A credit installment method and system based on big data

Abstract

The present invention discloses a credit installment method and system based on big data, which relates to the technical field of financial installment business of operators, obtains the identity information of customers, performs credit inquiry according to the identity information of customers, judges whether the customer credit reaches a first set value, and if so, generates a customer credit safety index; if not, screens the customer's credit inquiry information, and performs big data analysis on the screened customer financial data, and summarizes and generates a visualized customer financial record table; performs risk assessment on customers according to the visualized customer financial record table, obtains the customer financial risk assessment index, and judges whether the customer financial risk assessment index is lower than a second set value; if not, refuses to handle the customer financial installment business; if so, generates a corresponding customer credit limit according to the customer credit safety index and the customer financial risk assessment index, and the operator completes the handling of the customer financial installment business according to the customer credit limit.

Description

Technical Field

[0001] The present invention relates to the technical field of operator financial installment services, and in particular to a credit installment method and system based on big data. Background Art

[0002] Financial installment refers to a financial service provided by an operator, in which a user uses his or her personal credit to repay a loan or purchase a designated product in installments. In real life, as people's living standards continue to improve, financial installment services have become a trend for more and more users. However, at present, since operators are unable to conduct big data risk assessments on users, financial installment services are only provided to individual users who have never defaulted. On the one hand, this will affect the economic benefits of operators, and on the other hand, it will also affect those users who are in urgent need of funds. Therefore, the present invention provides a credit installment method and system based on big data to overcome the above problems. Summary of the invention

[0003] In view of the defects in the prior art, the present invention provides a credit installment method and system based on big data

[0004] To solve the problems raised by the background technology, the technical solution of the present invention is as follows:

[0005] A financial installment method based on big data includes the following steps:

[0006] S1: The operator obtains the customer's identity information according to the customer's financial installment business processing needs, performs a credit query based on the customer's identity information, and determines whether the customer's credit reaches a first set value. If so, a customer credit security index is generated and the process proceeds to step S4; if not, the process proceeds to step S2;

[0007] S2: Screen the customer's credit inquiry information to obtain the customer's financial data, perform big data analysis on the customer's financial data, and summarize the big data analysis data to generate a visual customer financial record table, and then proceed to step S3;

[0008] S3: Perform risk assessment on the customer according to the visualized customer financial record table to obtain the customer financial risk assessment index, and determine whether the customer financial risk assessment index is lower than the second set value. If so, proceed to step S4; if not, refuse to process the customer's financial installment business;

[0009] S4: Generate the corresponding customer credit line based on the customer credit security index and the customer financial risk assessment index. The operator completes the processing of the customer's financial installment business based on the customer's credit line.

[0010] Preferably, the big data analysis described in step S2 includes transaction behavior analysis, interest and hobby analysis, consumption capacity analysis, loan term analysis, overdue days analysis, education level analysis, and collection difficulty analysis.

[0011] Preferably, the customer's financial data in step S2 includes customer spending capacity data, customer Internet transaction data, and customer loan data.

[0012] Preferably, the risk assessment described in step S3 is specifically to perform risk assessment on the customer through a Bayes neural network model, and its mathematical expression is as follows:

[0013]

[0014]

[0015] Where p() is the probability density function; p(z|x) is the posterior probability; p(x|z) is the joint probability; p(x|z) is the likelihood; p(z) is the prior probability; and p(x) is the evidence.

[0016] Preferably, the risk assessment of the customer in step S3 is performed as follows:

[0017] S51: Obtain the operator's historical customer sample set;

[0018] S52: Establishing a Bayesian neural network model, inputting the historical customer sample set into the Bayesian neural network model for training, and obtaining a trained Bayesian neural network model;

[0019] S53: Calculate the visualized customer financial record table through the trained Bayes neural network model to obtain the customer financial risk assessment index.

[0020] Preferably, the Bayes neural network model described in step S52 has a neural network structure of an MLP model, and the variational distribution of the Bayes neural network parameters is solved by Kullback-Leibler divergence.

[0021] A credit installment system based on big data, comprising: a data collection end, a credit query end, a data storage end, a big data analysis end, a data aggregation end, a risk prediction end, a credit limit calculation end, and a business execution end; the data collection end is connected to the credit query end; the credit query end is respectively connected to the credit limit calculation end and the data storage end; the data storage end is connected to the big data analysis end; the big data analysis end is connected to the data aggregation end; the data aggregation end is connected to the risk prediction end; the risk prediction end is connected to the credit limit calculation end; the credit limit calculation end is connected to the business execution end.

[0022] The beneficial effects of the present invention are as follows: the present invention proposes a credit installment method and system based on big data, which not only overcomes the problem in the prior art that operators are unable to conduct big data risk assessment on users, resulting in financial installment services being only provided to individual users who have never defaulted; but also provides good market prospects for future financial installment services. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following is a brief introduction to the drawings required for the specific embodiments or the description of the prior art. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn according to the actual scale.

[0024] Figure 1 A schematic diagram of the process provided by the present invention;

[0025] Figure 2 A schematic diagram of the system provided by the present invention. DETAILED DESCRIPTION

[0026] The following embodiments of the technical solution of the present invention are described in detail in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and are therefore only used as examples, and cannot be used to limit the protection scope of the present invention.

[0027] It should be noted that, unless otherwise specified, the technical terms or scientific terms used in this application should have the common meanings understood by those skilled in the art to which the present invention belongs.

[0028] like Figure 1 As shown, Figure 1 As shown, a financial installment method based on big data includes the following steps:

[0029] S1: The operator obtains the customer's identity information according to the customer's financial installment business processing needs, performs a credit query based on the customer's identity information, and determines whether the customer's credit reaches a first set value. If so, a customer credit security index is generated and the process proceeds to step S4; if not, the process proceeds to step S2;

[0030] S2: Screen the customer's credit inquiry information to obtain the customer's financial data, perform big data analysis on the customer's financial data, and summarize the big data analysis data to generate a visual customer financial record table, and then proceed to step S3;

[0031] S3: Perform risk assessment on the customer according to the visualized customer financial record table to obtain the customer financial risk assessment index, and determine whether the customer financial risk assessment index is lower than the second set value. If so, proceed to step S4; if not, refuse to process the customer's financial installment business;

[0032] S4: Generate the corresponding customer credit line based on the customer credit security index and the customer financial risk assessment index. The operator completes the processing of the customer's financial installment business based on the customer's credit line.

[0033] Preferably, the big data analysis described in step S2 includes transaction behavior analysis, interest and hobby analysis, consumption capacity analysis, loan term analysis, overdue days analysis, education level analysis, and collection difficulty analysis.

[0034] Preferably, the customer's financial data in step S2 includes customer spending capacity data, customer Internet transaction data, and customer loan data.

[0035] Preferably, the risk assessment described in step S3 is specifically to perform risk assessment on the customer through a Bayes neural network model, and its mathematical expression is as follows:

[0036]

[0037]

[0038] Where p() is the probability density function; p(z|x) is the posterior probability; p(x, z) is the joint probability; p(x|z) is the likelihood; p(z) is the prior probability; and p(x) is the evidence.

[0039] Preferably, the risk assessment of the customer in step S3 is performed as follows:

[0040] S51: Obtain the operator's historical customer sample set;

[0041] S52: Establishing a Bayesian neural network model, inputting the historical customer sample set into the Bayesian neural network model for training, and obtaining a trained Bayesian neural network model;

[0042] S53: Calculate the visualized customer financial record table through the trained Bayes neural network model to obtain the customer financial risk assessment index.

[0043] Bayesian neural network is a combination of deep neural network and Bayesian reasoning. The deep neural network structure adopted in the present invention is a multi-layer perceptron MLP model. Bayesian neural network assumes that the network parameters of the MLP model obey the posterior distribution rather than a unique fixed value. Different from the feedforward calculation of neural network, the mathematical expression of Bayesian neural network prediction is shown in the following formula, and the prediction result is obtained by Monte Carlo sampling and solving the output expectation.

[0044] Because the posterior distribution of the Bayesian neural network parameters is not convenient to obtain directly, it is necessary to combine variational reasoning to obtain the variational distribution of the network parameters by minimizing the Kullback-Leibler divergence between the posterior distribution and the variational distribution, and use the variational distribution to approximate the posterior distribution. The mathematical expression of Kullback-Leibler divergence is:

[0045]

[0046] Reparameterization is used to train and optimize the variational distribution parameters. The variational posterior distribution of the weight parameters of the neural network is an independent Gaussian distribution. Each weight parameter can be obtained by random sampling based on the mean μ and standard deviation σ of the variational posterior Gaussian distribution. In order to ensure that the standard deviation is non-negative, the standard deviation is further parameterized.

[0047] Preferably, the Bayes neural network model described in step S52 has a neural network structure of an MLP model, and the variational distribution of the Bayes neural network parameters is solved by Kullback-Leibler divergence.

[0048] like Figure 2 As shown, a credit installment system based on big data includes: a data collection end, a credit query end, a data storage end, a big data analysis end, a data aggregation end, a risk prediction end, a credit limit calculation end, and a business execution end; the data collection end is connected to the credit query end; the credit query end is connected to the credit limit calculation end and the data storage end respectively; the data storage end is connected to the big data analysis end; the big data analysis end is connected to the data aggregation end; the data aggregation end is connected to the risk prediction end; the risk prediction end is connected to the credit limit calculation end; the credit limit calculation end is connected to the business execution end.

[0049] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. These modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be included in the scope of the claims and specification of the present invention.

Claims

1. A credit installment method based on big data, characterized in that: The steps include: S1: The operator obtains the customer's identity information according to the customer's financial installment business processing needs, performs a credit query based on the customer's identity information, and determines whether the customer's credit reaches a first set value. If so, a customer credit security index is generated and the process proceeds to step S4; if not, the process proceeds to step S2; S2: Screen the customer's credit inquiry information to obtain the customer's financial data, perform big data analysis on the customer's financial data, and summarize the big data analysis data to generate a visual customer financial record table, and then proceed to step S3; S3: Perform risk assessment on the customer according to the visualized customer financial record table to obtain the customer financial risk assessment index, and determine whether the customer financial risk assessment index is lower than the second set value. If so, proceed to step S4; if not, refuse to handle the customer's financial installment business; S4: Generate the corresponding customer credit line based on the customer credit security index and the customer financial risk assessment index, and the operator completes the processing of the customer's financial installment business based on the customer's credit line; The risk assessment described in step S3 is specifically to conduct risk assessment on customers through a Bayesian neural network model, and its mathematical expression is as follows: Where p() is the probability density function; p(z|x) is the posterior probability; p(x, z) is the joint probability; p(x|z) is the likelihood; p(z) is the prior probability; p(x) is the evidence; The process of risk assessment of the customer described in step S3 is as follows: S51: Obtain the operator's historical customer sample set; S52: Establishing a Bayesian neural network model, inputting the historical customer sample set into the Bayesian neural network model for training, and obtaining a trained Bayesian neural network model; S53: Calculate the visualized customer financial record table through the trained Bayes neural network model to obtain the customer financial risk assessment index.

2. A credit installment method based on big data according to claim 1, characterized in that: The big data analysis described in step S2 includes transaction behavior analysis, interest and hobby analysis, consumption capacity analysis, loan term analysis, overdue days analysis, education level analysis, and collection difficulty analysis.

3. The credit installment method based on big data according to claim 1, characterized in that: The customer's financial data in step S2 includes customer spending capacity data, customer Internet transaction data, and customer loan data.

4. The credit installment method based on big data according to claim 1, characterized in that: The Bayesian neural network model described in step S52 has a neural network structure of an MLP model, and the variational distribution of the Bayesian neural network parameters is solved by Kullback-Leibler divergence.

5. A credit installment system based on big data, characterized in that: The credit installment system is applicable to the method described in any one of claims 1 to 4, comprising: a data collection end, a credit query end, a data storage end, a big data analysis end, a data aggregation end, a risk prediction end, a credit limit calculation end, and a business execution end; the data collection end is connected to the credit query end; the credit query end is respectively connected to the credit limit calculation end and the data storage end; the data storage end is connected to the big data analysis end; the big data analysis end is connected to the data aggregation end; the data aggregation end is connected to the risk prediction end; the risk prediction end is connected to the credit limit calculation end; the credit limit calculation end is connected to the business execution end.

Images