Digital financial user credit assessment method and system based on big data

Through big data technology, users' historical credit scores are obtained and analyzed, average credit scores and score correction coefficients are calculated, credit ratings and maintenance cycles are determined, which solves the problem of insufficient convenience caused by the frequent credit assessment in digital finance, and improves loan approval Efficiency and accuracy of credit assessment.

CN118967301APending Publication Date: 2024-11-15LONGYAN UNIV
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202411037638.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-31
Publication Date
2024-11-15

AI Technical Summary

Technical Problem

In digital finance, frequent user credit assessment by financial institutions leads to insufficient convenience, especially for users whose credit rating meets the requirements but needs to urgently need loans. The existing technology cannot quickly and accurately evaluate, which affects the efficiency of loan approval.

Method used

Through a big data-based method, the user's historical credit score records are obtained, the average credit score and score correction coefficient are calculated, the credit rating and its range are determined, the target image is generated, the stable ratio and score difference are calculated, the maintenance period is corrected, and the credit rating is realized. Stability assessment and maintenance simplify the credit assessment process.

Benefits of technology

It improves the efficiency and convenience of loan approval, reduces frequent fluctuations in credit ratings, reduces the complexity of risk assessment of financial institutions, improves decision-making efficiency and accuracy, and ensures the accuracy and stability of credit assessment.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118967301A_ABST
    Figure CN118967301A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of big data, and particularly discloses a digital financial user credit assessment method and system based on big data, and the method comprises the following steps: S1, obtaining the historical credit score record of a user, and calculating the average credit score of the user; s2, determining the credit rating of the average credit score, and determining the credit score range of the credit rating; determining a target image, and determining a stable proportion; calculating a score difference value, and generating a difference value set; and S3, calculating a stability coefficient, and correcting a preset retention period according to the stability coefficient. According to the credit score fluctuation condition of the user, the time interval at which the credit rating of the user is kept unchanged can be determined, and the loan approval efficiency and convenience are improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of big data technology, and in particular to a digital financial user credit assessment method and system based on big data. Background Art

[0002] Digital finance refers to financial services and products provided using digital technology and innovative means. It covers the digital transformation of traditional financial services as well as emerging financial technology solutions, with the aim of improving the efficiency, convenience and inclusiveness of financial services.

[0003] In digital finance, the use of artificial intelligence and big data for credit assessment is a very important aspect. By evaluating the user's credit status, financial institutions can more accurately judge the borrower's repayment ability and default risk, thereby reducing the risk of loans and other financial products. At the same time, through accurate credit assessment, financial institutions can provide customized financial products and services to users with different credit status, such as differentiated loan interest rates, credit limits, etc., to enhance user experience.

[0004] When a user has a loan demand, the lending institution will be authorized to query the user's own credit report, and the lending institution will conduct a credit assessment on the user, and lend the user money after the user's credit rating meets the requirements. During the loan business, the user may have another loan demand, that is, to conduct a loan business again. Some financial institutions may not allow two outstanding loans to exist at the same time. Some financial institutions may choose to conduct a credit assessment on the user again and determine whether the user's credit rating meets the borrowing requirements. However, it usually takes a certain amount of time for financial institutions to conduct a credit assessment on the user again, which is not convenient for users who meet the credit rating requirements and have an urgent need for loans. Summary of the invention

[0005] The purpose of the present invention is to provide a digital financial user credit assessment method and system based on big data to solve the above technical problems.

[0006] The purpose of the present invention can be achieved through the following technical solutions:

[0007] The digital financial user credit assessment method based on big data includes the following steps:

[0008] S1: Obtain the user's historical credit score records within the preset collection period T and calculate the user's average credit score αi represents P i The score correction factor, P i represents the credit score after the i-th change in the historical credit score record, and m represents the total number of times the user's credit score has changed;

[0009] S2: setting a number of credit levels from low to high with a preset credit score interval, determining the credit level D where the average credit score P' is located, and determining the credit score range [PDsta, PDend] of the credit level D;

[0010] Generate data points (t i , P i ), t i Indicates the time when the user's credit score changes for the i-th time, and connects adjacent data points through straight lines to obtain the target image, and determines the duration T when the credit score on the target image is greater than or equal to PDsta fix , and calculate the stable ratio C = T fix / T;

[0011] Calculate the score difference ΔP i =|P i+1 -P i |, and generate a difference set ΔPJH = (ΔP1, ΔP2, ..., ΔP m-1 );

[0012] S3: Calculate the stability coefficient, and modify the preset retention period according to the stability coefficient, wherein the retention period is the time interval for keeping the user's credit rating unchanged.

[0013] As a further solution of the present invention: in step S1, the process of calculating the user's credit score specifically includes:

[0014] Obtain the user's loan data, including the total loan amount to be repaid Q due and the total outstanding loan amount Q qua , and determine whether the user has a loan overdue;

[0015] When a user has a loan overdue, the user's credit score is calculated through the following steps:

[0016] Get overdue data, including the total overdue amount Q due ' and the total number of overdue loans Q qua ';

[0017] The user's credit score P is calculated using the following formula:

[0018]

[0019] Where η1 represents the preset first correction coefficient, Q cre Indicates the remaining loan amount of the user;

[0020] When the user has no overdue loans, the user's credit score is calculated through the following steps:

[0021]

[0022] Wherein, η2 represents a preset second correction coefficient, and η2<η1.

[0023] As a further solution of the present invention: in the process of obtaining the user's credit score for the ath time, when the total amount of loans to be repaid in the user's loan data is 0, the time when the total amount of loans to be repaid in the user's loan data is not 0 is recorded as non-repayment time, and the average of the credit scores of the user when they are in the non-repayment time is calculated, and used as the credit score of the user obtained for the ath time.

[0024] As a further solution of the present invention: the process of obtaining the user's credit score also includes the following steps:

[0025] Calculate the overdue number A'=A*Qqua'*Qdue', where A is the overdue value. When the user has a loan overdue, the overdue value A=1; when the user has no loan overdue, the overdue value A=0;

[0026] The ratio B of the overdue number A'≥Ayc is counted, Ayc represents the preset overdue number threshold, when the ratio B≥B' of the overdue number A'≥Ayc is reached, the user's loan request is automatically rejected within the preset time period Tyc, and B' represents the preset ratio threshold.

[0027] As a further solution of the present invention: in the step S1, the process of obtaining the score correction coefficient specifically includes:

[0028] The collection period is divided into M coefficient periods of the same length, and the ranking coefficient αM'=δ*M' / M of the M'th coefficient period is calculated, where δ represents the preset ranking correction coefficient, and when the time when the credit score changes is within the M'th coefficient period, the ranking coefficient αM' is used as the corresponding score correction coefficient.

[0029] As a further solution of the present invention: in the step S3, the process of correcting the preset holding period according to the stability coefficient specifically includes:

[0030] Calculate the stability coefficient K = γ*C / (max(ΔPJH)*S 2 ), γ represents the preset fluctuation correction coefficient, S 2 represents the variance of the user's credit score;

[0031] Calculate the correction period Treg=K*Tori, where Tori represents the maintenance period. Before the time interval from the last calculation of the user's credit score reaches the length of the correction period Treg, the user's credit level is maintained at D.

[0032] As a further solution of the present invention: when the correction period Treg>Treg'=3Tori, let Treg=Treg'.

[0033] Digital financial user credit assessment system based on big data, including:

[0034] Credit scoring module: obtains the user's historical credit score records within the preset collection period T and calculates the user's average credit score αi represents P i The score correction factor, P i represents the credit score after the i-th change in the historical credit score record, and m represents the total number of times the user's credit score has changed;

[0035] Data optimization module: setting a number of credit levels from low to high with a preset credit score interval, determining the credit level D where the average credit score P' is located, and determining the credit score range [PDsta, PDend] of the credit level D;

[0036] Generate data points (t i , P i ), t i represents the time corresponding to the i-th change in the user's credit score, and connects adjacent data points through straight lines to obtain the target image, and determines the duration T when the credit score on the target image is greater than or equal to PDsta fix , and calculate the stable ratio C = T fix / T;

[0037] Calculate the score difference ΔP i =|P i+1 -P i |, and generate a difference set ΔPJH = (ΔP1, ΔP2, ..., ΔP m-1 );

[0038] Cycle correction module: calculates the stability coefficient and corrects the preset retention period according to the stability coefficient. The retention period is the time interval for keeping the user's credit rating unchanged.

[0039] Beneficial effects of the present invention: In the present invention, first, the historical credit score records of the user within the collection period are obtained, and the average credit score is calculated; it can be understood that the length of the collection period is the validity period of the credit report authorized by the user to the lending institution, and the score correction coefficient is to make the calculated average credit score closer to the level at the current time; then, the credit level D of the average credit score is determined, and the credit score range of the credit level D is determined, and the stability ratio is determined according to the target image and the credit score range of the credit level D; it is worth noting that the stability ratio is the ratio of the time that the user's credit score is at and / or higher than the credit level D in the collection period. The larger the stability ratio, the longer the time that the user's credit score is at and / or higher than the credit level D, that is, the user's credit level is more stably at and / or higher than the credit level D; then the score difference is calculated, and a difference set is generated; it should be noted that the larger the score difference, the more unstable the user's credit score is, that is, the user's credit level needs to be adjusted more frequently to avoid errors in the user's credit level, which affects the user's normal loan application and / or the loan risk of the financial institution; finally, the stability coefficient is calculated, and the stability coefficient is corrected according to the stability coefficient. Preset retention period; It is worth noting that the larger the stability coefficient is, the more stable the user's credit rating is, the longer the revised retention period is, and the user's credit rating remains unchanged during the revised retention period; The reasons for using credit rating D as the user's credit rating during the revised retention period include: 1. The average credit score can smooth out short-term fluctuations and reflect the user's credit status over a longer period. Therefore, using credit rating D (based on average credit rating) can more accurately represent the user's actual credit level, rather than a score affected by short-term abnormal fluctuations; 2. If each credit score may lead to a change in credit rating, the user's credit rating will fluctuate frequently, resulting in reduced stability of financial services such as loan approval. By using the average credit score to determine credit rating D and maintaining this level during the revised retention period, the frequent fluctuations in the score can be reduced, and the stability and continuity of financial services can be improved; 3. For financial institutions, frequent changes in credit ratings will increase the complexity of risk assessment and decision-making. By maintaining a stable credit rating D within a period, the credit approval and risk management process can be simplified, and the decision-making efficiency and accuracy can be improved. The present invention can determine the time interval for the user's credit rating to remain unchanged according to the fluctuation of the user's credit score, thereby improving the efficiency and convenience of loan approval. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] The present invention will be further described below in conjunction with the accompanying drawings.

[0041] Figure 1 It is a flow chart of the digital financial user credit assessment method based on big data of the present invention. DETAILED DESCRIPTION

[0042] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0043] See also Figure 1 As shown, the present invention is a digital financial user credit assessment method based on big data, comprising the following steps:

[0044] S1: Obtain the user's historical credit score records within the preset collection period T and calculate the user's average credit score αi represents P i The score correction factor, P i represents the credit score after the i-th change in the historical credit score record, and m represents the total number of times the user's credit score has changed;

[0045] S2: setting a number of credit levels from low to high with a preset credit score interval, determining the credit level D where the average credit score P' is located, and determining the credit score range [PDsta, PDend] of the credit level D;

[0046] Generate data points (t i , P i ), t i Indicates the time when the user's credit score changes for the i-th time, and connects adjacent data points through straight lines to obtain the target image, and determines the duration T when the credit score on the target image is greater than or equal to PDsta fix , and calculate the stable ratio C = T fix / T;

[0047] Calculate the score difference ΔP i =|P i+1 -P i |, and generate a difference set ΔPJH = (ΔP1, ΔP2, ..., ΔP m-1 );

[0048] S3: Calculate the stability coefficient, and modify the preset retention period according to the stability coefficient, wherein the retention period is the time interval for keeping the user's credit rating unchanged.

[0049] It should be noted that first, the historical credit score records of the user within the collection period are obtained, and the average credit score is calculated; it can be understood that the length of the collection period is the validity period of the credit report authorized by the user to the lending institution, and the score correction coefficient is to make the calculated average credit score closer to the current level; then, the credit level D of the average credit score is determined, and the credit score range of the credit level D is determined, and the stability ratio is determined according to the target image and the credit score range of the credit level D; it is worth noting that the stability ratio is the ratio of the time that the user's credit score is at and / or higher than the credit level D in the collection period. The larger the stability ratio, the longer the user's credit score is at and / or higher than the credit level D, that is, the user's credit level is more stably at and / or higher than the credit level D; then the score difference is calculated, and a difference set is generated; it should be noted that the larger the score difference, the more unstable the user's credit score is, that is, the user's credit level needs to be adjusted more frequently to avoid errors in the user's credit level, which affects the user's normal loan application and / or the loan risk of the financial institution; finally, the stability coefficient is calculated, and the preset stability coefficient is corrected according to the stability coefficient. holding period; it is worth noting that the larger the stability coefficient is, the more stable the user's credit rating is, the longer the revised holding period is, and the user's credit rating remains unchanged during the revised holding period; the reasons for using credit rating D as the user's credit rating during the revised holding period include: 1. The average credit score can smooth out short-term fluctuations and reflect the user's credit status over a longer period. Therefore, using credit rating D (based on the average credit score) can more accurately represent the user's actual credit level rather than a score affected by short-term abnormal fluctuations; 2. If each credit score may lead to a change in credit rating, the user's credit rating will fluctuate frequently, resulting in reduced stability in financial services such as loan approval. By using the average credit score to determine credit rating D and maintaining this level during the revised holding period, the frequent fluctuations in the score can be reduced, and the stability and continuity of financial services can be improved; 3. For financial institutions, frequent changes in credit ratings will increase the complexity of risk assessment and decision-making. By maintaining a stable credit rating D over a period, the credit approval and risk management processes can be simplified, and the decision-making efficiency and accuracy can be improved.

[0050] In another preferred implementation of the present invention, in step S1, the process of calculating the user's credit score specifically includes:

[0051] Obtain the user's loan data, including the total loan amount to be repaid Q due and the total outstanding loan amount Q qua , and determine whether the user has a loan overdue;

[0052] When a user has a loan overdue, the user's credit score is calculated through the following steps:

[0053] Get overdue data, including the total overdue amount Q due ' and the total number of overdue loans Q qua ';

[0054] The user's credit score P is calculated using the following formula:

[0055]

[0056] Where η1 represents the preset first correction coefficient, Q cre Indicates the remaining loan amount of the user;

[0057] When the user has no overdue loans, the user's credit score is calculated through the following steps:

[0058]

[0059] Wherein, η2 represents a preset second correction coefficient, and η2<η1.

[0060] It is worth noting that, when the total number of outstanding loans remains unchanged, the larger the total outstanding loan amount, the more lending institutions are willing to lend to the user, and therefore it can be reflected that the user's credit score is higher; the larger the remaining loanable amount, the higher the user's credit score. If the loanable amount is smaller, regardless of whether the total loanable amount is small or the user has already borrowed a larger amount, the user should not be given a higher credit score, otherwise the user's credit rating may be increased again, and then the user's loan amount may be increased, thereby causing loan risks (such as bad debts) for financial institutions.

[0061] In another preferred implementation of the present invention, during the process of obtaining the user's credit score for the ath time, when the total amount of loans to be repaid in the user's loan data is 0, the time when the total amount of loans to be repaid in the user's loan data is not 0 is recorded as non-repayment time, and the average of the credit scores of the user when they are in the non-repayment time is calculated and used as the credit score of the user obtained for the ath time.

[0062] It is understandable that when the user's outstanding loan amount is 0, this may be a special case, for example, the user has just paid off all loans. In this case, directly calculating the credit score may lead to abnormal fluctuations, because the sudden zero loan amount does not represent the user's overall credit situation; by using the historical average within the collection period as the current credit score, a more stable and consistent credit score can be provided to avoid drastic fluctuations in the score caused by special circumstances. At the same time, the use of the historical average can better reflect the user's long-term credit status rather than the short-term fluctuations of a single calculation, which helps financial institutions to more accurately assess the user's credit risk.

[0063] In another preferred implementation of the present invention, the process of obtaining the user's credit score further includes the following steps:

[0064] Calculate the overdue number A'=A*Qqua'*Qdue', where A is the overdue value. When the user has a loan overdue, the overdue value A=1; when the user has no loan overdue, the overdue value A=0;

[0065] The ratio B of the overdue number A'≥Ayc is counted, Ayc represents the preset overdue number threshold, when the ratio B≥B' of the overdue number A'≥Ayc is reached, the user's loan request is automatically rejected within the preset time period Tyc, and B' represents the preset ratio threshold.

[0066] It should be noted that by calculating the number of overdue payments, the user's overdue situation can be evaluated more comprehensively, reflecting the user's overall overdue risk in the loan, and identifying the frequency and severity of the user's overdue payments throughout the credit cycle; automatic rejection can prevent high-risk users from applying for loans again in a short period of time, effectively improving the risk control capabilities of financial institutions and avoiding potential bad debt losses. At the same time, the automatic rejection mechanism can reduce the workload of manual review and improve the efficiency of loan approval. When it is detected that the user's overdue ratio exceeds the standard, the loan application can be automatically rejected to save human resources.

[0067] In another preferred implementation of the present invention, in step S1, the process of obtaining the score correction coefficient specifically includes:

[0068] The collection period is divided into M coefficient periods of the same length, and the ranking coefficient αM'=δ*M' / M of the M'th coefficient period is calculated, where δ represents the preset ranking correction coefficient, and when the time when the credit score changes is within the M'th coefficient period, the ranking coefficient αM' is used as the corresponding score correction coefficient.

[0069] In another preferred implementation of the present invention, in step S3, the process of correcting the preset holding period according to the stability coefficient specifically includes:

[0070] Calculate the stability coefficient K = γ*C / (max(ΔPJH)*S 2 ), γ represents the preset fluctuation correction coefficient, S 2 represents the variance of the user's credit score;

[0071] Calculate the correction period Treg=K*Tori, where Tori represents the maintenance period. Before the time interval from the last calculation of the user's credit score reaches the length of the correction period Treg, the user's credit level is maintained at D.

[0072] It should be noted that S 2 The variance is calculated based on the credit score changes during the collection period. The larger the variance, the more unstable the user's credit score is, and the correction period needs to be shortened, that is, the stability coefficient needs to be lowered. The stability coefficient calculation formula is obtained based on the same reason; the more stable the user's historical credit score is, the less frequent the user's credit score needs to be adjusted, that is, the longer the correction period is.

[0073] In another preferred implementation of the present invention, when the correction period Treg>Treg'=3Tori, Treg=Treg'.

[0074] Digital financial user credit assessment system based on big data, including:

[0075] Credit scoring module: obtains the user's historical credit score records within the preset collection period T and calculates the user's average credit score αi represents P i The score correction factor, P i represents the credit score after the i-th change in the historical credit score record, and m represents the total number of times the user's credit score has changed;

[0076] Data optimization module: setting a number of credit levels from low to high with a preset credit score interval, determining the credit level D where the average credit score P' is located, and determining the credit score range [PDsta, PDend] of the credit level D;

[0077] Generate data points (t i , P i ), t i Indicates the time when the user's credit score changes for the i-th time, and connects adjacent data points through straight lines to obtain the target image, and determines the duration T when the credit score on the target image is greater than or equal to PDsta fix , and calculate the stable ratio C = T fix / T;

[0078] Calculate the score difference ΔP i =|Pi+1 -P i |, and generate a difference set ΔPJH = (ΔP1, ΔP2, ..., ΔP m-1 );

[0079] Cycle correction module: calculates the stability coefficient and corrects the preset retention period according to the stability coefficient. The retention period is the time interval for keeping the user's credit rating unchanged.

[0080] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A digital financial user credit assessment method based on big data, characterized in that: The following steps are involved: S1: Obtain the user's historical credit score records within the preset collection period T and calculate the user's average credit score αi represents P i The score correction factor, P i represents the credit score after the i-th change in the historical credit score record, and m represents the total number of times the user's credit score has changed; S2: setting a number of credit levels from low to high with a preset credit score interval, determining the credit level D where the average credit score P' is located, and determining the credit score range [PDsta, PDend] of the credit level D; Generate data points (t i , P i ), t i Indicates the time when the user's credit score changes for the i-th time, and connects adjacent data points through straight lines to obtain the target image, and determines the duration T when the credit score on the target image is greater than or equal to PDsta fix , and calculate the stable ratio C = T fix / T; Calculate the score difference ΔP i =|P i+1 -P i |, and generate a difference set ΔPJH = (ΔP1, ΔP2, ..., ΔP m-1 ); S3: Calculate the stability coefficient, and modify the preset retention period according to the stability coefficient, wherein the retention period is the time interval for keeping the user's credit rating unchanged.

2. The digital financial user credit assessment method based on big data according to claim 1 is characterized in that: In step S1, the process of obtaining the user's credit score specifically includes: Obtain the user's loan data, including the total loan amount to be repaid Q due and the total outstanding loan amount Q qua , and determine whether the user has a loan overdue; When a user has a loan overdue, the user's credit score is calculated through the following steps: Get overdue data, including the total overdue amount Q due ' and the total number of overdue loans Q qua '; The user's credit score P is calculated using the following formula: Where η1 represents the preset first correction coefficient, Q cre Indicates the remaining loan amount of the user; When the user has no overdue loans, the user's credit score is calculated through the following steps: Wherein, η2 represents a preset second correction coefficient, and η2<η1.

3. The digital financial user credit assessment method based on big data according to claim 2 is characterized in that: In the process of obtaining the user's credit score for the ath time, when the total amount of loans to be repaid in the user's loan data is 0, the time when the total amount of loans to be repaid in the user's loan data is not 0 is recorded as non-repayment time, and the average credit score of the user during the non-repayment time is calculated and used as the credit score of the user obtained for the ath time.

4. The digital financial user credit assessment method based on big data according to claim 2 is characterized in that: The process of obtaining a user's credit score also includes the following steps: Calculate the overdue number A'=A*Qqua'*Qdue', where A is the overdue value. When the user has a loan overdue, the overdue value A=1; when the user has no loan overdue, the overdue value A=0; The ratio B of the overdue number A'≥Ayc is counted, Ayc represents the preset overdue number threshold, when the ratio B≥B' of the overdue number A'≥Ayc is reached, the user's loan request is automatically rejected within the preset time period Tyc, and B' represents the preset ratio threshold.

5. The digital financial user credit assessment method based on big data according to claim 1 is characterized in that: In the step S1, the process of obtaining the score correction coefficient specifically includes: The collection period is divided into M coefficient periods of the same length, and the ranking coefficient αM'=δ*M' / M of the M'th coefficient period is calculated, where δ represents the preset ranking correction coefficient, and when the time when the credit score changes is within the M'th coefficient period, the ranking coefficient αM' is used as the corresponding score correction coefficient.

6. The digital financial user credit assessment method based on big data according to claim 1 is characterized in that: In the step S3, the process of correcting the preset holding period according to the stability coefficient specifically includes: Calculate the stability coefficient K = γ*C / (max(ΔPJH)*S 2 ), γ represents the preset fluctuation correction coefficient, S 2 represents the variance of the user's credit score; Calculate the correction period Treg=K*Tori, where Tori represents the maintenance period. Before the time interval from the last calculation of the user's credit score reaches the length of the correction period Treg, the user's credit level is maintained at D.

7. The digital financial user credit assessment method based on big data according to claim 6 is characterized in that: When the correction period Treg>Treg'=3Tori, it is assumed that Treg=Treg'.

8. A digital financial user credit assessment system based on big data, characterized in that: include: Credit scoring module: obtains the user's historical credit score records within the preset collection period T and calculates the user's average credit score αi represents P i The score correction factor, P i represents the credit score after the i-th change in the historical credit score record, and m represents the total number of times the user's credit score has changed; Data optimization module: setting a number of credit levels from low to high with a preset credit score interval, determining the credit level D where the average credit score P' is located, and determining the credit score range [PDsta, PDend] of the credit level D; Generate data points (t i , P i ), t i Indicates the time when the user's credit score changes for the i-th time, and connects adjacent data points through straight lines to obtain the target image, and determines the duration T when the credit score on the target image is greater than or equal to PDsta fix , and calculate the stable ratio C = T fix / T; Calculate the score difference ΔP i =|P i+1 -P i |, and generate a difference set ΔPJH = (ΔP1, ΔP2, ..., ΔP m-1 ); Cycle correction module: calculates the stability coefficient and corrects the preset retention period according to the stability coefficient. The retention period is the time interval for keeping the user's credit rating unchanged.

Citation Information

Cited By

  • Multi-source information-oriented credit data management method and system

    CN119398912A