CN116228405A

CN116228405A - Credit classification method, apparatus, device, medium and program product

Info

Publication number: CN116228405A
Application number: CN202310267676.6A
Authority: CN
Inventors: 宋媛媛; 杨彬; 朱建强; 袁宝
Original assignee: Industrial and Commercial Bank of China Ltd ICBC
Current assignee: Industrial and Commercial Bank of China Ltd ICBC
Priority date: 2023-03-20
Filing date: 2023-03-20
Publication date: 2023-06-06

Abstract

The present disclosure provides a credit classification method, which may be applied to the technical field of artificial intelligence. The credit classifying method comprises the following steps: acquiring credit data of a lender, wherein the credit data comprises data sets with N dimensions, and N is greater than or equal to 1; converting the data sets of the N dimensions into N credit values according to preset data mining logic based on the data sets of the N dimensions, wherein the preset data mining logic comprises N preset data mining sub-logics, and the preset data mining sub-logics are in one-to-one correspondence with the dimensions of the data sets; and outputting credit classification labels of lenders through a preset machine learning model based on the N credit values. The present disclosure also provides a credit classifying apparatus, device, storage medium and program product.

Description

Credit classification method, apparatus, device, medium and program product

Technical Field

The present disclosure relates to the field of artificial intelligence, and in particular to a credit classification method, apparatus, device, medium and program product.

Background

Loan transaction is an important property transaction of a bank, which is a lending act in which a commercial bank provides a certain amount of funds to a lender for use in accordance with a certain loan policy with the identity of the lender. Due to the complexity and uncertainty of loan transactions, there is a need to reduce the risk of loans by a variety of means.

In order to reduce the risk of loan business, in the current scheme, professional evaluators are mostly adopted to perform comprehensive credit analysis through a professional credit analysis method, and determine whether to give a loan to a debtor and provide corresponding loan conditions according to the comprehensive credit analysis. Although the loan risk condition of the debtor can be comprehensively estimated by a professional comprehensive credit analysis method, the scheme cannot be deeply quantized, and the estimation process is based on personnel estimation, so that the estimation time is long, and the smooth release of the loan is influenced.

Disclosure of Invention

In view of the foregoing, the present disclosure provides credit classification methods, apparatuses, devices, media, and program products that improve evaluation efficiency and evaluation accuracy.

According to a first aspect of the present disclosure, there is provided a credit classifying method comprising: acquiring credit data of a lender, wherein the credit data comprises data sets with N dimensions, and N is greater than or equal to 1; converting the data sets of the N dimensions into N credit values according to preset data mining logic based on the data sets of the N dimensions, wherein the preset data mining logic comprises N preset data mining sub-logics, and the preset data mining sub-logics are in one-to-one correspondence with the dimensions of the data sets; and outputting credit classification labels of lenders through a preset machine learning model based on the N credit values.

According to an embodiment of the present disclosure, the obtaining credit data of the lender includes: receiving a loan request from the lender; analyzing the loan request to obtain a loan main body and a repayment year; and obtaining the credit data of the lender based on the lender and the repayment age.

According to an embodiment of the present disclosure, the N dimensions include at least: the step of obtaining the credit data of the lender based on the lender body and the repayment year comprises the steps of: for the grid dimension, acquiring reference grid data and real-time grid data, wherein the reference grid data is acquired based on a first acquisition time period, the reference grid data comprises first repayment credit, first repayment amount and first repayment times, the real-time grid data is acquired based on a second acquisition time period, and the real-time grid data comprises second repayment amount and second repayment times; the converting the data set of the N dimensions into N credit values according to preset data mining logic, including: and calculating a lattice credit value based on the first repayment credit, the first repayment amount and the first repayment times and combining the second repayment amount and the second repayment times.

According to an embodiment of the present disclosure, the obtaining the credit data of the lender based on the lender and the repayment year includes: for the capability dimension, obtaining streaming data, the streaming data being acquired based on the first acquisition time period; the converting the data set of the N dimensions into N credit values according to preset data mining logic, including: based on the pipeline data, a capacity credit value is calculated.

According to an embodiment of the present disclosure, the obtaining the credit data of the lender based on the lender and the repayment year includes: for the asset dimension, collecting long-term fixed asset data and long-term liability data based on a third collection time period, and collecting short-term liquidity asset data and short-term liability data based on a fourth collection time period; the converting the data set of the N dimensions into N credit values according to preset data mining logic, including: and calculating an asset credit value based on the long-term fixed asset data and the long-term liability data in combination with the short-term liquidity asset data and the short-term liability data.

According to an embodiment of the present disclosure, the obtaining the credit data of the lender based on the lender and the repayment year includes: acquiring, for the guarantee dimension, a guarantee property price and a guarantee person revenue cash flow, wherein the guarantee person revenue cash flow is collected based on the first collection period or the guarantee person revenue cash flow is collected based on the third collection period; the converting the data set of the N dimensions into N credit values according to preset data mining logic, including: a vouching credit value is calculated based on the vouching property price and the revenue cash flow.

According to an embodiment of the present disclosure, the obtaining the credit data of the lender based on the lender and the repayment year includes: for the dimension of the environmental condition, acquiring stock price data, stock price ratio data, industry indexes and comprehensive indexes; the converting the data set of the N dimensions into N credit values according to preset data mining logic, including: and calculating an environmental condition credit value based on the stock price data, the stock price ratio data, the industry index and the comprehensive index.

According to an embodiment of the disclosure, the outputting, based on the N credit values, a credit classification label of the lender through a preset machine learning model includes: generating an N-dimensional vector corresponding to the lender based on the N modulo vectors; and outputting credit classification labels of lenders through a preset KNN model based on the N-dimensional vector.

In a second aspect of the present disclosure, there is provided a credit classifying apparatus including: the credit data acquisition module is used for acquiring credit data of lenders, wherein the credit data comprises data sets with N dimensions, and N is greater than or equal to 1; the data mining processing module is used for converting the data sets of the N dimensions into N credit values according to preset data mining logic based on the data sets of the N dimensions, wherein the preset data mining logic comprises N preset data mining sub-logics, and the preset data mining sub-logics are in one-to-one correspondence with the dimensions of the data sets; and the credit classification label output module is used for outputting credit classification labels of lenders through a preset machine learning model based on the N credit values.

According to an embodiment of the disclosure, the credit data acquisition module is further configured to receive a loan request from the lender; analyzing the loan request to obtain a loan main body and a repayment year; and obtaining the credit data of the lender based on the lender and the repayment age.

According to an embodiment of the present disclosure, the N dimensions include at least: the credit data acquisition module is further used for acquiring reference grid data and real-time grid data for the grid dimension, wherein the reference grid data are acquired based on a first acquisition time period, the reference grid data comprise first repayment credits, first repayment amounts and first repayment times, the real-time grid data are acquired based on a second acquisition time period, and the real-time grid data comprise second repayment amounts and second repayment times; the credit classification label output module is further configured to calculate a lattice credit value based on the first repayment credit, the first repayment amount, and the first repayment number, and in combination with the second repayment amount and the second repayment number.

According to an embodiment of the disclosure, the credit data acquisition module is further configured to acquire, for the capability dimension, streaming data, the streaming data being acquired based on the first acquisition time period; the converting the data set of the N dimensions into N credit values according to preset data mining logic, including: based on the pipeline data, a capacity credit value is calculated.

According to an embodiment of the disclosure, the credit data acquisition module is further configured to acquire, for the asset dimension, long-term fixed asset data and long-term liability data based on a third acquisition time period, short-term liquidity asset data and short-term liability data based on a fourth acquisition time period; the credit classification tag output module is further configured to calculate an asset credit value based on the long-term fixed asset data and the long-term liability data in combination with the short-term liquidity asset data and the short-term liability data.

According to an embodiment of the disclosure, the credit data collection module is further configured to obtain, for the guarantee dimension, a guarantee property price and a guarantee earner cash flow, where the guarantee earner cash flow is collected based on the first collection period or the guarantee earner cash flow is collected based on the third collection period; the credit category label output module is further configured to calculate a vouching credit value based on the vouching property price and the revenue cash flow.

According to an embodiment of the disclosure, the credit data acquisition module is further configured to acquire, for the environmental condition dimension, stock price data, stock price ratio data, an industry index, and a comprehensive index; the credit classification label output module is further configured to calculate an environmental condition credit value based on the stock price data, the stock price ratio data, the industry index, and the composite index.

According to an embodiment of the disclosure, the credit classification label output module is further configured to generate an N-dimensional vector corresponding to the lender based on the N modulo vectors; and outputting credit classification labels of lenders through a preset KNN model based on the N-dimensional vector.

In a third aspect of the present disclosure, there is provided an electronic device, comprising: one or more processors; and a memory for storing one or more programs, wherein the one or more programs, when executed by the one or more processors, cause the one or more processors to perform the credit classification method described above.

In a fourth aspect of the present disclosure, there is also provided a computer-readable storage medium having stored thereon executable instructions that, when executed by a processor, cause the processor to perform the credit classification method described above.

In a fifth aspect of the present disclosure, there is also provided a computer program product comprising a computer program which, when executed by a processor, implements the credit classification method described above.

In the embodiment of the disclosure, the credit data of different dimensions of the lender are converted into the vectors corresponding to the dimensions according to the mining conversion logic specific to the dimensions, so that the different treatment of the data of different dimensions is realized, and the information quantity contained in the original data can be mined to the greatest extent. By adopting the vectors to carry out modular classification operation, the output credit classified data is more accurate and reliable, and the financial condition and borrowing capability of a lender can be effectively evaluated, so that the accurate release of loans is realized.

Drawings

The foregoing and other objects, features and advantages of the disclosure will be more apparent from the following description of embodiments of the disclosure with reference to the accompanying drawings, in which:

FIG. 1 schematically illustrates an application scenario diagram of a credit classification method according to an embodiment of the disclosure;

FIG. 2 schematically illustrates a flow chart of a credit classification method according to an embodiment of the disclosure;

FIG. 3 schematically illustrates a flow chart of a credit data acquisition method according to an embodiment of the disclosure;

FIG. 4 schematically illustrates a flow chart of a credit data acquisition and credit calculation method according to an embodiment of the disclosure;

FIG. 5 schematically illustrates a flow chart of a credit data acquisition and credit calculation method according to an embodiment of the disclosure;

FIG. 6 schematically illustrates a flow chart of a credit data acquisition and credit calculation method according to an embodiment of the disclosure;

FIG. 7 schematically illustrates a flow chart of a credit data acquisition and credit calculation method according to an embodiment of the disclosure;

FIG. 8 schematically illustrates a flow chart of a credit data acquisition and credit calculation method according to an embodiment of the disclosure;

FIG. 9 schematically illustrates a flow chart of a credit classification method according to an embodiment of the disclosure;

FIG. 10 schematically illustrates a block diagram of a credit classification device according to an embodiment of the disclosure; and

fig. 11 schematically illustrates a block diagram of an electronic device adapted to implement a credit classification method according to an embodiment of the disclosure.

Detailed Description

Hereinafter, embodiments of the present disclosure will be described with reference to the accompanying drawings. It should be understood that the description is only exemplary and is not intended to limit the scope of the present disclosure. In the following detailed description, for purposes of explanation, numerous specific details are set forth in order to provide a thorough understanding of the embodiments of the present disclosure. It may be evident, however, that one or more embodiments may be practiced without these specific details. In addition, in the following description, descriptions of well-known structures and techniques are omitted so as not to unnecessarily obscure the concepts of the present disclosure.

The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the disclosure. The terms "comprises," "comprising," and/or the like, as used herein, specify the presence of stated features, steps, operations, and/or components, but do not preclude the presence or addition of one or more other features, steps, operations, or components.

All terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art unless otherwise defined. It should be noted that the terms used herein should be construed to have meanings consistent with the context of the present specification and should not be construed in an idealized or overly formal manner.

Where expressions like at least one of "A, B and C, etc. are used, the expressions should generally be interpreted in accordance with the meaning as commonly understood by those skilled in the art (e.g.," a system having at least one of A, B and C "shall include, but not be limited to, a system having a alone, B alone, C alone, a and B together, a and C together, B and C together, and/or A, B, C together, etc.).

In the technical scheme of the disclosure, the related processes of collecting, storing, using, processing, transmitting, providing, disclosing, applying and the like of the personal information of the user all conform to the regulations of related laws and regulations, necessary security measures are adopted, and the public order harmony is not violated.

In the technical scheme of the disclosure, the authorization or consent of the user is obtained before the personal information of the user is obtained or acquired.

To solve the technical problems existing in the prior art, an embodiment of the present disclosure provides a credit classifying method, including: acquiring credit data of a lender, wherein the credit data comprises data sets with N dimensions, and N is greater than or equal to 1; converting the data sets of the N dimensions into N credit values according to preset data mining logic based on the data sets of the N dimensions, wherein the preset data mining logic comprises N preset data mining sub-logics, and the preset data mining sub-logics are in one-to-one correspondence with the dimensions of the data sets; and outputting credit classification labels of lenders through a preset machine learning model based on the N credit values.

Fig. 1 schematically illustrates an application scenario diagram of a credit classification method according to an embodiment of the disclosure.

As shown in fig. 1, an application scenario 100 according to this embodiment may include

terminal devices

101, 102, 103, a network 104, and a server 105. The network 104 is used as a medium to provide communication links between the

terminal devices

101, 102, 103 and the server 105. The network 104 may include various connection types, such as wired, wireless communication links, or fiber optic cables, among others.

The user may interact with the server 105 via the network 104 using the

terminal devices

101, 102, 103 to receive or send messages or the like. Various communication client applications, such as shopping class applications, web browser applications, search class applications, instant messaging tools, mailbox clients, social platform software, etc. (by way of example only) may be installed on the

terminal devices

101, 102, 103.

The

terminal devices

101, 102, 103 may be a variety of electronic devices having a display screen and supporting web browsing, including but not limited to smartphones, tablets, laptop and desktop computers, and the like.

The server 105 may be a server providing various services, such as a background management server (by way of example only) providing support for websites browsed by users using the

terminal devices

101, 102, 103. The background management server may analyze and process the received data such as the user request, and feed back the processing result (e.g., the web page, information, or data obtained or generated according to the user request) to the terminal device.

It should be noted that the credit classifying method provided by the embodiments of the present disclosure may be generally performed by the server 105. Accordingly, the credit classifying apparatus provided by the embodiments of the present disclosure may be generally provided in the server 105. The credit classification method provided by the embodiments of the present disclosure may also be performed by a server or a server cluster that is different from the server 105 and is capable of communicating with the

terminal devices

101, 102, 103 and/or the server 105. Accordingly, the credit classifying apparatus provided by the embodiments of the present disclosure may also be provided in a server or a server cluster different from the server 105 and capable of communicating with the

terminal devices

101, 102, 103 and/or the server 105.

It should be understood that the number of terminal devices, networks and servers in fig. 1 is merely illustrative. There may be any number of terminal devices, networks, and servers, as desired for implementation.

The credit classifying method of the disclosed embodiment will be described in detail below with reference to fig. 2 to 9 based on the scenario described in fig. 1.

Fig. 2 schematically illustrates a flow chart of a credit classification method according to an embodiment of the disclosure.

As shown in fig. 2, the credit classifying method of this embodiment includes operations S210 to S230, and the credit classifying method may be performed by the server 105.

In operation S210, credit data of the lender is acquired, the credit data including a data set of N dimensions, where N is 1 or more.

In embodiments of the present disclosure, the user's consent or authorization may be obtained prior to obtaining the user's information. For example, before operation S210, a request to acquire user information may be issued to the user. In case the user agrees or authorizes that the user information can be acquired, the operation S210 is performed.

Specifically, the lender includes a natural person and an enterprise, so the credit classification method in the embodiment of the disclosure may be performed by classifying the natural person as well as classifying the enterprise as the credit classification method in the embodiment of the disclosure. The acquired credit data is made up of data sets of different dimensions, and the processing logic for these credit data is also divided by dimension as described below. Similarly, when credit data is acquired, data in different dimensions are scattered in different data platforms, and data sets in the N dimensions are acquired in different data platforms, and due to the specificity of the data on the different data platforms, the acquired data needs to be uniformly processed so as to keep the data of the different platforms to realize real-time synchronous acquisition on a service level.

Fig. 3 schematically illustrates a flow chart of a credit data acquisition method according to an embodiment of the disclosure.

As shown in fig. 3, the credit data acquisition method of this embodiment includes operations S310 to S330, and the operations S310 to S330 may perform at least the above-described operation S210.

In operation S310, a loan request of the lender is received.

In operation S320, the loan request is parsed to obtain a loan subject and a repayment year.

In operation S330, the credit data of the lender is acquired based on the lender and the repayment year.

The repayment period is defined according to the duration of the loan of the lender, and for convenience of explanation, it is assumed that the defined repayment period includes a short-term loan and a long-term loan, and then the acquisition of credit data is required according to whether the lender is a natural person or an enterprise and according to the short-term loan or the long-term loan of the lender.

In operation S220, the N-dimensional data sets are converted into N credit values according to a preset data mining logic based on the N-dimensional data sets, where the preset data mining logic includes N preset data mining sub-logics, and the preset data mining sub-logics are in one-to-one correspondence with the dimensions of the data sets.

Specifically, the acquired data sets of each dimension are subjected to calculation processing according to preset mining logic, so that information which can be used for credit evaluation classification and is contained in the data in each data set can be converted into a modulus vector through the preset mining logic, wherein the data sets of each dimension correspond to the mining logic of the dimension.

It should be noted that, the collection period of the credit data may be a collection period of the credit data determined by the lending body alone or a collection period of the credit data determined by the repayment year alone, and the collection period of the credit data may also be a collection period of the credit data determined by the lending body and the repayment year together.

In operation S230, a credit classification tag of the lender is output through a preset machine learning model based on the N credit values.

Specifically, the preset machine learning model is a supervised machine learning model, and thus, training of the model is also based on the completion of training of the modulo vector and the pre-set labels, which is not described here too much.

Fig. 4 schematically illustrates a flow chart of a credit data acquisition and credit calculation method according to an embodiment of the disclosure.

As shown in fig. 4, the credit data acquisition and credit value calculation method of this embodiment includes operations S410 to S420.

According to an embodiment of the present disclosure, the N dimensions include at least: lattice dimension, capability dimension, asset dimension, vouching dimension and environmental condition dimension,

in operation S410, for the grid dimension, reference grid data and real-time grid data are acquired, wherein the reference grid data is acquired based on a first acquisition period, the reference grid data includes a first payment credit, a first payment amount, and a first payment number, the real-time grid data is acquired based on a second acquisition period, and the real-time grid data includes a second payment amount and a second payment number. The operation S410 may at least partially perform the above operation S310.

In operation S420, a lattice credit value is calculated based on the first payment credit, the first payment amount, and the first number of payments, in combination with the second payment amount and the second number of payments. The operation S420 may at least partially perform the above-described operation S220.

The price is a broad concept, which not only means that the lender has the capability of paying the debt, but also has the responsibility and willingness of bearing the debt, and the price is quantized to better reflect the default probability of the client, and the better the price is, the smaller the default probability is; the data may be provided by a plurality of data center platforms, with the data provider timing the acquisition of customer credit data, the data provider typically quantifies the lender's credit based on past lender records and expiration records. Assuming the collected original data, the client credit module additionally improves the quantitative value of the credit of the lender for timely repayment of larger amount and frequent repayment times in the past, and reflects the client magnitude.

Specifically, relevant client credit system data are called, the highest client credit value of the credit in the range of a first acquisition time period (for example, ten years) is selected as a reference value, and the current client credit value is calculated; the collection time period of the historical payment amount and the payment number related to the current client credit value is determined by an evaluation period (namely, a second collection time period), and if the credit is a short-term loan, the second collection time period can set the parameter to be one year, and if the credit is a long-term loan, the second collection time period can set the parameter to be five years. The baseline highest customer credit value is calculated once per month, i.e.: the standard highest client credit value is updated in real time on a fixed date per month, so that the standard data can be replaced in time.

For example, based on the most frequent lender credit (noted as the screen, the first payoff credit) and the real-time payoff amount (noted as the value, the second payoff amount) and the real-time payoff number (noted as the frequency, the second payoff number) of historical payoff amounts (noted as the svalue, the first payoff amount) in the past decade (i.e., the first collection period). The credit value representing its lattice is calculated for each customer in the following manner:

x _i (value_frequency) s screen/(svalue_frequency) formula (1)

Fig. 5 schematically illustrates a flow chart of a credit data acquisition and credit calculation method according to an embodiment of the disclosure.

As shown in fig. 5, the credit data acquisition and credit value calculation method of this embodiment includes operations S510 to S520.

In operation S510, for the capability dimension, pipeline data is acquired, the pipeline data being acquired based on the first acquisition time period. The operation S510 may at least partially perform the above operation S310.

In operation S520, a capability credit value is calculated based on the pipeline data. Wherein, this operation S520 may at least partially perform the above-described operation S220.

Lender capability embodies the ability to use funds in the future to bring cash flow and to repay loans. If the lender is personal, the assessment emphasizes consideration of future cash flows of the lender, and if the loan applicant is n years in this period of loan, the assessment considers cash flow predictions for at least n years in the future. The module predicts the future cash flow based on the past first collection time period (which can be set as n) of the bank card flowing water (ssalary) of the lender, and adjusts the predicted flowing water according to the current market interest rate and the currency expansion rate, wherein the calculation mode is as follows:

y _i ＝PV(sum(ssalary*(1+π) ⁿ ) Arbitrary (2)

Wherein, the liquid crystal display device comprises a liquid crystal display device,

wherein, n-phase income flows, each phase is separated by the same time, r is the interest rate, r is the month interest rate if one month is one phase, and r is the annual interest rate if one year is one phase.

If the borrowing and lending person is an enterprise, the assessment mainly calculates the free cash flow of the enterprise according to the cash flow meter in the financial statement of the enterprise, and comprehensively balances and calculates the free cash flow as a quantification index of the borrowing and lending person capability according to various incomes and extra incomes in the profit table.

Fig. 6 schematically illustrates a flow chart of a credit data acquisition and credit calculation method according to an embodiment of the disclosure.

As shown in fig. 6, the credit data acquisition and credit value calculation method of this embodiment includes operations S610 to S620.

In operation S610, long-term fixed asset data and long-term liability data are acquired based on the third acquisition period, and short-term liquidity asset data and short-term liability data are acquired based on the fourth acquisition period for the asset dimension. The operation S610 may at least partially perform the above-described operation S310.

In operation S620, asset credit values are calculated based on the long-term fixed asset data and long-term liability data in combination with the short-term liquidity asset data and short-term liability data. The operation S620 may at least partially perform the above-described operation S220.

For individual customers, the index focuses on the conditions of the lender's long-term fixed assets current market price A1, short-term liquidity asset A2, and family long-term liabilities B1, short-term liabilities B2, quantified by the difference or ratio.

z _j ＝(k ₁ *A ₁ +k ₂ *A ₂ )/(k ₃ *B ₁ +k ₄ *B ₂ ) (5)

Wherein k is ₁ 、k ₂ 、k ₃ K ₄ Respectively different adjustment coefficients.

For enterprises, according to the enterprise asset liability list, the asset and owner equity and long-short-period liability difference are quantified, different weights are given according to the duration of the asset (liability), and the asset (liability) with long duration is moderately reduced in weight due to larger risk of variation.

Fig. 7 schematically illustrates a flow chart of a credit data acquisition and credit calculation method according to an embodiment of the disclosure.

As shown in fig. 7, the credit data acquisition and credit value calculation method of this embodiment includes operations S710 to S720.

In operation S710, a vouching property price and a vouchers revenue cash flow are acquired for the vouching dimension, wherein the vouching revenue cash flow is collected based on the first collection period or the vouching revenue cash flow is collected based on the third collection period. The operation S710 may at least partially perform the above-described operation S310.

In operation S720, a vouching credit value is calculated based on the vouching property price and the revenue cash flow. The operation S720 may at least partially perform the above operation S220.

The guarantee provides a protection for the loan of the bank, and when the lender presents repayment crisis, the bank can dispense the guarantee or collect the return to the guarantee. The index may be quantified by the market price of the guaranty (price after discount) (denoted as sasset), and the net income cash flow of the guaranty (denoted as PV (income)), as follows:

α _j = (sasset) +pv (income) formula (6)

Fig. 8 schematically illustrates a flow chart of a credit data acquisition and credit calculation method according to an embodiment of the disclosure.

As shown in fig. 8, the credit data acquisition and credit value calculation method of this embodiment includes operations S810 to S820.

In operation S810, stock price data, stock price ratio data, industry index, and composite index are acquired for the environmental condition dimension. The operation S810 may at least partially perform the above-described operation S310.

In operation S820, an environmental condition credit value is calculated based on the stock price data, the stock price ratio data, the industry index, and the composite index. The operation S820 may at least partially perform the above-described operation S220.

The dimension of the environmental condition needs to consider the rights and interests of the lender and the risk condition of the industry; without taking into account systematic risks, stock prices (p ₁ ) Current stock price ratio (P) to industry of enterprise ₁ ) As a quantitative index, reflecting the operating environment condition of the lender; in view of industry lifecycle, an available industry index (i ₁ ) And complex index (I) ₁ ) The ratio represents the lender's business risk.

b _j ＝(p ₁ *i ₁ )/(P ₁ *I ₁ ) (7)

Wherein the closer the ratio is to 1, the less risk that represents the lender's market; the closer the ratio is to 0, the greater the market risk that represents the lender.

Fig. 9 schematically illustrates a flow chart of a credit classification method according to an embodiment of the disclosure.

As shown in fig. 9, the credit classifying method of this embodiment includes operations S910 to S920, and the operations S910 to S920 may at least partially perform the above-described operation S230.

In operation S910, an N-dimensional vector corresponding to the lender is generated based on the N modulo vectors.

In operation S920, a credit classification tag of the lender is output through a preset KNN model based on the N-dimensional vector.

For example, the five-dimensional vector a for each lender may be generated by the acquisition processing of the above-described five-dimensional data _i ＝(x _j ，y _j ，z _j ，a _j ，b _j ). The five-dimensional vector is used as modeling data, and credit classification labels of lenders are output through a pre-trained KNN model. In particular, the method is consistent with the classification of loan quality by banks. Banks divide the quality of loans into five categories, namely normal category, attention, secondary, suspicious, and loss, according to the risk level of the loans, wherein the latter three are bad loans. The classification label is a final standard for determining loan quality by business personnel, and can more truly reflect bad loan conditions.

The K-nearest neighbor algorithm (KNN) is a common supervised learning algorithm, which is assumed to be given to a training data set, and the class of the examples is determined, and when classifying data, class prediction is performed on new data examples according to the class of training examples of K nearest neighbors. Because the KNN model is fast in training time and insensitive to abnormal values, a good prediction effect can be achieved, the credit condition of a lender is analyzed and predicted from five dimensions by adopting a KNN algorithm, the purpose of predicting the real-time credit condition of the lender is achieved, the credit evaluation flow is simplified, the loan issuing speed is accelerated, and business personnel are helped to evaluate the risk condition of loan issuing in real time.

The test and training method for the KNN model is as follows:

collecting historical data of past lenders, preparing five-dimensional characteristic data and final classification label data of the historical lenders, and mixing sample data according to 8: and splitting the training set and the verification set according to the proportion 2. The choice of k value determines the model predictive effect to a great extent. If the k value is smaller, the KNN model is more complex, and the risk of over fitting is generated; if the k value is large, the model is simpler but there is a risk of under fitting. Because the final classification label number of the scheme is 5 (i.e. normal class, concern, secondary class, suspicious class and loss), and the classification decision rule considering KNN is a majority vote, namely the label classification with highest occurrence frequency in k samples around the unknown sample is determined as the class of the unknown sample, the k value is at least required to be greater than 6. Starting from k value of 6, the k value is continuously increased, variance or misclassification rate of data prediction in the verification set is calculated, the variance is small, and the label prediction accuracy is high, so that the k value is selected as the k value on which the final decision is based when the variance is minimum. And respectively calculating the space distance between the unknown sample data and a data set formed by each known sample, selecting k samples with the minimum Euler distance, and selecting a classification label with the highest occurrence frequency as a final prediction result of the unknown sample.

Based on the credit classifying method, the disclosure also provides a credit classifying device. The device will be described in detail below in connection with fig. 10.

Fig. 10 schematically shows a block diagram of a credit classifying apparatus according to an embodiment of the disclosure.

As shown in fig. 10, the credit classifying apparatus 1000 of this embodiment includes a credit data acquisition module 1010, a data mining processing module 1020, and a credit classification tag output module 1030.

The credit data collection module 1010 is configured to obtain credit data of a lender, where the credit data includes a data set with N dimensions, where N is greater than or equal to 1. In an embodiment, the credit data collection module 1010 may be used to perform the operation S210 described above, which is not described herein.

The data mining processing module 1020 is configured to convert the data sets of the N dimensions into N credit values according to a preset data mining logic based on the data sets of the N dimensions, where the preset data mining logic includes N preset data mining sub-logics, and the preset data mining sub-logics are in one-to-one correspondence with the dimensions of the data sets. In an embodiment, the data mining processing module 1020 may be configured to perform the operation S220 described above, which is not described herein.

The credit category label output module 1030 is configured to output a credit category label of the lender through a preset machine learning model based on the N credit values. In an embodiment, the credit classification tag output module 1030 may be used to perform the operation S230 described above, which is not described herein.

Any of the credit data collection module 1010, the data mining processing module 1020, and the credit category label output module 1030 may be combined in one module or any of the modules may be split into multiple modules according to embodiments of the present disclosure. Alternatively, at least some of the functionality of one or more of the modules may be combined with at least some of the functionality of other modules and implemented in one module. At least one of the credit data acquisition module 1010, the data mining processing module 1020, and the credit category label output module 1030 may be implemented at least in part as hardware circuitry, such as a Field Programmable Gate Array (FPGA), a Programmable Logic Array (PLA), a system on a chip, a system on a substrate, a system on a package, an Application Specific Integrated Circuit (ASIC), or may be implemented in hardware or firmware in any other reasonable manner of integrating or packaging circuitry, or in any one of or a suitable combination of three of software, hardware, and firmware. Alternatively, at least one of the credit data collection module 1010, the data mining processing module 1020, and the credit category label output module 1030 may be implemented at least in part as a computer program module that, when executed, performs the corresponding functions.

As shown in fig. 11, an electronic device 1100 according to an embodiment of the present disclosure includes a processor 1101 that can perform various appropriate actions and processes according to a program stored in a Read Only Memory (ROM) 1102 or a program loaded from a storage section 1108 into a Random Access Memory (RAM) 1103. The processor 1101 may include, for example, a general purpose microprocessor (e.g., a CPU), an instruction set processor and/or an associated chipset and/or a special purpose microprocessor (e.g., an Application Specific Integrated Circuit (ASIC)), or the like. The processor 1101 may also include on-board memory for caching purposes. The processor 1101 may comprise a single processing unit or a plurality of processing units for performing the different actions of the method flow according to embodiments of the present disclosure.

In the RAM 1103, various programs and data necessary for the operation of the electronic device 1100 are stored. The processor 1101, ROM 1102, and RAM 1103 are connected to each other by a bus 1104. The processor 1101 performs various operations of the method flow according to the embodiments of the present disclosure by executing programs in the ROM 1102 and/or the RAM 1103. Note that the program may be stored in one or more memories other than the ROM 1102 and the RAM 1103. The processor 1101 may also perform various operations of the method flow according to embodiments of the present disclosure by executing programs stored in the one or more memories.

According to an embodiment of the disclosure, the electronic device 1100 may also include an input/output (I/O) interface 1105, the input/output (I/O) interface 1105 also being connected to the bus 1104. The electronic device 1100 may also include one or more of the following components connected to the I/O interface 1105: an input section 1106 including a keyboard, a mouse, and the like; an output portion 1107 including a Cathode Ray Tube (CRT), a Liquid Crystal Display (LCD), and the like, a speaker, and the like; a storage section 1108 including a hard disk or the like; and a communication section 1109 including a network interface card such as a LAN card, a modem, and the like. The communication section 1109 performs communication processing via a network such as the internet. The drive 1110 is also connected to the I/O interface 1105 as needed. Removable media 1111, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, or the like, is installed as needed in drive 1110, so that a computer program read therefrom is installed as needed in storage section 1108.

The present disclosure also provides a computer-readable storage medium that may be embodied in the apparatus/device/system described in the above embodiments; or may exist alone without being assembled into the apparatus/device/system. The computer-readable storage medium carries one or more programs which, when executed, implement methods in accordance with embodiments of the present disclosure.

According to embodiments of the present disclosure, the computer-readable storage medium may be a non-volatile computer-readable storage medium, which may include, for example, but is not limited to: a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the context of this disclosure, a computer-readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. For example, according to embodiments of the present disclosure, the computer-readable storage medium may include ROM1102 and/or RAM 1103 described above and/or one or more memories other than ROM1102 and RAM 1103.

Embodiments of the present disclosure also include a computer program product comprising a computer program containing program code for performing the methods shown in the flowcharts. The program code, when executed in a computer system, causes the computer system to perform the methods provided by embodiments of the present disclosure.

The above-described functions defined in the system/apparatus of the embodiments of the present disclosure are performed when the computer program is executed by the processor 1101. The systems, apparatus, modules, units, etc. described above may be implemented by computer program modules according to embodiments of the disclosure.

In one embodiment, the computer program may be based on a tangible storage medium such as an optical storage device, a magnetic storage device, or the like. In another embodiment, the computer program can also be transmitted, distributed over a network medium in the form of signals, downloaded and installed via the communication portion 1109, and/or installed from the removable media 1111. The computer program may include program code that may be transmitted using any appropriate network medium, including but not limited to: wireless, wired, etc., or any suitable combination of the foregoing.

In such an embodiment, the computer program can be downloaded and installed from a network via the communication portion 1109, and/or installed from the removable media 1111. The above-described functions defined in the system of the embodiments of the present disclosure are performed when the computer program is executed by the processor 1101. The systems, devices, apparatus, modules, units, etc. described above may be implemented by computer program modules according to embodiments of the disclosure.

According to embodiments of the present disclosure, program code for performing computer programs provided by embodiments of the present disclosure may be written in any combination of one or more programming languages, and in particular, such computer programs may be implemented in high-level procedural and/or object-oriented programming languages, and/or assembly/machine languages. Programming languages include, but are not limited to, such as Java, c++, python, "C" or similar programming languages. The program code may execute entirely on the user's computing device, partly on the user's device, partly on a remote computing device, or entirely on the remote computing device or server. In the case of remote computing devices, the remote computing device may be connected to the user computing device through any kind of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or may be connected to an external computing device (e.g., connected via the Internet using an Internet service provider).

The flowcharts and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams or flowchart illustration, and combinations of blocks in the block diagrams or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.

Those skilled in the art will appreciate that the features recited in the various embodiments of the disclosure and/or in the claims may be provided in a variety of combinations and/or combinations, even if such combinations or combinations are not explicitly recited in the disclosure. In particular, the features recited in the various embodiments of the present disclosure and/or the claims may be variously combined and/or combined without departing from the spirit and teachings of the present disclosure. All such combinations and/or combinations fall within the scope of the present disclosure.

The embodiments of the present disclosure are described above. However, these examples are for illustrative purposes only and are not intended to limit the scope of the present disclosure. Although the embodiments are described above separately, this does not mean that the measures in the embodiments cannot be used advantageously in combination. The scope of the disclosure is defined by the appended claims and equivalents thereof. Various alternatives and modifications can be made by those skilled in the art without departing from the scope of the disclosure, and such alternatives and modifications are intended to fall within the scope of the disclosure.

Claims

1. A credit classification method, comprising:

acquiring credit data of a lender, wherein the credit data comprises data sets with N dimensions, and N is greater than or equal to 1;

Converting the data sets of the N dimensions into N credit values according to preset data mining logic based on the data sets of the N dimensions, wherein the preset data mining logic comprises N preset data mining sub-logics, and the preset data mining sub-logics are in one-to-one correspondence with the dimensions of the data sets; and

and outputting credit classification labels of lenders through a preset machine learning model based on the N credit values.

2. The method of claim 1, wherein the acquiring credit data for the lender comprises:

receiving a loan request from the lender;

analyzing the loan request to obtain a loan main body and a repayment year; and

and acquiring the credit data of the lender based on the lender body and the repayment year.

3. The method of claim 2, wherein the N dimensions comprise at least: lattice dimension, capability dimension, asset dimension, vouching dimension and environmental condition dimension,

the obtaining the credit data of the lender based on the lender and the repayment age includes:

for the grid dimension, acquiring reference grid data and real-time grid data, wherein the reference grid data is acquired based on a first acquisition time period, the reference grid data comprises first repayment credit, first repayment amount and first repayment times, the real-time grid data is acquired based on a second acquisition time period, and the real-time grid data comprises second repayment amount and second repayment times;

The converting the data set of the N dimensions into N credit values according to preset data mining logic, including:

and calculating a lattice credit value based on the first repayment credit, the first repayment amount and the first repayment times and combining the second repayment amount and the second repayment times.

4. The method of claim 3, wherein,

for the capability dimension, obtaining streaming data, the streaming data being acquired based on the first acquisition time period;

based on the pipeline data, a capacity credit value is calculated.

5. The method according to claim 3 or 4, wherein,

for the asset dimension, collecting long-term fixed asset data and long-term liability data based on a third collection time period, and collecting short-term liquidity asset data and short-term liability data based on a fourth collection time period;

and calculating an asset credit value based on the long-term fixed asset data and the long-term liability data in combination with the short-term liquidity asset data and the short-term liability data.

6. The method of claim 5, wherein,

acquiring, for the guarantee dimension, a guarantee property price and a guarantee person revenue cash flow, wherein the guarantee person revenue cash flow is collected based on the first collection period or the guarantee person revenue cash flow is collected based on the third collection period;

a vouching credit value is calculated based on the vouching property price and the revenue cash flow.

7. The method of claim 6, wherein,

For the dimension of the environmental condition, acquiring stock price data, stock price ratio data, industry indexes and comprehensive indexes;

and calculating an environmental condition credit value based on the stock price data, the stock price ratio data, the industry index and the comprehensive index.

8. The method of claim 1, wherein the outputting the credit classification tag of the lender through a preset machine learning model based on the N credit values comprises:

generating an N-dimensional vector corresponding to the lender based on the N modulo vectors; and

and outputting credit classification labels of lenders through a preset KNN model based on the N-dimensional vector.

9. A credit classifying apparatus, comprising:

the credit data acquisition module is used for acquiring credit data of lenders, wherein the credit data comprises data sets with N dimensions, and N is greater than or equal to 1;

the data mining processing module is used for converting the data sets of the N dimensions into N credit values according to preset data mining logic based on the data sets of the N dimensions, wherein the preset data mining logic comprises N preset data mining sub-logics, and the preset data mining sub-logics are in one-to-one correspondence with the dimensions of the data sets; and

And the credit classification label output module is used for outputting credit classification labels of lenders through a preset machine learning model based on the N credit values.

10. An electronic device, comprising:

one or more processors;

storage means for storing one or more programs,

wherein the one or more programs, when executed by the one or more processors, cause the one or more processors to perform the method of any of claims 1-8.

11. A computer readable storage medium having stored thereon executable instructions which, when executed by a processor, cause the processor to perform the method according to any of claims 1-8.

12. A computer program product comprising a computer program which, when executed by a processor, implements the method according to any one of claims 1 to 8.