Real-time early warning method and device for credit data

Abstract

The invention discloses a real-time early warning method and device for credit data, and relates to the technical field of bank credit evaluation. The method comprises the following steps: receiving a lending request of a first user; processing the target data to obtain a first evaluation result; acquiring historical behavior information according to the first user, and evaluating the historical behavior information to obtain a second evaluation result; inputting the first evaluation result and the second evaluation result into a preset model for processing to obtain a credit rating; determining a preset loan amount according to the credit rating, and judging whether the first amount is smaller than or equal to the preset loan amount; and when the first amount is greater than the preset loan amount, determining that the first user is in a first abnormal state, and generating early warning information according to the first abnormal state so as to process the first user according to the early warning information. By implementing the technical scheme provided by the invention, delay and errors possibly occurring in manual auditing are avoided, and the overall auditing efficiency is improved.

Description

Technical Field

[0001] The present application relates to the technical field of bank credit assessment, and in particular to a real-time early warning method and device for credit data. Background Art

[0002] With the continuous evolution of financial technology, individuals and enterprises are able to deposit idle capital in banks and use the diversified investment and wealth management products provided by banks to flexibly select appropriate investment strategies based on their own risk preferences and investment goals, thereby realizing the preservation and appreciation of assets.

[0003] As a core financial institution, banks actively respond to the financing needs of individuals and enterprises in terms of capital turnover, business expansion, venture capital, etc. through credit services, thereby injecting vitality into economic growth. The essence of credit services is to effectively allocate idle funds in society to those in need of funds, so as to achieve the optimal use of funds. However, given that credit business involves the circulation of huge amounts of funds, its risk prevention and control is particularly important. To ensure the safety of funds, banks must assess the credit status of credit applicants to prevent capital losses caused by bad credit or fraudulent behavior. At present, in the credit approval process, the assessment of applicants' credit mainly relies on manual review, but in the face of massive amounts of personal information and financial data, manual review is inefficient.

[0004] Therefore, there is an urgent need for a real-time early warning method and device for credit data that can solve the above technical problems. Summary of the invention

[0005] The present application provides a real-time early warning method and device for credit data. The method automatically receives and processes a loan request from a first user to obtain a first evaluation result and a second evaluation result, quickly conducts a comprehensive analysis on the first evaluation result and the second evaluation result according to a preset model to obtain a credit rating, determines a preset loan amount according to the credit rating, and immediately identifies that the first user is in a first abnormal state when the first amount is greater than the preset loan amount, generates early warning information, and improves the overall review efficiency.

[0006] In a first aspect, the present application provides a real-time early warning method for credit data, which is applied to an early warning platform, and the method includes: receiving a loan request from a first user, the loan request includes target data and a first amount, the first amount is the amount that the first user needs to borrow, and the first user is an applicant with a loan demand; processing the target data to obtain a first evaluation result; obtaining historical behavior information based on the first user, evaluating the historical behavior information, and obtaining a second evaluation result; inputting the first evaluation result and the second evaluation result into a preset model for processing to obtain a credit rating; determining a preset loan amount based on the credit rating, and judging whether the first amount is less than or equal to the preset loan amount; when the first amount is greater than the preset loan amount, confirming that the first user is in a first abnormal state, and generating early warning information based on the first abnormal state, so that the first user can be processed according to the early warning information.

[0007] By adopting the above technical solution, a loan request from a first user is automatically received, the loan request includes target data and historical behavior information, and automatic processing greatly reduces the need for manual intervention. The first evaluation result and the second evaluation result can be quickly analyzed comprehensively through a preset model to obtain a credit rating. This rapid evaluation enables the credit approval process to be completed in a shorter time, and then the preset loan amount is determined based on the credit rating, and it is determined whether the first amount meets the preset loan amount. When the first amount exceeds the preset loan amount, it can immediately identify and confirm that the first user is in a first abnormal state, and then generate early warning information. This efficient decision-making mechanism avoids delays and errors that may occur in manual review, and improves the overall review efficiency.

[0008] Optionally, before processing the target data to obtain the first evaluation result, the method also includes: determining the application institution according to the loan request, obtaining a preset material list of the application institution, the preset material list being material list information that applicants in the application institution need to provide; extracting multiple target material names from the target data, and matching the multiple target material names with a list in the preset material list; if there is a list in the preset material list that is not matched, confirming that the target data is in a data missing state, so that the first user can supplement the target data according to the data missing state.

[0009] By adopting the above technical solution, the applicant institution is automatically determined from the loan request, and the preset material list of the institution is obtained, which reduces the time of manual query and confirmation. Multiple target material names are automatically extracted from the target data and matched with the preset material list. Based on the matching process, the missing materials in the target data can be discovered in time. If there is a list in the preset material list that is not matched, it means that the target data is in a missing material state. This material integrity check helps to confirm that the materials provided by the applicant are complete and avoid approval delays due to incomplete materials.

[0010] Optionally, the target data is processed to obtain a first evaluation result, specifically including: if all lists in the preset material list are matched, confirming that the target data is in a material completion state; extracting the target data according to the material completion state to obtain proof information, the proof information including identity card information, household registration information, marital status proof information, income proof information, personal asset proof information, loan plan information, work proof information and academic qualification proof information; judging whether the proof information is consistent with the preset information; when the proof information is inconsistent with the preset information, confirming that the target data is in a false state, and outputting the false state as the first evaluation result; when the proof information is consistent with the preset information, confirming that the target data is in a true state, and outputting the true state as the first evaluation result.

[0011] By adopting the above technical solution, when all the lists in the preset material list are matched, it can automatically confirm that the target data is in a complete data state; extract key certification information from the complete target data, including identity card information, household registration information, marital status certification information, etc., which are important bases for evaluating the applicant's credit and repayment ability; compare the extracted certification information with the preset information to determine the authenticity of the information; this verification process helps to ensure that the information used in the approval process is accurate and reliable. When the certification information is inconsistent with the preset information, it can be confirmed that the target data is in a false state and output it as the first evaluation result, which helps to discover potential fraud in a timely manner; when the certification information is consistent with the preset information, it can be confirmed that the target data is in a true state and output it as the first evaluation result, which enhances the credibility of the approval process and significantly reduces the time of manual review.

[0012] Optionally, based on the historical behavior information obtained from the first user, the historical behavior information is evaluated to obtain a second evaluation result, specifically including: obtaining the historical behavior of the first user, the historical behavior including personal loan behavior, credit card usage behavior, communication payment behavior, water and electricity payment behavior, repayment behavior, litigation behavior and overdue behavior; processing the historical behavior to obtain the number of exceptions, comparing the number of exceptions with the preset number of times, the number of exceptions being the number of times abnormal historical behaviors occur, and outputting the historical behavior as historical behavior information; when the number of exceptions is less than the preset number of times, confirming that the credit is good, and outputting the good credit as the second evaluation result; when the number of exceptions is equal to the preset number of times, confirming that the credit is average, and outputting the average credit as the second evaluation result; when the number of exceptions is greater than the preset number of times, confirming that the credit is poor, and outputting the poor credit as the second evaluation result.

[0013] By adopting the above technical solution, the first user's various historical behaviors are obtained, including personal loan behavior, credit card usage behavior, communication payment behavior, water and electricity payment behavior, repayment behavior, litigation behavior, and overdue behavior, etc., which can comprehensively and objectively evaluate the user's credit status. This comprehensive evaluation helps to more accurately reflect the user's credit risk and repayment ability. The historical behavior is processed to obtain the number of abnormalities, and the number of abnormalities is compared with the preset number, which can quantify the user's abnormal behavior. According to the comparison result of the number of abnormalities and the preset number, the credit status of the first user is divided into three levels: good credit, medium credit, and poor credit, which is helpful to formulate corresponding measures for subsequent credit levels according to the credit status.

[0014] Optionally, after determining whether the first amount is less than or equal to the preset loan amount, the method also includes: when the first amount is less than or equal to the preset loan amount, confirming the acquisition of transaction information of the first user, and extracting multiple user information from the transaction information; determining whether there is a second user in the preset user table, the preset user table is a list of prohibited borrowers, and the second user is any one of the multiple user information; when the second user does not exist in the preset user table, confirming that the first user is in a normal state.

[0015] By adopting the above technical solution, after confirming that the first amount is less than or equal to the preset loan amount, the transaction information of the first user is further obtained and in-depth analysis is performed. This process helps financial institutions implement more stringent risk prevention and control measures during the loan approval process. Automatically comparing user information with the preset user table can quickly determine whether the first user is associated with a member of the prohibited borrowing list. If the first user does not have an association with the preset member table, it is confirmed that the first user is in a normal state. This automated processing not only improves the approval efficiency, but also significantly improves the accuracy of the approval, avoiding credit risks caused by human negligence or misjudgment. Optionally, after determining whether there is a second user in the preset user table, the method also includes: when the second user exists in the preset user table, confirming that the first user is in a second abnormal state, marking the second user according to the second abnormal state, so as to subsequently monitor the person in the second abnormal state.

[0016] By adopting the above technical solution, when there is a user (i.e., the second user) in the preset member table that matches the transaction information of the first user, it can be quickly confirmed that the first user is in the second abnormal state, so that it can promptly discover and respond to potential credit risks, thereby effectively preventing and controlling the occurrence of non-performing loans; marking and monitoring users in the second abnormal state will help improve the compliance of financial institutions' businesses. Optionally, after the second user exists in the preset user table, the method also includes: obtaining a target number of people, the target number of people being the number of users in the preset user table; determining whether the target number of people is less than or equal to a preset threshold; when the target number of people is greater than the preset threshold, confirming that the first user is in a third abnormal state.

[0017] By adopting the above technical solution, obtaining the target number of people and comparing the target number of people with the preset threshold, the risk concentration directly or indirectly associated with the first user can be monitored. When the target number of people exceeds the preset threshold, it means that the first user may be involved in a higher-risk environment or group, which helps financial institutions to identify potential risks in a timely manner.

[0018] In a second aspect of the present application, a real-time early warning device for credit data is provided, the device is an early warning platform, and the early warning platform includes a receiving unit, a processing unit and a confirmation unit; the receiving unit receives a loan request from a first user, the loan request includes target data and a first amount, the first amount is the amount that the first user needs to borrow, and the first user is an applicant with a loan demand; the processing unit processes the target data to obtain a first evaluation result; obtains historical behavior information based on the first user, evaluates the historical behavior information, and obtains a second evaluation result; inputs the first evaluation result and the second evaluation result into a preset model for processing to obtain a credit rating; determines a preset loan amount based on the credit rating, and determines whether the first amount is less than or equal to the preset loan amount; the confirmation unit confirms that the first user is in a first abnormal state when the first amount is greater than the preset loan amount, and generates early warning information based on the first abnormal state, so as to process the first user based on the early warning information.

[0019] Optionally, the receiving unit is used to determine the application institution according to the loan request, and obtain the preset material list of the application institution, which is the material list information that applicants in the application institution need to provide; the processing unit is used to extract multiple target material names from the target data, and match the multiple target material names with the lists in the preset material list; the confirmation unit is used to confirm that the target data is in a missing data state if there is a list in the preset material list that is not matched, so that the first user can supplement the target data according to the missing data state.

[0020] Optionally, the confirmation unit is used to confirm that the target data is in a data completion state if all lists in the preset material list are matched; the processing unit is used to extract the target data according to the data completion state to obtain proof information, the proof information including identity card information, household registration information, marital status proof information, income proof information, personal asset proof information, loan plan information, work proof information and academic qualification proof information; determine whether the proof information is consistent with the preset information; the confirmation unit is used to confirm that the target data is in a false state when the proof information is inconsistent with the preset information, and output the false state as the first evaluation result; when the proof information is consistent with the preset information, confirm that the target data is in a true state, and output the true state as the first evaluation result.

[0021] Optionally, the receiving unit is used to obtain the historical behavior of the first user, the historical behavior including personal loan behavior, credit card usage behavior, communication payment behavior, water and electricity payment behavior, repayment behavior, litigation behavior and overdue behavior; the processing unit is used to process the historical behavior, obtain the number of exceptions, compare the number of exceptions with the preset number of times, the number of exceptions is the number of times abnormal historical behavior occurs, and output the historical behavior as historical behavior information; the confirmation unit is used to confirm that the credit is good when the number of exceptions is less than the preset number of times, and output the good credit as the second evaluation result; when the number of exceptions is equal to the preset number of times, confirm that the credit is medium, and output the medium credit as the second evaluation result; when the number of exceptions is greater than the preset number of times, confirm that the credit is poor, and output the poor credit as the second evaluation result.

[0022] Optionally, the receiving unit is used to confirm the acquisition of transaction information of the first user when the first amount is less than or equal to the preset loan amount, and extract multiple user information from the transaction information; the processing unit is used to determine whether there is a second user in the preset user table, the preset user table is a list of prohibited borrowers, and the second user is any one of the multiple user information; the confirmation unit is used to confirm that the first user is in a normal state when the second user does not exist in the preset user table.

[0023] Optionally, the confirmation unit is used to confirm that the first user is in a second abnormal state when there is a second user in the preset user table, and mark the second user according to the second abnormal state to facilitate subsequent monitoring of the person in the second abnormal state.

[0024] Optionally, the receiving unit is used to obtain a target number of people, which is the number of users in a preset user table; the processing unit is used to determine whether the target number of people is less than or equal to a preset threshold; and the confirmation unit is used to confirm that the first user is in a third abnormal state when the target number of people is greater than the preset threshold.

[0025] In a third aspect of the present application, an electronic device is provided, which includes a processor, a memory, a user interface and a network interface, the memory is used to store instructions, the user interface and the network interface are used to communicate with other devices, and the processor is used to execute the instructions stored in the memory, so that an electronic device executes any one of the methods described above in the present application.

[0026] In a fourth aspect of the present application, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores instructions, and when the instructions are executed, any one of the above methods of the present application is executed.

[0027] In summary, one or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages: 1. Automatically receive the loan request of the first user, which includes target data and historical behavior information. Automatic processing greatly reduces the need for manual intervention. The first evaluation result and the second evaluation result can be quickly analyzed through the preset model to obtain the credit rating. This rapid evaluation enables the credit approval process to be completed in a shorter time. The preset loan amount is then determined based on the credit rating, and it is determined whether the first amount meets the preset loan amount. When the first amount exceeds the preset loan amount, it can immediately identify and confirm that the first user is in the first abnormal state, and then generate early warning information. This efficient decision-making mechanism avoids delays and errors that may occur in manual review and improves the overall review efficiency.

[0028] 2. Automatically identify the applicant institution from the loan request and obtain the preset material list of the institution, reducing the time for manual query and confirmation. Automatically extract multiple target material names from the target data and match them with the preset material list. Based on the matching process, it is possible to promptly discover missing materials in the target data. If there is a list in the preset material list that is not matched, it means that the target data is in a missing data state. This data integrity check helps to confirm that the materials provided by the applicant are complete and avoid approval delays due to incomplete materials. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 It is a flowchart of a real-time early warning method for credit data provided by an embodiment of the present application; Figure 2 It is a structural diagram of a real-time early warning device for credit data provided by an embodiment of the present application; Figure 3 It is a structural schematic diagram of an electronic device disclosed in an embodiment of the present application.

[0030] Description of reference numerals: 201, receiving unit; 202, processing unit; 203, confirmation unit; 300, electronic device; 301, processor; 302, memory; 303, user interface; 304, network interface; 305, communication bus. DETAILED DESCRIPTION

[0031] In order to enable technicians in this field to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the drawings in the embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this application, not all of the embodiments.

[0032] In the description of the embodiments of the present application, words such as "for example" or "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "for example" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "for example" or "for example" is intended to present related concepts in a specific way.

[0033] In the description of the embodiments of the present application, the meaning of the term "multiple" refers to two or more. For example, multiple systems refer to two or more systems, and multiple screen terminals refer to two or more screen terminals. In addition, the terms "first" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance or implicitly indicating the indicated technical features. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. The terms "include", "comprise", "have" and their variations all mean "including but not limited to", unless otherwise specifically emphasized.

[0034] With the continuous evolution of financial technology, individuals and enterprises are able to deposit idle capital in banks and use the diversified investment and wealth management products provided by banks to flexibly select appropriate investment strategies based on their own risk preferences and investment goals, thereby realizing the preservation and appreciation of assets.

[0035] As a core financial institution, banks actively respond to the financing needs of individuals and enterprises in terms of capital turnover, business expansion, venture capital, etc. through credit services, thereby injecting vitality into economic growth. The essence of credit services is to effectively allocate idle funds in society to those in need of funds, so as to achieve the optimal use of funds. However, given that credit business involves the circulation of huge amounts of funds, its risk prevention and control is particularly important. To ensure the safety of funds, banks must assess the credit status of credit applicants to prevent capital losses caused by bad credit or fraudulent behavior. At present, in the credit approval process, the assessment of applicants' credit mainly relies on manual review, but in the face of massive amounts of personal information and financial data, manual review is inefficient.

[0036] Therefore, how to solve the problem of low efficiency of manual review in the loan approval process. The real-time early warning method of credit data provided in the embodiment of the present application is applied to the early warning platform. The early warning platform of the present application can be a platform that provides credit assessment for financial structures. Figure 1 This is a flow chart of a real-time early warning method for credit data provided by an embodiment of the present application, with reference to Figure 1 The method includes the following steps S101-S106.

[0037] S101: Receive a loan request from a first user, where the loan request includes target data and a first amount.

[0038] In the above S101, the financial institution platform will provide a loan application interface, the financial institution refers to a bank institution, the first user refers to an applicant with a loan demand, and the first user fills in the relevant information through the interface and submits the loan request. The loan request includes target data and a first amount. The target data may include personal identity information, financial information, occupational information, etc., which is used to evaluate the credit status and repayment ability of the first user. The first amount is the specific amount that the first user wants to borrow.

[0039] S102: Process the target data to obtain a first evaluation result.

[0040] In the above S102, after receiving the loan request, the financial institution platform first conducts a preliminary review and verification of the target data to ensure its authenticity and integrity. That is, the application institution is determined according to the loan request, and the preset material list of the application institution is obtained. The preset material list is the material list information that the applicant in the application institution needs to provide; multiple target material names are extracted from the target data, and the multiple target material names are matched with the list in the preset material list; if there is a list in the preset material list that is not matched, it is confirmed that the target data is in a data missing state, so that the first user can supplement the target data according to the data missing state. Specifically, a loan request from the first user is received. The request contains key information such as the target data and the first amount. The information in the loan request is parsed to determine the application institution for the first user to apply for the loan. At this time, the application institution refers to the name of the financial institution where the first user handles the loan business, and the application institution is determined according to the name. Then the specific application institution is determined. The application institution can be based on user selection or based on the recommendation of the platform. Once the application institution is determined, the preset material list of the institution will be retrieved from the database. This list lists in detail the various materials that the applicant needs to provide, such as identity proof, income proof, property proof, etc. Different application institutions may require different numbers of materials from applicants when handling loan business, so a preset material list is determined according to the application institution. Parse the target data to identify the various information contained therein. Use natural language processing or text parsing technology to extract multiple target material names related to the application materials from the target data. These names usually correspond to the material names in the preset material list. Store the extracted target material names in a temporary variable or database for subsequent matching with the preset material list. Use string matching, regular expression or other matching algorithms to match the target material names with the materials in the preset material list one by one. Record the matching results of each target material name, including whether the match is successful, the matched material name and the matching degree. Check the matching results of all target material names, especially those material names that are not successfully matched. If there is at least one unmatched material name in the preset material list, it is confirmed that the target data is in a missing data state. This means that the materials provided by the first user are incomplete and need to be supplemented. Generate a reminder message to inform the first user which materials are missing and suggest that the first user supplement these materials as soon as possible. This information can be sent to the first user by email, SMS or system notification. Wait for the first user to supplement the target data according to the reminder information. The first user can supplement the materials by uploading new files, filling in new information or modifying existing information. Once the first user submits the supplementary materials, these materials are re-verified to ensure that they meet the requirements and regulations of the applicant organization.The above method can help improve the approval efficiency and accuracy of loan requests, while reducing the risk of approval delays or rejections due to incomplete materials.

[0041] In addition, if all lists in the preset material list are matched, it is confirmed that the target data is in a data completion state; the target data is extracted according to the data completion state to obtain proof information, which includes identity card information, household registration information, marital status proof information, income proof information, personal asset proof information, loan plan information, work proof information and academic qualification proof information; it is determined whether the proof information is consistent with the preset information; when the proof information is inconsistent with the preset information, it is confirmed that the target data is in a false state, and the false state is output as the first evaluation result; when the proof information is consistent with the preset information, it is confirmed that the target data is in a true state, and the true state is output as the first evaluation result.

[0042] Specifically, the results of matching all lists in the preset material list with the target data will be checked. This step is performed automatically and depends on the matching algorithm. If it is found that all lists in the preset material list have been successfully matched, it will confirm that the target data is in a complete data state. This means that the first user has provided all the necessary application materials without omission. After confirming that the target data is in a complete data state, the target data will be deeply analyzed to extract various certification information. This information includes but is not limited to identity card information, household registration information, marital status certification information, income certification information, personal asset certification information, loan plan information, work certification information, and academic certification information. According to the identity information provided by the first user, the preset information of the first user can be obtained from the authorized third-party channel, and then the extracted certification information is compared with the preset information one by one to check whether they are consistent. This step may involve a variety of comparison algorithms and technologies to ensure the accuracy and efficiency of the comparison. If the certification information is marital status certification information, the marital status is queried according to the identity information of the first user, and the queried marital status is output as the preset information. When querying the corresponding preset information, the authorization authority of the channel has been obtained in advance, and the relevant information can be queried. If the proof information is income proof information, the bank flow information is retrieved according to the identity information of the first user, and the bank flow information is output as the preset information. The income proof information is then compared with the bank flow information. If the two are inconsistent, it is confirmed that the proof information is inconsistent with the preset information. At this time, the target data is in a false state, which means that the first user may have provided false or forged income information. After confirming that the target data is in a false state, this result will be immediately output as the first evaluation result. If the two are consistent, it is confirmed that the proof information is consistent with the preset information. At this time, the target data is in a true state, which means that the income information in the application materials provided by the first user is true and valid. After confirming that the target data is in a true state, this result will also be output as the first evaluation result. In this application, as long as there is a situation in the proof information that is inconsistent with the preset information, it can be confirmed that the target data provided by the first user is in a false state. The first evaluation result reflects the authenticity of the target data provided by the first user.

[0043] S103: Acquire historical behavior information according to the first user, evaluate the historical behavior information, and obtain a second evaluation result.

[0044] In the above S103, the financial institution or loan platform obtains the historical behavior information of the first user through an internal database or a third-party data source. According to the historical behavior information obtained by the first user, the historical behavior is evaluated to obtain a second evaluation result, which specifically includes: obtaining the historical behavior of the first user, the historical behavior includes personal loan behavior, credit card usage behavior, communication payment behavior, water and electricity payment behavior, repayment behavior, litigation behavior and overdue behavior; processing the historical behavior to obtain the number of abnormalities, comparing the number of abnormalities with the preset number of times, the abnormal number of times is the number of times abnormal historical behaviors occur, and the historical behavior is output as historical behavior information; when the number of abnormalities is less than the preset number of times, confirming that the credit is good, and outputting the good credit as the second evaluation result; when the number of abnormalities is equal to the preset number of times, confirming that the credit is medium, and outputting the medium credit as the second evaluation result; when the number of abnormalities is greater than the preset number of times, confirming that the credit is poor, and outputting the poor credit as the second evaluation result.

[0045] Specifically, the historical behavior data of the first user is collected from multiple data sources. These data sources may include banks, credit card companies, communication operators, water and electricity companies, courts, etc., which respectively record the user's loan, credit card usage, communication payment, water and electricity payment, repayment, litigation, and overdue behavior. The collected data may come from different formats and platforms, so it is necessary to integrate and standardize these data for subsequent analysis. In the process of integrating data, data cleaning is also required, including removing duplicate data, filling missing values, correcting erroneous data, etc., to ensure the accuracy and completeness of the data. After obtaining historical behaviors, it is necessary to first clarify what abnormal behaviors include, such as overdue repayments, frequent litigation, malicious credit card overdrafts, etc., which may be considered abnormal. For each user, their historical behaviors will be analyzed and the number of abnormal behaviors will be counted. This step may involve complex algorithms and models to accurately determine which behaviors are abnormal. After the analysis is completed, the historical behavior information of the first user (including abnormal behaviors and their number) will be output for subsequent evaluation. Compare the number of abnormalities with the preset number. The preset number is pre-set and used to judge the user's credit status. These thresholds may be set based on industry standards, institutional policies or data analysis results. Compare the number of abnormalities of the user with the preset number. If the number of abnormalities is less than the preset number, the user's credit is judged to be good; if it is equal to the preset number, the credit is judged to be medium; if it is greater than the preset number, the credit is judged to be poor. Based on the comparison result, the corresponding second assessment result (good credit, medium credit or poor credit) will be output. This result may be used in subsequent loan approval, credit assessment or risk management processes. When processing the user's historical behavior data, it is necessary to strictly comply with relevant laws, regulations and privacy policies to ensure the security and privacy of user data.

[0046] For example, if the first user has overdue payments and credit card overdrafts, in this application, when the user only has one or two abnormal behaviors, the number of times the user has each abnormal behavior in the past year can be checked, and then the number of times each abnormal behavior occurred in the past year can be added up to get the number of abnormalities.

[0047] Get the number of overdue payments for the first user in the past year. If the number of overdue payments is 3 and the number of credit card overdrafts is 2, then the number of abnormal behaviors in the past year needs to be superimposed to obtain 5 abnormal times. When there is a time limit, that is, there is no need to check the number of times each abnormal behavior of the user occurred in the past year, the preset number needs to be set to 4 times. If the number of abnormal behaviors is greater than the preset number, it indicates that the first user has a medium credit rating. When the user has more than two abnormal behaviors, the number of abnormal behaviors of the user can be directly obtained. In this case, one abnormal behavior corresponds to one time. If the first user has overdue payments, credit card overdrafts, and lawsuits, the number of abnormalities is 3. When there is no time limit, the preset number can be set to 2 times. If the number of abnormal behaviors is greater than the preset number, it indicates that the first user has a poor credit rating.

[0048] S104: Input the first evaluation result and the second evaluation result into a preset model for processing to obtain a credit rating.

[0049] In the above S104, the target data is processed to obtain a first evaluation result, which includes a true state and a false state. The historical behavior information is processed to obtain a second evaluation result, and the third evaluation result includes good credit, medium credit, and poor credit. The first evaluation result and the second evaluation result are used as inputs and input into the preset model for processing. The preset model accurately evaluates the user's credit rating according to the corresponding situations of the first evaluation result and the second evaluation result. Since there are two situations for the first evaluation result and three situations for the second evaluation result, the credit rating can be divided into 6 levels, and the 6 credit levels are introduced in turn. When the first evaluation result is a true state and the second evaluation result is a good credit, the credit rating can be set to excellent and A level can be used, because the true state and good credit mean that the individual has a good credit record. When the first evaluation is a true state and the second evaluation result is a medium credit, the credit rating can be set to good and B level can be used, indicating that the credit record is general but still true. When the first evaluation is a true state and the second evaluation result is a poor credit, the credit rating can be set to general and C level can be used, indicating that although the information is true, the credit record is not good. When the first evaluation result is false, no matter the second evaluation result is good, medium or poor credit, since the first evaluation result is false, it means that there is a problem with credit. When the first evaluation result is false, the credit rating can be set to untrustworthy when the second evaluation result is any of good, medium and poor credit, and D can be used. However, in order to more carefully reflect the difference in credit situation, it can still be slightly distinguished: if the second evaluation result is good credit, the credit rating can be set to D1, indicating that there is a false credit record but the credit record is good. The second evaluation result is medium credit: the credit rating is D2, indicating that it is false and the credit record is average. The second evaluation result is poor credit: the credit rating is D3, indicating that it is false and the credit record is poor. Credit rating is a numerical value or classification label used to indicate the user's credit status and repayment ability.

[0050] S105: Determine a preset loan amount according to the credit rating, and determine whether the first amount is less than or equal to the preset loan amount.

[0051] In the above S105, the financial institution platform determines a preset loan amount according to the credit rating of the first user. This amount is usually set according to the user's credit status, repayment ability and potential risk. As mentioned above, there are 6 credit ratings, and different credit ratings correspond to different preset loan amounts. The higher the rating, the higher the loan amount, and the lower the rating, the lower the loan amount. The first amount (i.e., the amount the user wants to borrow) is compared with the preset loan amount.

[0052] S106: When the first amount is greater than the preset loan amount, confirm that the first user is in a first abnormal state, generate warning information according to the first abnormal state, so as to process the first user according to the warning information.

[0053] In the above S106, if the first amount is greater than the preset loan amount, it is considered that the first user's loan request exceeds the range allowed by his credit status and repayment ability, so the first user is marked as the first abnormal state. When the user is in the first abnormal state, an early warning message is automatically generated. The early warning information may include the user's identity information, loan request details, abnormal state description, etc. After receiving the early warning information, the relevant personnel of the financial institution platform can conduct further review and investigation on the first user to decide whether to approve his loan request or take other measures. Avoid fraudulent behavior of the first user due to failure to investigate the first user. The user's loan request is comprehensively evaluated and processed through the above method to ensure that the risks of the financial institution or loan platform are controllable and the legitimate rights and interests of the user are protected.

[0054] For example, when the first user's credit rating is C, the loan amount corresponding to C is 100,000 yuan. In the loan request sent by the first user, the amount to be borrowed is 300,000 yuan. At this time, the first amount sent by the first user is greater than the preset loan amount. It is confirmed that the first user is marked as a first abnormal state to prevent the first user from defrauding funds.

[0055] In addition, when the first amount is less than or equal to the preset loan amount, confirm that the transaction information of the first user is obtained, and extract multiple user information from the transaction information; determine whether there is a second user in the preset user table, the preset user table is a list of prohibited borrowers, and the second user is any one of the multiple user information; when the second user does not exist in the preset user table, confirm that the first user is in a normal state. Specifically, if the first amount is less than or equal to the preset loan amount, the transaction information of the first user will be further obtained. This information may include the user's bank account transaction record, third-party payment platform transaction record, etc. Parse the transaction information and extract multiple user information from it. This information may include the name of the transaction counterparty, transaction amount, transaction time, transaction type, etc. Retrieve the preset user table from the database, which is a list of prohibited borrowers, including user information that is prohibited from borrowing for various reasons. Match the extracted multiple user information (at this time as the information to be matched) with the user information in the preset user table one by one to determine whether there is a second user (that is, a user in the list of prohibited borrowers). Record the matching results of each user information, including whether the match is successful, the matched user information, and the matching degree. Check the matching results of all user information, especially those that are successfully matched. If there is no second user matching the extracted user information in the preset user table, then it will be confirmed that the first user is in a normal state, that is, other users who have transactions with the first user are not included in the list of prohibited borrowers, and the subsequent loan approval process for the first user can continue. The normal state can also be understood as the first user is a normal borrower, and the loan process for the first user can continue to be approved.

[0056] Further, when there is a second user in the preset user table, it is confirmed that the first user is in the second abnormal state, and the second user is marked according to the second abnormal state, so as to monitor the person in the second abnormal state in the future. Specifically, after matching the user information, the matching result will be checked. If it is found that there is a second user matching the extracted user information in the preset user table, it means that the first user is associated with a user in the list of prohibited borrowers. Based on the matching result, the first user is immediately determined to be in the second abnormal state. This state indicates that the first user may be involved in illegal borrowing, fraudulent behavior or other risk situations, and further investigation and processing are required. Prepare marking information, including the identity of the second user, the associated first user information, the abnormal state type (i.e., the second abnormal state), the marking time, etc. Update the marking information to the database, especially the record associated with the second user. This ensures that the person in the second abnormal state can be easily queried and tracked in the future. The marking method may include setting a specific status field in the database, adding remark information, or generating a special monitoring list, etc. The specific method depends on the design and business requirements of the system. Establish a monitoring mechanism for people in the second abnormal state. This may include regularly querying the database, setting an alarm system, or linking with other systems, etc. Through the monitoring mechanism, the behavior of the second user (and the first user associated with it) is continuously monitored. If new abnormal behavior or risk situations are found, an alarm will be triggered immediately and handled. Depending on the nature and severity of the abnormal behavior, the system will formulate corresponding handling measures. This may include rejecting loan applications, restricting account operations, conducting further investigations, etc.

[0057] Furthermore, after determining that the second user exists in the preset member table, it is also necessary to obtain the number of times the user appears in the predicted member table, and judge the current risk state of the first user according to the number of times, so as to realize the identification and monitoring of potential high-risk users. Obtain the target number of people, which is the number of users in the preset user table; judge whether the target number of people is less than or equal to the preset threshold; when the target number of people is greater than the preset threshold, it is confirmed that the first user is in the third abnormal state. Specifically, it is necessary to access and query the preset user table, which contains all user information marked as prohibited lending or other specific states. Query other users who have transaction records with the first user, match the other users with transaction records with the preset member table in turn, obtain matching results, and then convert the matching results into the target number of people, that is, a single user matches the preset member table, and the target number of people is 1 at this time. If two users match the members on the preset member table, the target number of people is 2 at this time. After obtaining the target number of people, compare the target number of people with the preset threshold, which is set according to business rules, risk strategies or policy requirements, and is used to determine whether the target number of people has reached a level that requires special attention. If the target number is greater than the preset threshold, it means that the first user is associated with a high-risk user and the first user is in a high-risk group, which is composed of people in the preset member table. At this time, the first user has a higher risk or problem. When it is determined that the first user is closely connected with the high-risk group, this close relationship meets the conditions of the third abnormal state defined in the business rules or risk strategy, and the first user will be confirmed to be in the third abnormal state. Once it is confirmed that the first user is in the third abnormal state, appropriate follow-up measures need to be taken, such as suspending his loan application or conducting further investigation and analysis. For example, other users with transaction records with the first user are obtained. If there are 20 other users, then determine whether there are other users in the preset member table in turn. If there are 4 users in the preset member table, the target number is 4 at this time, and the preset threshold can be set to 3. At this time, the target number is greater than the preset threshold, and the first user can be set to the third abnormal state. The third abnormal state indicates that the first user has fraudulent behavior. In addition, when the target number is less than the preset threshold, it is confirmed that the first user is in the first abnormal state.

[0058] The present application also provides a real-time early warning device for credit data. Figure 2 This is a schematic diagram of the structure of a real-time early warning device for credit data provided by an embodiment of the present application, with reference to Figure 2 The device is an early warning platform, which includes a receiving unit 201, a processing unit 202 and a confirmation unit 203.

[0059] The receiving unit 201 receives a loan request from a first user, where the loan request includes target data and a first amount, where the first amount is the amount that the first user needs to borrow, and the first user is an applicant with a loan demand.

[0060] The processing unit 202 processes the target data to obtain a first evaluation result; obtains historical behavior information based on the first user, evaluates the historical behavior information, and obtains a second evaluation result; inputs the first evaluation result and the second evaluation result into a preset model for processing to obtain a credit rating; determines a preset loan amount based on the credit rating, and determines whether the first amount is less than or equal to the preset loan amount.

[0061] The confirmation unit 203 confirms that the first user is in a first abnormal state when the first amount is greater than a preset loan limit, and generates warning information according to the first abnormal state so as to process the first user according to the warning information.

[0062] In a possible implementation, the receiving unit 201 is used to determine the application institution according to the loan request, and obtain the preset material list of the application institution, which is the material list information that the applicants in the application institution need to provide; the processing unit 202 is used to extract multiple target material names from the target data, and match the multiple target material names with the lists in the preset material list; the confirmation unit 203 is used to confirm that the target data is in a missing data state if there is a list in the preset material list that is not matched, so that the first user can supplement the target data according to the missing data state.

[0063] In one possible implementation, the confirmation unit 203 is used to confirm that the target data is in a data completion state if all lists in the preset material list are matched; the processing unit 202 is used to extract the target data according to the data completion state to obtain proof information, the proof information including identity card information, household registration information, marital status proof information, income proof information, personal asset proof information, loan plan information, work proof information and academic qualification proof information; determine whether the proof information is consistent with the preset information; the confirmation unit 203 is used to confirm that the target data is in a false state when the proof information is inconsistent with the preset information, and output the false state as the first evaluation result; when the proof information is consistent with the preset information, confirm that the target data is in a true state, and output the true state as the first evaluation result.

[0064] In a possible implementation, the receiving unit 201 is used to obtain the historical behavior of the first user, the historical behavior including personal loan behavior, credit card usage behavior, communication payment behavior, water and electricity payment behavior, repayment behavior, litigation behavior and overdue behavior; the processing unit 202 is used to process the historical behavior, obtain the number of exceptions, compare the number of exceptions with the preset number of times, the number of exceptions is the number of times abnormal historical behaviors occur, and output the historical behavior as historical behavior information; the confirmation unit 203 is used to confirm that the credit is good when the number of exceptions is less than the preset number of times, and output the good credit as the second evaluation result; when the number of exceptions is equal to the preset number of times, confirm that the credit is medium, and output the medium credit as the second evaluation result; when the number of exceptions is greater than the preset number of times, confirm that the credit is poor, and output the poor credit as the second evaluation result.

[0065] In one possible implementation, the receiving unit 201 is used to confirm the acquisition of the transaction information of the first user when the first amount is less than or equal to the preset loan amount, and extract multiple user information from the transaction information; the processing unit 202 is used to determine whether there is a second user in the preset user table, the preset user table is a list of prohibited borrowers, and the second user is any one of the multiple user information; the confirmation unit 203 is used to confirm that the first user is in a normal state when the second user does not exist in the preset user table.

[0066] In a possible implementation, the confirmation unit 203 is used to confirm that the first user is in a second abnormal state when there is a second user in the preset user table, and mark the second user according to the second abnormal state to facilitate subsequent monitoring of the person in the second abnormal state.

[0067] In one possible implementation, the receiving unit 201 is used to obtain a target number of people, which is the number of users in a preset user table; the processing unit 202 is used to determine whether the target number of people is less than or equal to a preset threshold; and the confirmation unit 203 is used to confirm that the first user is in a third abnormal state when the target number of people is greater than the preset threshold.

[0068] It should be noted that: when the device provided in the above embodiment realizes its function, only the division of the above functional modules is used as an example. In actual application, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the device and method embodiments provided in the above embodiment belong to the same concept, and the specific implementation process is detailed in the method embodiment, which will not be repeated here.

[0069] The present application also discloses an electronic device. Figure 3 , Figure 3The electronic device 300 may include: at least one processor 301 , at least one network interface 304 , a user interface 303 , a memory 302 , and at least one communication bus 305 .

[0070] The communication bus 305 is used to realize the connection and communication between these components.

[0071] The user interface 303 may include a display screen (Display) and a camera (Camera). Optionally, the user interface 303 may also include a standard wired interface and a wireless interface.

[0072] The network interface 304 may optionally include a standard wired interface or a wireless interface (such as a WI-FI interface).

[0073] Among them, the processor 301 may include one or more processing cores. The processor 301 uses various interfaces and lines to connect various parts in the entire server, and executes various functions of the server and processes data by running or executing instructions, programs, code sets or instruction sets stored in the memory 302, and calling data stored in the memory 302. Optionally, the processor 301 can be implemented in at least one hardware form of digital signal processing (Digital Signal Processing, DSP), field programmable gate array (Field-Programmable Gate Array, FPGA), and programmable logic array (Programmable Logic Array, PLA). The processor 301 can integrate one or a combination of a central processing unit (Central Processing Unit, CPU), a graphics processing unit (Graphics Processing Unit, GPU) and a modem. Among them, the CPU mainly processes the operating system, user interface and application requests; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor 301, and it can be implemented separately through a chip.

[0074] Among them, the memory 302 may include a random access memory (RAM) or a read-only memory (Read-Only Memory). Optionally, the memory 302 includes a non-transitory computer-readable storage medium. The memory 302 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 302 may include a program storage area and a data storage area, wherein the program storage area. Instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned various method embodiments, etc. can be stored; the data storage area can store data involved in the above-mentioned various method embodiments, etc. The memory 302 can also be optionally at least one storage device located away from the aforementioned processor 301.

[0075] like Figure 3 As shown, the memory 302 as a computer storage medium may include an operating system, a network communication module, a user interface module, and an application for real-time early warning of credit data.

[0076] exist Figure 3 In the electronic device 300 shown, the user interface 303 is mainly used to provide an input interface for the user and obtain data input by the user; and the processor 301 can be used to call the real-time warning application for credit data stored in the memory 302. When executed by one or more processors, the electronic device executes one or more methods described in the above embodiments.

[0077] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are all described as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the order of the actions described, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required for the present application.

[0078] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0079] In the several embodiments provided in the present application, it should be understood that the disclosed devices can be implemented in other ways. For example, the device embodiments described above are only schematic, such as the division of the units, which is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some service interfaces, and the indirect coupling or communication connection of devices or units can be electrical or other forms.

[0080] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0081] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0082] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a memory and includes several instructions for a computer device (which can be a personal computer, server or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned memory includes: various media that can store program codes, such as USB flash drives, mobile hard drives, magnetic disks or optical disks.

[0083] The above is only an exemplary embodiment of the present disclosure and cannot be used to limit the scope of the present disclosure. That is, any equivalent changes and modifications made according to the teachings of the present disclosure are still within the scope of the present disclosure. After considering the disclosure of the specification and the truth of practice, it will be easy for those skilled in the art to think of other embodiments of the present disclosure. This application is intended to cover any variation, use or adaptive change of the present disclosure, which follows the general principles of the present disclosure and includes common knowledge or customary technical means in the technical field that are not recorded in the present disclosure.

Claims

1. A real-time early warning method for credit data, characterized in that: Applied to an early warning platform, the method comprises: receiving a loan request from a first user, the loan request including target data and a first amount, the first amount being the amount that the first user needs to borrow, and the first user being an applicant with a loan demand; Processing the target data to obtain a first evaluation result; Acquiring historical behavior information of the first user, evaluating the historical behavior information to obtain a second evaluation result; Inputting the first evaluation result and the second evaluation result into a preset model for processing to obtain a credit rating; Determine a preset loan amount according to the credit rating, and determine whether the first amount is less than or equal to the preset loan amount; When the first amount is greater than the preset loan amount, it is confirmed that the first user is in a first abnormal state, and warning information is generated according to the first abnormal state, so as to process the first user according to the warning information.

2. The method according to claim 1, characterized in that Before processing the target data to obtain the first evaluation result, the method further includes: Determine the applicant institution according to the loan request, and obtain a preset material list of the applicant institution, wherein the preset material list is a list of materials that the applicant in the applicant institution needs to provide; Extracting a plurality of target material names from the target data, and matching the plurality of target material names with a list in a preset material list; If there is a list in the preset material list that is not matched, it is confirmed that the target data is in a data missing state, so that the first user can supplement the target data according to the data missing state.

3. The method according to claim 2, characterized in that The processing of the target data to obtain a first evaluation result specifically includes: If all lists in the preset material list are matched, it is confirmed that the target data is in a data complete state; Extract the target data according to the data completion status to obtain certification information, which includes identity card information, household registration information, marital status certification information, income certification information, personal asset certification information, loan plan information, work certification information and academic certification information; Determine whether the certification information is consistent with the preset information; When the certification information is inconsistent with the preset information, confirming that the target data is in a false state, and outputting the false state as the first evaluation result; When the certification information is consistent with the preset information, it is confirmed that the target data is in a true state, and the true state is output as the first evaluation result.

4. The method according to claim 1, characterized in that: The acquiring of historical behavior information according to the first user and evaluating the historical behavior information to obtain a second evaluation result specifically includes: Acquiring the first user's historical behaviors, the historical behaviors including personal loan behaviors, credit card usage behaviors, communication payment behaviors, utility payment behaviors, repayment behaviors, litigation behaviors, and overdue behaviors; Processing the historical behavior to obtain an abnormal number, comparing the abnormal number with a preset number, the abnormal number being the number of times the abnormal historical behavior occurs, and outputting the historical behavior as the historical behavior information; When the number of abnormalities is less than the preset number, confirming that the credit is good, and outputting the good credit as the second evaluation result; When the number of abnormalities is equal to the preset number of times, the credit rating is determined to be average, and the average credit rating is output as the second evaluation result; When the number of abnormalities is greater than the preset number, the credit rating is determined to be poor, and the poor credit rating is output as the second evaluation result.

5. The method according to claim 1, characterized in that After determining whether the first amount is less than or equal to a preset loan amount, the method further includes: When the first amount is less than or equal to the preset loan amount, confirming to obtain the transaction information of the first user, and extracting multiple user information from the transaction information; Determine whether there is a second user in the preset user table, the preset user table is a list of prohibited borrowers, and the second user is any one of the multiple user information; When the second user does not exist in the preset user table, it is confirmed that the first user is in a normal state.

6. The method according to claim 5, characterized in that After determining whether there is a second user in the preset user table, the method further includes; When the second user exists in the preset user table, it is confirmed that the first user is in a second abnormal state, and the second user is marked according to the second abnormal state, so as to subsequently monitor the person in the second abnormal state.

7. The method according to claim 6, characterized in that After the second user exists in the preset user table, the method further includes: Obtaining a target number of people, the target number of people being the number of users in the preset user table; Determine whether the target number of people is less than or equal to a preset threshold; When the target number of people is greater than the preset threshold, it is confirmed that the first user is in a third abnormal state.

8. A real-time early warning device for credit data, characterized in that: The device is an early warning platform, comprising a receiving unit (201), a processing unit (202) and a confirmation unit (203); The receiving unit (201) receives a loan request from a first user, the loan request comprising target data and a first amount, the first amount being the amount that the first user needs to borrow, and the first user being an applicant with a loan demand; The processing unit (202) processes the target data to obtain a first evaluation result; obtains historical behavior information of the first user, evaluates the historical behavior information, and obtains a second evaluation result; Inputting the first evaluation result and the second evaluation result into a preset model for processing to obtain a credit rating; Determine a preset loan amount according to the credit rating, and determine whether the first amount is less than or equal to the preset loan amount; A confirmation unit (203) confirms that the first user is in a first abnormal state when the first amount is greater than the preset loan limit, and generates warning information according to the first abnormal state so as to process the first user according to the warning information.

9. An electronic device, characterized in that: The electronic device (300) comprises a processor (301), a memory (302), a user interface (303) and a network interface (304), wherein the memory (302) is used to store instructions, the user interface (303) and the network interface (304) are used to communicate with other devices, and the processor (301) is used to execute the instructions stored in the memory (302) so that the electronic device (300) executes the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores instructions, and when the instructions are executed, the method according to any one of claims 1 to 7 is executed.

Images