold value according to the borrower information, the equipment information, the historical transaction behavior information of the user and a preset risk early warning model, external data of the user are obtained, wherein the risk early warning model is a machine learning model, and the external data are non-prestored data;
and determining the score of the user according to the loan application information, the external data and a preset scoring rule, and outputting an approval result according to the score.
In a second aspect, an embodiment of the present application provides a loan approval device based on machine learning, including:
the first acquisition module is used for acquiring loan application information input by a user and equipment information of an input equipment currently used by the user, wherein the loan application information comprises loan information and borrower information;
the second acquisition module is used for acquiring external data of the user when the repayment probability of the user is determined to be greater than a preset first threshold value according to the borrower information, the equipment information, the historical transaction behavior information of the user and a preset risk early warning model, wherein the risk early warning model is a machine learning model, and the external data is non-prestored data;
And the processing module is used for determining the score of the user according to the loan application information, the external data and a preset scoring rule and outputting an approval result according to the score.
In a third aspect, an embodiment of the present application provides a computer device, including a memory and a processor, where the memory stores a computer program, and the processor implements the method for approving a loan provided in the first aspect of the embodiment of the present application when executing the computer program.
In a fourth aspect, an embodiment of the present application provides a computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements the method for approving a loan provided by the first aspect of the embodiment of the present application.
The loan approval method, the device, the equipment and the storage medium based on machine learning provided by the embodiment of the application are that computer equipment acquires loan application information input by a user and equipment information of input equipment currently used by the user; acquiring external data of the user when the repayment probability of the user is determined to be greater than a preset first threshold value according to the borrower information, the equipment information, the historical transaction behavior information of the user and a preset risk early warning model in the loan application information; and determining the score of the user according to the loan application information, the external data and a preset scoring rule, and outputting an approval result according to the score. In the whole loan approval process, the computer equipment can automatically determine the repayment probability of the user through a machine learning model, namely a risk early warning model, based on loan application information input by the user, equipment information of the input equipment and historical transaction behavior information of the user, and when the repayment probability of the user is greater than a first threshold value, the computer equipment can determine the score of the user through a preset scoring rule based on the acquired external data of the user and the loan application information of the user, and output an approval result according to the determined score. Namely, the whole loan approval process is automatically completed by the computer equipment, and a customer manager is not required to conduct offline approval based on experience, so that the approval efficiency of the loan is improved, and the approval risk brought by manually approving the loan is reduced.
Drawings
FIG. 1 is a schematic flow chart of a loan approval method based on machine learning according to an embodiment of the application;
FIG. 2 is a schematic flow chart of another method for approving loans based on machine learning according to an embodiment of the application;
FIG. 3 is a schematic diagram of an information relationship network according to an embodiment of the present application;
fig. 4 is a schematic structural diagram of a loan approval device based on machine learning according to an embodiment of the application;
fig. 5 is a schematic structural diagram of a computer device according to an embodiment of the present application.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the technical solutions in the embodiments of the present application will be further described in detail by the following embodiments with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the application.
It should be noted that, the execution subject of the method embodiment described below may be a loan approval device based on machine learning, and the device may be implemented as part or all of a computer device by software, hardware, or a combination of software and hardware. The following method embodiments are described taking the implementation subject as a computer device as an example.
Fig. 1 is a schematic flow chart of a loan approval method based on machine learning according to an embodiment of the application. This embodiment relates to the specific process of how a computer device automatically approves loans for a user. As shown in fig. 1, the method may include:
s101, obtaining loan application information input by a user and equipment information of input equipment currently used by the user.
Specifically, the user may input loan application information through input means such as client software, applet, public number, etc. of the bank, so that the computer device obtains the loan application information input by the user. The loan application information may include loan information and borrower information, and the borrower information may include principal lender information and may also include participant information. Meanwhile, the computer device needs to collect device information of an input device currently used by a user, wherein the device information may include location information of the device, a media access control (Media Access Control, MAC) address, an internet protocol (Internet Protocol, IP) address, and the like of the device, and when the input device is a mobile device, the device information may further include an international mobile subscriber identity (International Mobile Subscriber Identification Number, IMSI), an international mobile equipment identity (International Mobile Equipment Identity, IMEI), a device fingerprint, a mobile phone number, and the like.
S102, when the repayment probability of the user is determined to be larger than a preset first threshold value according to the borrower information, the equipment information, the historical transaction behavior information of the user and a preset risk early warning model, external data of the user are obtained.
Specifically, the computer device obtains historical transaction behavior information of the user according to the identification of the user. The historical transaction behavior information can be historical loan behavior information, historical consumption behavior information, social data, e-commerce data and the like. And the computer equipment inputs the obtained borrower information, equipment information and historical transaction behavior information of the user into a risk early warning model to obtain repayment probability of the user. The risk early warning model is a machine learning model, and can predict risk factors of the user through the early warning model, and the repayment probability of the user is determined based on the predicted risk factors.
And when the repayment probability of the user is determined to be larger than a preset first threshold value, acquiring external data of the user. The external data is data which is not stored in advance, namely, data acquired through an external channel. Such as credit data, public accumulation data, social security data, and related data of the user obtained through business and customs channels.
When the repayment probability of the user is less than or equal to the first threshold value, the computer device outputs a refusal credit result to the user.
S103, determining the score of the user according to the loan application information, the external data and a preset scoring rule, and outputting an approval result according to the score.
Specifically, the external data may include first income data of the user, and the loan application information may include second income data of the user, where the first income data is income data obtained by the computer device according to the external data of the user, and the second income data is income data input by the user, so that the computer device needs to determine whether the first income data and the second income data match with the target loan amount, respectively. Optionally, in the step S103, the process of determining the score of the user according to the loan application information, the external data and the preset scoring rule may be: respectively judging whether the first income data and the second income data in the loan application information are matched with the target loan amount according to a preset mapping relation; and if at least one piece of income data is matched, determining the score of the user according to the credit investigation data in the external data by a preset scoring rule.
Wherein the mapping relation comprises the corresponding relation between different income data and loan amount. When the computer device determines that at least one of the income data matches, the computer device determines the score of the user according to the credit data in the external data, the preset exclusion policy, calculation standard, hard policy, special screening policy and screening policy, and outputs the approval result based on the determined score.
Optionally, if the first revenue data and the second revenue data are not matched, outputting a refusal result; and if the first income data are not matched and the second income data are matched, outputting first prompt information, wherein the first prompt information is used for indicating that income demonstration needs to be provided.
After obtaining the score for the user, the computer device may output an approval result based on the obtained score. Optionally, when the score of the user is greater than or equal to a preset second threshold, the computer device outputs a first approval result, where the first approval result is used to characterize that the loan approval passes; and when the score of the user is smaller than the second threshold value, the computer equipment outputs a second approval result, wherein the second approval result is used for indicating approval of the loan through an offline approval process.
According to the loan approval method based on machine learning, which is provided by the embodiment of the application, the computer equipment acquires loan application information input by a user and equipment information of input equipment currently used by the user; acquiring external data of the user when the repayment probability of the user is determined to be greater than a preset first threshold value according to the borrower information, the equipment information, the historical transaction behavior information of the user and a preset risk early warning model in the loan application information; and determining the score of the user according to the loan application information, the external data and a preset scoring rule, and outputting an approval result according to the score. In the whole loan approval process, the computer equipment can automatically determine the repayment probability of the user through a machine learning model, namely a risk early warning model, based on loan application information input by the user, equipment information of the input equipment and historical transaction behavior information of the user, and when the repayment probability of the user is greater than a first threshold value, the computer equipment can determine the score of the user through a preset scoring rule based on the acquired external data of the user and the loan application information of the user, and output an approval result according to the determined score. Namely, the whole loan approval process is automatically completed by the computer equipment, and a customer manager is not required to conduct offline approval based on experience, so that the approval efficiency of the loan is improved, and the approval risk brought by manually approving the loan is reduced.
In the practical application process, the loan application information may further include house information of a house to be purchased, in order to further reduce the risk of loan approval, so as to provide the user with a credit amount matched with the credit amount, and optionally, the process of outputting the approval result by the computer device according to the score may further be: determining house properties of the house to be purchased according to the house information when the score is greater than or equal to the second threshold; when the house to be purchased is a second house, judging whether the purchase price of the house to be purchased is matched with the market price; if so, outputting the first trial result; if the first examination result is not matched with the second examination result, outputting second prompt information, wherein the first examination result is used for representing that the loan examination and approval passes, and the second prompt information is used for indicating that a house evaluation report of the house to be purchased needs to be provided.
Fig. 2 is a schematic flow chart of another loan approval method based on machine learning according to an embodiment of the application. The embodiment relates to a specific process of how the computer equipment determines the repayment probability of the user based on the risk early warning model. Based on the above embodiment, optionally, as shown in fig. 2, the process of determining the repayment probability of the user according to the borrower information, the device information, the historical transaction behavior information of the user, and the preset risk early warning model in S102 may be:
S201, constructing an information relation network of the user according to the borrower information, the equipment information and the historical transaction behavior information of the user.
Specifically, the computer device converts the obtained loaner information, the device information of the input device and the historical transaction behavior information of the user in the loan application information into graph data by using unstructured data extraction technologies such as texts, maps and the like, processes the graph data, screens out nodes and edges required for constructing an information relationship network, and constructs the information relationship network of the user according to the nodes and edges. Wherein, the node refers to the original data related to the user. For example, the nodes are an account number of the user, a device identifier (such as a device MAC address, an IP address, an IMEI, and an IMSI of a mobile phone card) of an input device currently used by the user, a mobile phone number, an identity card number, current location information, a home address, a company address, and the like. The side refers to the association relationship between nodes, such as the account owned by the user, the user using an input device, whether the user using a mobile phone card, a home address or a work address matches the current location information, whether the application mobile phone number when applying for loans matches the bank card account opening mobile phone number, and whether the resources in one account are transferred to another account. In this way, the computer device can build an information relationship network as shown in FIG. 3. The client ID (Identification) in fig. 3 is a user identifier, and the global positioning system (Global Positioning System, GPS) address is the current location information of the user.
S202, extracting risk characteristic information corresponding to the user from the information relation network.
Specifically, the computer device adopts a dual-mode network feature extraction framework to extract risk feature information corresponding to the user from the information relationship network. The dual-mode network feature extraction framework may include a self-central network risk feature group, a local network cluster coefficient risk feature group, a global network risk feature group, a network information consistency risk feature group, and a quadrangle risk feature group. The computer equipment can extract the number of input equipment used by the user and the number of common use people of the input equipment in preset time through a self-centering network risk feature group, can extract the transaction times existing between the user and other users through a local network clustering coefficient risk feature group, can extract the relative importance degree, the central degree and the like of the user in a transaction network through a global network risk feature group, can extract whether the mobile phone number of the user is a common mobile phone number, the times of shifting the current position information and the home address by a preset distance and the like through a network information consistency risk feature group, and can extract the common account number of the input equipment and the times of common application loan of the user in the current position in preset time (such as 1 day) through a quadrangle risk feature group.
S203, the risk characteristic information is respectively input into a plurality of preset risk early warning models to obtain corresponding output results.
Specifically, different risk early warning models are used for predicting different repayment risks of the user, wherein the repayment risks comprise any one of overdue probability, repayment capacity and repayment willingness. The computer equipment respectively inputs the obtained risk characteristic information into a plurality of risk early warning models, so that a plurality of output results can be obtained, and a plurality of repayment risks are obtained.
Optionally, the plurality of risk early warning models may include a clustering algorithm, a document theme generation model, a natural language processing model, a long-term and short-term memory model, and other deep learning models. Each risk early warning model is trained through a large amount of historical training data.
S204, weighting calculation is carried out on all output results, and the repayment probability of the user is obtained.
Specifically, different output results have corresponding weight values, so that the computer equipment can perform weighted calculation on all the output results according to each output result and the weight value corresponding to each output result to obtain the repayment probability of the user.
In this embodiment, the computer device may construct an information relationship network of the user based on the obtained borrower information, the device information of the input device, and the historical transaction behavior information of the user, extract risk feature information corresponding to the user from the information relationship network, and input the risk feature information into a plurality of risk early warning models to obtain each repayment risk representing the user, and determine a repayment probability of the user based on each repayment risk. Through the information relation network, the real information of the user can be more comprehensively and deeply known, so that the obtained risk characteristic information is more comprehensive and accurate, the accuracy of the repayment risk predicted by the risk early-warning model is improved, the accuracy of the calculated repayment probability is further improved, and the risk of loan approval is reduced.
Optionally, after obtaining the information relationship network of the user, the computer device may further analyze the information relationship network of the user, determine a target risk identifier, and add the target risk identifier to the blacklist.
Specifically, the computer device determines the target risk identifier by analyzing the information relationship network, i.e., by analyzing similarities and centrality between nodes in the information relationship network. The target risk identifier may be an IP address with risk, a region range with risk, a device number with risk, and the like. The computer equipment adds the IP addresses with risks, the area ranges with risks and the equipment numbers with risks into the existing blacklist to supplement the existing blacklist information, so that the information in the blacklist is more perfect, and in the next loan approval process, the computer equipment can pre-approve loan application users based on the blacklist, so that the risk of loan approval is further reduced.
Fig. 4 is a schematic structural diagram of a loan approval device based on machine learning according to an embodiment of the application. As shown in fig. 4, the apparatus may include: a first acquisition module 10, a second acquisition module 11 and a processing module 12.
Specifically, the first obtaining module 10 is configured to obtain loan application information input by a user and device information of an input device currently used by the user, where the loan application information includes loan information and borrower information;
the second obtaining module 11 is configured to obtain external data of the user when it is determined that the repayment probability of the user is greater than a preset first threshold according to the borrower information, the equipment information, the historical transaction behavior information of the user, and a preset risk early warning model, where the risk early warning model is a machine learning model, and the external data is non-prestored data;
the processing module 12 is configured to determine a score of the user according to the loan application information, the external data, and a preset scoring rule, and output an approval result according to the score.
According to the loan approval device based on machine learning, which is provided by the embodiment of the application, the computer equipment acquires loan application information input by a user and equipment information of input equipment currently used by the user; acquiring external data of the user when the repayment probability of the user is determined to be greater than a preset first threshold value according to the borrower information, the equipment information, the historical transaction behavior information of the user and a preset risk early warning model in the loan application information; and determining the score of the user according to the loan application information, the external data and a preset scoring rule, and outputting an approval result according to the score. In the whole loan approval process, the computer equipment can automatically determine the repayment probability of the user through a machine learning model, namely a risk early warning model, based on loan application information input by the user, equipment information of the input equipment and historical transaction behavior information of the user, and when the repayment probability of the user is greater than a first threshold value, the computer equipment can determine the score of the user through a preset scoring rule based on the acquired external data of the user and the loan application information of the user, and output an approval result according to the determined score. Namely, the whole loan approval process is automatically completed by the computer equipment, and a customer manager is not required to conduct offline approval based on experience, so that the approval efficiency of the loan is improved, and the approval risk brought by manually approving the loan is reduced.
On the basis of the above embodiment, optionally, the apparatus may further include a determining module, where the determining module may include: the device comprises a construction unit, an extraction unit, a prediction unit and a determination unit;
specifically, the construction unit is configured to construct an information relationship network of the user according to the borrower information, the equipment information and the historical transaction behavior information of the user;
the extraction unit is used for extracting risk characteristic information corresponding to the user from the information relation network;
the prediction unit is used for respectively inputting the risk characteristic information into a plurality of preset risk early warning models to obtain corresponding output results, wherein different risk early warning models are used for predicting different repayment risks of the user, and the repayment risks comprise any one of overdue probability, repayment capacity and repayment willingness;
and the determining unit is used for carrying out weighted calculation on all the output results to obtain the repayment probability of the user.
On the basis of the above embodiment, optionally, the determining module may further include a processing unit;
specifically, the processing unit is configured to analyze the information relationship network of the user, determine a target risk identifier, and add the target risk identifier to a blacklist.
On the basis of the above embodiment, optionally, the external data includes first income data of the user, and the processing module 12 is specifically configured to determine, according to a preset mapping relationship, whether the first income data and second income data in the loan application information match with a target loan amount; and if at least one income data is matched, determining the score of the user according to the credit data in the external data and a preset scoring rule, wherein the mapping relation comprises the corresponding relation between different income data and loan amount.
Based on the above embodiment, optionally, the processing module 12 is further configured to output a refusal result if neither the first revenue data nor the second revenue data match; and if the first income data are not matched and the second income data are matched, outputting first prompt information, wherein the first prompt information is used for indicating that income demonstration needs to be provided.
On the basis of the above embodiment, optionally, the processing module 12 is further configured to output a first approval result when the score is greater than or equal to a preset second threshold, where the first approval result is used to characterize that the loan approval passes; and when the score is smaller than the second threshold, outputting a second approval result, wherein the second approval result is used for indicating approval of the loan through an offline approval process.
On the basis of the above embodiment, optionally, the loan application information further includes house information of a house to be purchased, and the processing module 12 is further configured to determine, according to the house information, a house property of the house to be purchased when the score is greater than or equal to the second threshold; when the house to be purchased is a second house, judging whether the purchase price of the house to be purchased is matched with the market price; if so, outputting the first trial result; and if the house evaluation report is not matched with the house evaluation report, outputting second prompt information, wherein the second prompt information is used for indicating that the house evaluation report of the house to be purchased needs to be provided.
In one embodiment, a computer device is provided, the schematic structure of which may be as shown in fig. 5. The computer device includes a processor, a memory, a network interface, and a database connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The database of the computer device is used to store data during machine learning based loan approval. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program, when executed by a processor, implements a machine learning based loan approval method.
It will be appreciated by those skilled in the art that the structure shown in FIG. 5 is merely a block diagram of some of the structures associated with the present inventive arrangements and is not limiting of the computer device to which the present inventive arrangements may be applied, and that a particular computer device may include more or fewer components than shown, or may combine some of the components, or have a different arrangement of components.
In one embodiment, a computer device is provided, the computer device comprising a memory and a processor, the memory having stored therein a computer program, the processor when executing the computer program performing the steps of:
acquiring loan application information input by a user and equipment information of an input device currently used by the user, wherein the loan application information comprises loan information and borrower information;
when the repayment probability of the user is determined to be larger than a preset first threshold value according to the borrower information, the equipment information, the historical transaction behavior information of the user and a preset risk early warning model, external data of the user are obtained, wherein the risk early warning model is a machine learning model, and the external data are non-prestored data;
And determining the score of the user according to the loan application information, the external data and a preset scoring rule, and outputting an approval result according to the score.
In one embodiment, the processor when executing the computer program further performs the steps of: constructing an information relationship network of the user according to the borrower information, the equipment information and the historical transaction behavior information of the user; extracting risk characteristic information corresponding to the user from the information relation network; respectively inputting the risk characteristic information into a plurality of preset risk early-warning models to obtain corresponding output results, wherein different risk early-warning models are used for predicting different repayment risks of the user, and the repayment risks comprise any one of overdue probability, repayment capacity and repayment willingness; and carrying out weighted calculation on all output results to obtain the repayment probability of the user.
In one embodiment, the processor when executing the computer program further performs the steps of: and analyzing the information relation network of the user, determining a target risk identifier, and adding the target risk identifier into a blacklist.
In one embodiment, the external data comprises first revenue data for the user, the processor when executing the computer program further implementing the steps of: respectively judging whether the first income data and the second income data in the loan application information are matched with the target loan amount according to a preset mapping relation, wherein the mapping relation comprises the corresponding relation between different income data and the loan amount; and if at least one piece of income data is matched, determining the score of the user according to the credit investigation data in the external data by a preset scoring rule.
In one embodiment, the processor when executing the computer program further performs the steps of: outputting a refusal result if the first income data and the second income data are not matched; and if the first income data are not matched and the second income data are matched, outputting first prompt information, wherein the first prompt information is used for indicating that income demonstration needs to be provided.
In one embodiment, the processor when executing the computer program further performs the steps of: when the score is greater than or equal to a preset second threshold, outputting a first examination result, wherein the first examination result is used for representing that the loan examination passes; and when the score is smaller than the second threshold, outputting a second approval result, wherein the second approval result is used for indicating approval of the loan through an offline approval process.
In one embodiment, the loan application information further includes house information of the house to be purchased, and the processor when executing the computer program further performs the steps of: determining house properties of the house to be purchased according to the house information when the score is greater than or equal to the second threshold; when the house to be purchased is a second house, judging whether the purchase price of the house to be purchased is matched with the market price; if so, outputting the first trial result; and if the house evaluation report is not matched with the house evaluation report, outputting second prompt information, wherein the second prompt information is used for indicating that the house evaluation report of the house to be purchased needs to be provided.
In one embodiment, a computer readable storage medium is provided having a computer program stored thereon, which when executed by a processor, performs the steps of:
acquiring loan application information input by a user and equipment information of an input device currently used by the user, wherein the loan application information comprises loan information and borrower information;
when the repayment probability of the user is determined to be larger than a preset first threshold value according to the borrower information, the equipment information, the historical transaction behavior information of the user and a preset risk early warning model, external data of the user are obtained, wherein the risk early warning model is a machine learning model, and the external data are non-prestored data;
And determining the score of the user according to the loan application information, the external data and a preset scoring rule, and outputting an approval result according to the score.
In one embodiment, the computer program when executed by the processor further performs the steps of: constructing an information relationship network of the user according to the borrower information, the equipment information and the historical transaction behavior information of the user; extracting risk characteristic information corresponding to the user from the information relation network; respectively inputting the risk characteristic information into a plurality of preset risk early-warning models to obtain corresponding output results, wherein different risk early-warning models are used for predicting different repayment risks of the user, and the repayment risks comprise any one of overdue probability, repayment capacity and repayment willingness; and carrying out weighted calculation on all output results to obtain the repayment probability of the user.
In one embodiment, the computer program when executed by the processor further performs the steps of: and analyzing the information relation network of the user, determining a target risk identifier, and adding the target risk identifier into a blacklist.
In one embodiment, the external data comprises first revenue data for the user, the computer program when executed by the processor further implementing the steps of: respectively judging whether the first income data and the second income data in the loan application information are matched with the target loan amount according to a preset mapping relation, wherein the mapping relation comprises the corresponding relation between different income data and the loan amount; and if at least one piece of income data is matched, determining the score of the user according to the credit investigation data in the external data by a preset scoring rule.
In one embodiment, the computer program when executed by the processor further performs the steps of: outputting a refusal result if the first income data and the second income data are not matched; and if the first income data are not matched and the second income data are matched, outputting first prompt information, wherein the first prompt information is used for indicating that income demonstration needs to be provided.
In one embodiment, the computer program when executed by the processor further performs the steps of: when the score is greater than or equal to a preset second threshold, outputting a first examination result, wherein the first examination result is used for representing that the loan examination passes; and when the score is smaller than the second threshold, outputting a second approval result, wherein the second approval result is used for indicating approval of the loan through an offline approval process.
In one embodiment, the loan application information further includes house information of the house to be purchased, and the computer program when executed by the processor further performs the steps of: determining house properties of the house to be purchased according to the house information when the score is greater than or equal to the second threshold; when the house to be purchased is a second house, judging whether the purchase price of the house to be purchased is matched with the market price; if so, outputting the first trial result; and if the house evaluation report is not matched with the house evaluation report, outputting second prompt information, wherein the second prompt information is used for indicating that the house evaluation report of the house to be purchased needs to be provided.
The loan approval device, the computer equipment and the storage medium based on machine learning provided in the above embodiment can execute the loan approval method based on machine learning provided in any embodiment of the application, and have the corresponding functional modules and beneficial effects of executing the method. Technical details not described in detail in the above embodiments may be referred to the loan approval method based on machine learning provided in any embodiment of the present application.
Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in embodiments provided herein may include non-volatile and/or volatile memory. The nonvolatile memory can include Read Only Memory (ROM), programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double Data Rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous Link DRAM (SLDRAM), memory bus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), among others.
The technical features of the above-described embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above-described embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.
The foregoing examples illustrate only a few embodiments of the application and are described in detail herein without thereby limiting the scope of the application. It should be noted that it will be apparent to those skilled in the art that several variations and modifications can be made without departing from the spirit of the application, which are all within the scope of the application. Accordingly, the scope of protection of the present application is to be determined by the appended claims.
Claims (8)
1. A machine learning-based loan approval method, comprising:
acquiring loan application information input by a user and equipment information of an input device currently used by the user, wherein the loan application information comprises loan information and borrower information;
when the repayment probability of the user is determined to be larger than a preset first threshold value according to the borrower information, the equipment information, the historical transaction behavior information of the user and a preset risk early warning model, external data of the user are obtained, wherein the risk early warning model is a machine learning model, and the external data are non-prestored data;
Determining the score of the user according to the loan application information, the external data and a preset scoring rule, and outputting an approval result according to the score;
wherein the determining the repayment probability of the user according to the borrower information, the equipment information, the historical transaction behavior information of the user and a preset risk early warning model comprises:
constructing an information relationship network of the user according to the borrower information, the equipment information and the historical transaction behavior information of the user;
extracting risk characteristic information corresponding to the user from the information relation network;
respectively inputting the risk characteristic information into a plurality of preset risk early-warning models to obtain corresponding output results, wherein different risk early-warning models are used for predicting different repayment risks of the user, and the repayment risks comprise any one of overdue probability, repayment capacity and repayment willingness;
weighting calculation is carried out on all output results to obtain repayment probability of the user;
correspondingly, the extracting risk characteristic information corresponding to the user from the information relation network comprises the following steps:
extracting the number of used input devices of the user and the number of commonly used people of the input devices within preset time from the information relation network through a self-centering network risk feature family;
Extracting the transaction times existing between the user and other users from the information relation network through a local network clustering coefficient risk characteristic family;
extracting the relative importance degree and the central degree of the user in the transaction network from the information relation network through a global network risk feature family;
extracting whether the mobile phone number of the user is a common mobile phone number or not and the number of times that the current position information and the home address deviate by a preset distance from the information relation network through a network information consistency risk feature family;
extracting the number of common account numbers of the input equipment from the information relation network through a quadrangle risk feature family, and the number of times that the user applies for loans at the current position within preset time;
the external data includes first income data of the user, and the determining the score of the user according to the loan application information, the external data and a preset scoring rule includes:
respectively judging whether the first income data and the second income data in the loan application information are matched with the target loan amount according to a preset mapping relation, wherein the mapping relation comprises the corresponding relation between different income data and the loan amount;
And if at least one piece of income data is matched, determining the score of the user according to the credit investigation data in the external data by a preset scoring rule.
2. The method as recited in claim 1, further comprising:
and analyzing the information relation network of the user, determining a target risk identifier, and adding the target risk identifier into a blacklist.
3. The method as recited in claim 1, further comprising:
outputting a refusal result if the first income data and the second income data are not matched;
and if the first income data are not matched and the second income data are matched, outputting first prompt information, wherein the first prompt information is used for indicating that income demonstration needs to be provided.
4. A method according to any one of claims 1 to 3, wherein said outputting an approval result according to said score comprises:
when the score is greater than or equal to a preset second threshold, outputting a first examination result, wherein the first examination result is used for representing that the loan examination passes;
and when the score is smaller than the second threshold, outputting a second approval result, wherein the second approval result is used for indicating approval of the loan through an offline approval process.
5. The method of claim 4, wherein the loan application information further comprises house information for a house to be purchased, the method further comprising:
determining house properties of the house to be purchased according to the house information when the score is greater than or equal to the second threshold;
when the house to be purchased is a second house, judging whether the purchase price of the house to be purchased is matched with the market price;
if so, outputting the first trial result;
and if the house evaluation report is not matched with the house evaluation report, outputting second prompt information, wherein the second prompt information is used for indicating that the house evaluation report of the house to be purchased needs to be provided.
6. A loan approval device based on machine learning, comprising:
the first acquisition module is used for acquiring loan application information input by a user and equipment information of an input equipment currently used by the user, wherein the loan application information comprises loan information and borrower information;
the second acquisition module is used for acquiring external data of the user when the repayment probability of the user is determined to be greater than a preset first threshold value according to the borrower information, the equipment information, the historical transaction behavior information of the user and a preset risk early warning model, wherein the risk early warning model is a machine learning model, and the external data is non-prestored data; the external data includes first revenue data for the user;
The processing module is used for determining the scores of the users according to the loan application information, the external data and preset scoring rules and outputting approval results according to the scores;
the determining module is used for constructing an information relation network of the user according to the borrower information, the equipment information and the historical transaction behavior information of the user; extracting risk characteristic information corresponding to the user from the information relation network; respectively inputting the risk characteristic information into a plurality of preset risk early-warning models to obtain corresponding output results, wherein different risk early-warning models are used for predicting different repayment risks of the user, and the repayment risks comprise any one of overdue probability, repayment capacity and repayment willingness; weighting calculation is carried out on all output results to obtain repayment probability of the user;
the system comprises a determining module, a processing module and a processing module, wherein the determining module is specifically used for extracting the number of used input devices of the user and the number of commonly used people of the input devices in preset time from the information relation network through a self-centering network risk feature family; extracting the transaction times existing between the user and other users from the information relation network through a local network clustering coefficient risk characteristic family; extracting the relative importance degree and the central degree of the user in the transaction network from the information relation network through a global network risk feature family; extracting whether the mobile phone number of the user is a common mobile phone number or not and the number of times that the current position information and the home address deviate by a preset distance from the information relation network through a network information consistency risk feature family; extracting the number of common account numbers of the input equipment from the information relation network through a quadrangle risk feature family, and the number of times that the user applies for loans at the current position within preset time;
The processing module is specifically configured to respectively determine whether the first income data and the second income data in the loan application information are matched with a target loan amount according to a preset mapping relationship; if at least one income data is matched, determining the score of the user according to the credit investigation data in the external data by a preset scoring rule; wherein the mapping relation comprises the corresponding relation between different income data and loan amount.
7. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the method of any one of claims 1 to 5 when the computer program is executed.
8. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the method according to any one of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010147886.8A CN111369342B (en) | 2020-03-05 | 2020-03-05 | Loan approval method, device, equipment and storage medium based on machine learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010147886.8A CN111369342B (en) | 2020-03-05 | 2020-03-05 | Loan approval method, device, equipment and storage medium based on machine learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111369342A CN111369342A (en) | 2020-07-03 |
| CN111369342B true CN111369342B (en) | 2023-10-17 |
Family
ID=71211727
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010147886.8A Active CN111369342B (en) | 2020-03-05 | 2020-03-05 | Loan approval method, device, equipment and storage medium based on machine learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111369342B (en) |
Families Citing this family (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111813399B (en) * | 2020-07-23 | 2022-05-31 | 平安医疗健康管理股份有限公司 | Machine learning-based auditing rule processing method and device and computer equipment |
| CN113034263A (en) * | 2020-08-20 | 2021-06-25 | 深圳大学 | Credit approval method, credit approval device, credit approval server and credit approval medium |
| CN111932377A (en) * | 2020-08-21 | 2020-11-13 | 深圳技术大学 | Asset security product early warning method and device, electronic equipment and storage medium |
| CN112037039A (en) * | 2020-09-02 | 2020-12-04 | 中国银行股份有限公司 | Loan assessment method and device |
| CN112200656A (en) * | 2020-09-17 | 2021-01-08 | 中国建设银行股份有限公司 | On-line pre-approval method, device, medium and electronic equipment for house loan |
| CN112163944A (en) * | 2020-09-18 | 2021-01-01 | 中国建设银行股份有限公司 | Loan qualification scoring method and device for customer, computer equipment and storage medium |
| CN112184154A (en) * | 2020-09-23 | 2021-01-05 | 中国建设银行股份有限公司 | Business approval method and device |
| CN112184417A (en) * | 2020-09-25 | 2021-01-05 | 中国建设银行股份有限公司 | Business approval method, device, medium and electronic equipment |
| CN112529691A (en) * | 2020-12-16 | 2021-03-19 | 深圳市卡牛科技有限公司 | Behavior risk result display method, behavior risk result display device, terminal and storage medium |
| CN112651822A (en) * | 2020-12-22 | 2021-04-13 | 李福福 | Financial big data processing method based on block chain |
| CN112767131A (en) * | 2021-01-26 | 2021-05-07 | 福建卓融信息技术有限公司 | Loan approval-based risk control method and device |
| CN112862472A (en) * | 2021-03-15 | 2021-05-28 | 深圳市心版图科技有限公司 | Loan payment system based on business display industry chain and loan payment judgment method |
| CN113393317A (en) * | 2021-06-10 | 2021-09-14 | 罗忠明 | Automobile financial loan wind control system based on algorithm, big data and block chain |
| CN114282988A (en) * | 2021-12-27 | 2022-04-05 | 深圳前海微众银行股份有限公司 | Abnormal application identification method and equipment |
| CN115187393A (en) * | 2022-09-14 | 2022-10-14 | 深圳市明源云科技有限公司 | Loan risk detection method and device, electronic equipment and readable storage medium |
| CN115760368A (en) * | 2022-11-24 | 2023-03-07 | 中电金信软件有限公司 | Credit business approval method and device and electronic equipment |
| CN116151965B (en) * | 2023-04-04 | 2023-07-18 | 成都新希望金融信息有限公司 | Risk feature extraction method and device, electronic equipment and storage medium |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106600399A (en) * | 2016-12-05 | 2017-04-26 | 深圳前海微众银行股份有限公司 | Loan approval method and system |
| CN107657525A (en) * | 2017-08-29 | 2018-02-02 | 深圳市佰仟金融服务有限公司 | One kind loan measures and procedures for the examination and approval and server |
| CN108846520A (en) * | 2018-06-22 | 2018-11-20 | 北京京东金融科技控股有限公司 | Overdue loan prediction technique, device and computer readable storage medium |
| CN108876588A (en) * | 2018-04-28 | 2018-11-23 | 重庆小雨点小额贷款有限公司 | A kind of loan measures and procedures for the examination and approval, device, server and storage medium |
| CN109670936A (en) * | 2018-09-26 | 2019-04-23 | 深圳壹账通智能科技有限公司 | Loan examination & approval processing method, platform, equipment and computer readable storage medium |
| CN110276677A (en) * | 2019-04-24 | 2019-09-24 | 武汉众邦银行股份有限公司 | Refund prediction technique, device, equipment and storage medium based on big data platform |
-
2020
- 2020-03-05 CN CN202010147886.8A patent/CN111369342B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106600399A (en) * | 2016-12-05 | 2017-04-26 | 深圳前海微众银行股份有限公司 | Loan approval method and system |
| CN107657525A (en) * | 2017-08-29 | 2018-02-02 | 深圳市佰仟金融服务有限公司 | One kind loan measures and procedures for the examination and approval and server |
| CN108876588A (en) * | 2018-04-28 | 2018-11-23 | 重庆小雨点小额贷款有限公司 | A kind of loan measures and procedures for the examination and approval, device, server and storage medium |
| CN108846520A (en) * | 2018-06-22 | 2018-11-20 | 北京京东金融科技控股有限公司 | Overdue loan prediction technique, device and computer readable storage medium |
| CN109670936A (en) * | 2018-09-26 | 2019-04-23 | 深圳壹账通智能科技有限公司 | Loan examination & approval processing method, platform, equipment and computer readable storage medium |
| CN110276677A (en) * | 2019-04-24 | 2019-09-24 | 武汉众邦银行股份有限公司 | Refund prediction technique, device, equipment and storage medium based on big data platform |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111369342A (en) | 2020-07-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111369342B (en) | Loan approval method, device, equipment and storage medium based on machine learning | |
| CN109165840B (en) | Risk prediction processing method, risk prediction processing device, computer equipment and medium | |
| CN109272396B (en) | Customer risk early warning method, device, computer equipment and medium | |
| CN109345374B (en) | Risk control method and device, computer equipment and storage medium | |
| CN108876133B (en) | Risk assessment processing method, device, server and medium based on business information | |
| CN109543925B (en) | Risk prediction method and device based on machine learning, computer equipment and storage medium | |
| CN109063921B (en) | Optimization processing method and device for client risk early warning, computer equipment and medium | |
| CN111507638A (en) | Risk information output and risk information construction method and device | |
| US20220300903A1 (en) | System and method for dynamically predicting fraud using machine learning | |
| CN109949154A (en) | Customer information classification method, device, computer equipment and storage medium | |
| CN109523117A (en) | Risk Forecast Method, device, computer equipment and storage medium | |
| CN109377347B (en) | Network credit early warning method and system based on feature selection and electronic equipment | |
| CN110135943B (en) | Product recommendation method, device, computer equipment and storage medium | |
| CN110619065A (en) | Resource scheduling service processing method and device, computer equipment and storage medium | |
| CN111259952A (en) | Abnormal user identification method and device, computer equipment and storage medium | |
| CN114186760A (en) | Analysis method and system for stable operation of enterprise and readable storage medium | |
| AU2016364120A1 (en) | User data sharing method and device | |
| CN116151954A (en) | Real-time group-partner anti-fraud detection method and system | |
| CN114493902A (en) | Multi-mode information anomaly monitoring method and device, computer equipment and storage medium | |
| CN112417315A (en) | User portrait generation method, device, equipment and medium based on website registration | |
| CN109711984B (en) | Pre-loan risk monitoring method and device based on collection urging | |
| CN109241249B (en) | Method and device for determining burst problem | |
| CN114201650A (en) | Method and device for determining equipment fingerprint homology and computer equipment | |
| CN110517133B (en) | Associated account searching method and device, computer equipment and storage medium | |
| CN111506486B (en) | Data processing method and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20220920 Address after: 25 Financial Street, Xicheng District, Beijing 100033 Applicant after: CHINA CONSTRUCTION BANK Corp. Address before: 25 Financial Street, Xicheng District, Beijing 100033 Applicant before: CHINA CONSTRUCTION BANK Corp. Applicant before: Jianxin Financial Science and Technology Co.,Ltd. |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |