Internal and external data security interaction method based on data analysis

Abstract

The invention relates to the technical field of data interaction, and discloses a data analysis-based internal and external data security interaction method, which comprises the steps of obtaining target submission information, forming a preliminary evaluation model, forming acquisition data, forming judgment data, forming reference data, forming analysis data and generating an evaluation result of a target customer. According to the internal and external data security interaction method based on data analysis, in the internal and external data interaction process, communication permission, communication mode agreement and communication message security analysis are performed on an external communication party, internal standardization processing, communication encryption and communication flow monitoring are performed on an external interface message, and internal and external data interaction is performed through an open platform; the data acquisition channel is more diversified and safer, enterprise evaluation errors caused by a single data channel are avoided, and enterprise losses are reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of data interaction, and in particular to a method for secure interaction of internal and external data based on data analysis. Background Art

[0002] Data security exchange refers to taking a series of measures to protect the integrity, confidentiality and availability of data during the data transmission process. Data security exchange between networks requires data interaction across networks, network segments and even organizations, which has higher requirements for data transmission. The core of the demand for intranet and extranet data exchange lies in "security, efficiency and compliance", and the diversity of application scenarios requires customized solutions for different enterprises. By combining business characteristics and security level requirements, enterprises can choose appropriate technical solutions to achieve efficient and secure intranet and extranet data exchange.

[0003] To ensure data and network security, consumer finance companies have implemented strict physical isolation between external networks and the Internet. However, in some scenarios, they must conduct business-related network communications with the Internet to meet business development needs. Therefore, they have established a network communication intermediate system based on data analysis and business distribution. By introducing data analysis technology, they can meet the actual needs of the business while ensuring network communication security.

[0004] The existing internal and external data security interaction methods based on data analysis cannot automatically form an evaluation model for the borrower based on the borrower's income information, social security information and loan information to determine whether his repayment is stable. It cannot form an interactive database based on the data of the People's Bank of China Credit Information Center for internal and external data interaction, making the data source single, resulting in inaccurate and incomplete evaluation data. It cannot judge whether the borrower has concealed his historical loan records based on the borrower's historical loan records and data interaction with partners. When the borrower's historical loan records show that the corresponding repayments were not submitted in time, it is impossible to judge whether the behavior was caused by the borrower's active and deliberate default, which is easy to cause misjudgment or wrong judgment of the customer, resulting in the customer being unable to borrow money reasonably due to the improper behavior of others. Its practicality has certain limitations. Summary of the invention

[0005] The present invention provides a method for secure interaction of internal and external data based on data analysis, which is used to promote the solution of the problems mentioned in the background technology.

[0006] The present invention provides the following technical solution: a method for secure interaction of internal and external data based on data analysis, comprising:

[0007] Obtain target submission information from target customers;

[0008] The target submitted information includes loan information and personal information;

[0009] Submit information based on the target and form a preliminary evaluation model;

[0010] Submit information based on the target and form collected data through data collection strategy;

[0011] According to the collected data, the judgment data is formed through the data judgment strategy;

[0012] The judgment data includes whether the customer data is correct or incorrect;

[0013] If the data is judged to be customer data, the collected data will be identified as reference data;

[0014] If the data is judged to be incorrect customer data, a data correction strategy is executed to form reference data;

[0015] Based on the preliminary evaluation model and reference data, analytical data is generated through data analysis strategies;

[0016] Generate evaluation results of target customers based on the analyzed data.

[0017] As an optional solution of the internal and external data security interaction method based on data analysis of the present invention, the forming of the preliminary evaluation model is specifically as follows:

[0018] Obtaining loan information, wherein the loan information includes loan data and loan duration;

[0019] Calculate repayment data, repayment data = loan data ÷ loan duration;

[0020] Access personal information, including income and security data;

[0021] Obtain target submission information, wherein the target submission information includes submission time;

[0022] Set the collection duration;

[0023] Calculate the collection time, collection time = submission time - collection duration;

[0024] The analysis period is formed by taking the collection time as the starting time point and the submission time as the ending time point;

[0025] Obtain the target customer's income data during the analysis period to form an income set;

[0026] Sequentially extract two adjacent elements in the income set and define them as the first and last elements respectively;

[0027] If the latter element is ≥ the former element × 90%, the target customer is determined to meet the first stable data;

[0028] If the latter element is less than the former element × 90%, it is determined that the target customer does not meet the first stable data;

[0029] Get the number of elements in the collection, which is defined as the collection quantity;

[0030] Calculate the average value of the elements, the average value of the elements = the sum of all elements in the collection / the number of collections;

[0031] If the repayment data is ≤ the element average value × 50%, the target customer is determined to meet the second stable data;

[0032] If the repayment data > element average value × 50%, it is determined that the target customer does not meet the second stable data;

[0033] Implement a stable analysis strategy.

[0034] As an optional solution of the internal and external data security interaction method based on data analysis described in the present invention, the stable analysis strategy is specifically:

[0035] Obtaining safeguard data;

[0036] Get the analysis period;

[0037] Get the collection quantity;

[0038] Obtain all security data within the analysis period to form a security collection set;

[0039] Get the number of elements in the guaranteed collection set and set it as the analysis number;

[0040] If the analysis quantity = the collection quantity, the target customer is determined to meet the third stable data;

[0041] If the analysis quantity ≠ the collection quantity, it is determined that the target customer does not meet the third stable data;

[0042] Obtain the unit information of each element in the collection set and define it as the analysis unit;

[0043] If all analysis units in the collection set are guaranteed to be the same unit, the target customer is determined to meet the fourth stable data;

[0044] If all analysis units in the guaranteed collection set have different units, it is determined that the target customer does not meet the fourth stable data.

[0045] As an optional solution of the internal and external data security interaction method based on data analysis described in the present invention, the data collection strategy is specifically:

[0046] Obtain personal information, including historical application data;

[0047] Obtain the applicant enterprise of each historical application data;

[0048] If there is an applicant enterprise, obtain the interactive database;

[0049] Extracting the application information of personal information in the interactive database to analyze the application information;

[0050] Get the number of historical application data and set it as the first application number;

[0051] The number of analysis application information is obtained and determined as the second application number;

[0052] If the first application quantity ≥ the second application quantity, the customer is determined to meet the fifth stable data, and all historical application data are obtained to form an application analysis set;

[0053] If the first application quantity is less than the second application quantity, the customer is determined not to meet the fifth stable data, and all analysis application information is obtained to form an application analysis set;

[0054] If the applicant enterprise does not exist, obtain the interactive database;

[0055] Extract the application information of personal information in the interactive database and define it as target application information;

[0056] If target application information exists, the customer is determined not to meet the fifth stable data, and all target application information is obtained to form an application analysis set;

[0057] If the target application information does not exist, it is determined that the customer meets the fifth stable data.

[0058] As an optional solution of the internal and external data security interaction method based on data analysis described in the present invention, the data determination strategy is specifically as follows:

[0059] Get the application analysis collection;

[0060] Identify each element in the application analysis set as the target element in turn;

[0061] Obtain the evaluation data of the target element and set it as the initial evaluation data for analysis;

[0062] The analysis of the preliminary assessment data includes the first risk and the second risk;

[0063] Execute element evaluation strategy on target element;

[0064] Extract secondary evaluation data of target elements;

[0065] If the target element's initial analysis data is the first risk and the target element's secondary analysis data is the first risk, the historical analysis data is determined to be unchanged;

[0066] If the target element's initial analysis data is the first risk but the target element's secondary evaluation data is the second risk, determine the change in historical evaluation data;

[0067] If the target element's initial analysis data is the second risk but the target element's secondary evaluation data is the first risk, determine the change in historical evaluation data;

[0068] If the target element's initial analysis data is the second risk and the target element's secondary evaluation data is the second risk, the historical evaluation data is judged to be unchanged;

[0069] If there are elements of historical evaluation data changes in the application analysis set, the judgment data is judged to be incorrect customer data;

[0070] If all elements in the application analysis set are historical assessment data and remain unchanged, it is determined that the data is customer data and is correct.

[0071] As an optional solution of the internal and external data security interaction method based on data analysis of the present invention, the element evaluation strategy is specifically:

[0072] Get the scheduled repayment time and actual repayment time of the target element;

[0073] Get the scheduled repayment data and actual repayment data of the target element;

[0074] If the actual repayment time ≤ the scheduled repayment time and the actual repayment data = the scheduled repayment data, the customer is considered to be actively repaying;

[0075] If the actual repayment time is greater than the scheduled repayment time or the actual repayment data is less than the scheduled repayment data, the customer analysis strategy is executed;

[0076] Obtain the number of abnormal repayment data that indicates that the customer is not actively repaying the loan, and define this as the number of abnormal repayments;

[0077] Withdraw the predetermined amount;

[0078] If the number of abnormal repayments is ≤ the predetermined number × 15%, the secondary assessment data of the target element is determined to be the first risk;

[0079] If the abnormal repayment amount>predetermined amount×15%, the secondary assessment data of the target element is determined to be the second risk.

[0080] As an optional solution of the internal and external data security interaction method based on data analysis described in the present invention, the customer analysis strategy is specifically:

[0081] Extract the target customer's actual repayment time under the target element > the scheduled repayment time or the actual repayment data < the scheduled repayment data, and define it as abnormal repayment data;

[0082] Get the number of all scheduled repayment times of the target customer under the target element and set it as the scheduled number;

[0083] The number of abnormal repayment data is obtained and defined as the abnormal number;

[0084] If the abnormal amount is ≤ the predetermined amount × 10%, the customer is judged to be not active in repayment;

[0085] If the abnormal amount > the predetermined amount × 10%, then the collection data of the target customer under the abnormal repayment data is obtained;

[0086] If there is collection data, it is determined that the customer is not active in repayment;

[0087] If there is no collection data, the repayment submission data of the target customer under the abnormal repayment data is obtained;

[0088] If the repayment submission data is greater than 1, the repayment submission time of each repayment submission data is obtained;

[0089] If the repayment submission time is ≤ the scheduled repayment time, it is considered as on-time data;

[0090] The number of on-time data is obtained and is defined as the on-time number;

[0091] Get the number of repayment submission data, which is defined as the repayment submission number;

[0092] If the on-time quantity is greater than 0 and the on-time quantity is less than or equal to the repayment submission quantity - 1, the customer is considered to be actively repaying;

[0093] If the on-time quantity = 0, the customer is judged to be not active in repayment;

[0094] If the repayment submission data = 1, it is determined that the customer is not active in repayment.

[0095] As an optional solution of the internal and external data security interaction method based on data analysis described in the present invention, the data correction strategy is specifically:

[0096] Get the number of elements with historical evaluation data changes in the application analysis set, which is defined as the number of changes;

[0097] Get the number of elements in the application analysis set and set it as the number of historical applications;

[0098] If the number of changes is ≥ the number of historical applications × 30%, the element corresponding to the number of changes in the application analysis set is extracted and defined as the changed element;

[0099] The application analysis set with the changed elements removed is identified as the unchanged set;

[0100] Extract the secondary assessment data as the number of changing elements of the first risk, which is defined as the second variable quantity

[0101] Extract the number of elements in the invariant set that are the first risk in the initial assessment data and define it as the initial assessment number;

[0102] If the number of secondary changes + the number of initial evaluations ≥ the number of historical applications × 80%, the customer is judged to meet the repayment stability data;

[0103] If the number of secondary changes + the number of initial evaluations is less than the number of historical applications × 80%, the customer is judged to be ineligible for stable repayment data;

[0104] If the number of changes is less than the number of historical applications × 30%, the number of elements with the first risk in the initial evaluation data of the application analysis set is extracted and determined as the stable number;

[0105] If the stable number ≥ the historical application number × 80%, the customer is judged to meet the repayment stability data;

[0106] If the stable quantity is less than the historical application quantity × 80%, the customer is judged as not meeting the repayment stability data.

[0107] As an optional solution of the internal and external data security interaction method based on data analysis described in the present invention, the data analysis strategy is specifically:

[0108] Setting assessment scores;

[0109] If the customer meets the first stable data, the first score is calculated, the first score = [(assessment score × 60%) × 10%] × 80%;

[0110] If the customer does not meet the first stable data, the first score is calculated, the first score = [(assessment score × 60%) × 10%] × 20%;

[0111] If the customer meets the second stable data, the second score is calculated, the second score = [(assessment score × 60%) × 30%] × 80%;

[0112] If the customer does not meet the second stable data, calculate the second score, the second score = [(assessment score × 60%) × 30%] × 20%;

[0113] If the customer meets the third stable data, the third score is calculated, the third score = [(assessment score × 60%) × 20%] × 80%;

[0114] If the customer does not meet the third stable data, the third score is calculated, the third score = [(assessment score × 60%) × 20%] × 20%;

[0115] If the customer meets the fourth stable data, the fourth score is calculated, the fourth score = [(assessment score × 60%) × 10%] × 80%;

[0116] If the customer does not meet the fourth stable data, the fourth score is calculated, the fourth score = [(assessment score × 60%) × 10%] × 20%;

[0117] If the customer meets the fifth stable data, the fifth score is calculated, the fifth score = [(assessment score × 60%) × 30%] × 80%;

[0118] If the customer does not meet the fifth stable data, the fifth score is calculated, the fifth score = [(assessment score × 60%) × 30%] × 20%;

[0119] If the customer meets the repayment stability data, the sixth score is calculated, the sixth score = (assessment score × 40%) × 80%;

[0120] If the customer does not meet the repayment stability data, the sixth score is calculated, the sixth score = (assessment score × 40%) × 20%;

[0121] Calculate and analyze the data, analysis data = first score + second score + third score + fourth score + fifth score + sixth score.

[0122] The present invention has the following beneficial effects:

[0123] 1. This internal and external data security interaction method based on data analysis automatically forms an evaluation model for the borrower by obtaining the borrower's income information, social security information and loan information, and determines whether the borrower's work and income are stable, thereby determining whether the borrower's repayment is stable, reducing misjudgments or wrong judgments of customers, and preventing customers from being unable to borrow money reasonably due to the improper behavior of others.

[0124] 2. This internal and external data security interaction method based on data analysis obtains the borrower's historical loan records and the data interaction between the enterprise and its partners to determine whether the historical loan records provided by the borrower are consistent with the historical loan records recorded in the interactive database, thereby determining whether the borrower has concealed his or her historical loan records, reducing misjudgments or wrong judgments against customers, and preventing customers from being unable to borrow money reasonably due to the improper behavior of others.

[0125] 3. This internal and external data security interaction method based on data analysis, when the borrower's historical loan record contains the behavior of failing to submit the corresponding repayment in time, by obtaining the borrower's repayment submission record and the borrowing behavior of the borrowing enterprise, it is judged whether the borrower has actively and deliberately defaulted on repayment, thereby reducing misjudgment or wrong judgment of the customer and avoiding the customer's inability to borrow money reasonably. When reviewing the borrower's submitted information, the enterprise can refer to the borrower's borrowing information and post-loan performance of other consumer finance enterprises, so as to conduct a more accurate and comprehensive risk assessment of the borrower, improve the information sensitivity of the enterprise, and reduce the enterprise's property loss. BRIEF DESCRIPTION OF THE DRAWINGS

[0126] Figure 1 The present invention is a flow chart of the method for secure interaction of internal and external data based on data analysis. DETAILED DESCRIPTION

[0127] 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.

[0128] Embodiment 1, a method for secure interaction of internal and external data based on data analysis, see Figure 1 ,include:

[0129] Obtain target submission information from target customers;

[0130] The target submitted information includes loan information and personal information;

[0131] Submit information based on the target and form a preliminary evaluation model;

[0132] Submit information based on the target and form collected data through data collection strategy;

[0133] According to the collected data, the judgment data is formed through the data judgment strategy;

[0134] The judgment data includes whether the customer data is correct or incorrect;

[0135] If the data is judged to be customer data, the collected data will be identified as reference data;

[0136] If the data is judged to be incorrect customer data, a data correction strategy is executed to form reference data;

[0137] Based on the preliminary evaluation model and reference data, analytical data is generated through data analysis strategies;

[0138] Generate evaluation results of target customers based on the analyzed data.

[0139] This embodiment also provides that the implementation process of secure interaction of internal and external data is as follows:

[0140] Internal and external API governance: interface governance for internal and external network communications within the company, governance of internal and external interfaces with different interaction mechanisms and different data standards, and meeting internal and external interface communication needs through customized docking development and data cleaning;

[0141] Data encryption and decryption: In order to ensure data security of network communications, the company's external communication interfaces all have message encryption requirements. Each external communication unit also has different requirements for encryption methods and encryption strategies to facilitate secure interaction of internal and external data.

[0142] Abnormal traffic monitoring and malicious behavior identification: Use big data analysis technology to analyze pre-processed data and build security models and business models;

[0143] Among them, the security model is used to identify potential security threats, including abnormal traffic detection, malicious behavior identification, etc.

[0144] Among them, the business model is used to identify business needs and normal business patterns to ensure that legitimate business requests are processed in a timely manner;

[0145] Network communication strategy formulation: Based on the analysis results of the security model, formulate security strategies for internal and external network communications, including flow control, access control, data encryption, etc.;

[0146] Based on the analysis results of the business model, formulate a business priority strategy to ensure that key business data can be transmitted first.

[0147] Deployment and configuration of intermediate systems: Deploy intermediate systems between internal and external networks as a bridge for data interaction;

[0148] The intermediate system screens and processes data according to the established network communication strategy to ensure that only data that complies with the security strategy can pass;

[0149] Real-time monitoring and dynamic adjustment: real-time monitoring of intranet and extranet communications, collection of new data for analysis, and updating of security and business models;

[0150] Dynamically adjust network communication strategies based on the latest analysis results to ensure that the system is always in the best security state;

[0151] Logging and auditing: Detailed logging of all intranet and extranet communications, including the time, content, source and purpose of data transmission;

[0152] Audit logs regularly to identify and address potential security issues;

[0153] Through the above method, an evaluation model is automatically formed for the borrower based on the borrower's communication data to determine whether its repayment is stable. Based on the borrower's historical borrowing records and data interaction with partners, it is determined whether the borrower has concealed its historical borrowing records. When the borrower's historical borrowing records show that the corresponding repayments have not been submitted in a timely manner, it is determined whether the behavior is caused by the borrower's active and intentional default, thereby reducing misjudgments or wrong judgments on customers and avoiding customers from being unable to borrow money reasonably. The interactive database of interbank credit and post-loan conditions established based on the data of the People's Bank of China Credit Information Center is used for internal and external data interaction, so that the data sources are diversified, thereby making the evaluation data accurate and comprehensive.

[0154] Embodiment 2: This embodiment is an improvement made on the basis of embodiment 1. The internal and external data security interaction method based on data analysis, the formation of the preliminary evaluation model, is specifically:

[0155] Obtaining loan information, the loan information includes loan data and loan duration, the loan data is the total loan amount of the borrower, and the loan duration is the total repayment period promised by the borrower when applying for the loan. For example, if the total repayment period promised by the borrower when applying for the loan is 5 years, the loan duration is 60 months;

[0156] Calculate repayment data, repayment data = loan data ÷ loan duration;

[0157] Obtaining personal information, the personal information including income data and security data, the income data being the monthly income amount submitted by the borrower when applying for a loan, and the security data being the monthly social security payment status submitted by the borrower when applying for a loan;

[0158] Obtain target submission information, wherein the target submission information includes submission time;

[0159] Set the collection period, which is 2 years;

[0160] Calculate the collection time, collection time = submission time - collection duration;

[0161] The analysis period is formed by taking the collection time as the starting time point and the submission time as the ending time point;

[0162] Obtain the target customer's income data during the analysis period to form an income set, that is, obtain the target customer's monthly income within 2 years;

[0163] Sequentially extract two adjacent elements in the income set and define them as the first and last elements respectively;

[0164] If the later income element ≥ the earlier income element × 90%, the target customer is judged to meet the first stable data, that is, the target customer's monthly income is similar during the analysis period, and his work and income are relatively stable, indicating that the repayment is relatively stable;

[0165] If the later income element is less than the earlier income element × 90%, it is determined that the target customer does not meet the first stable data, that is, the target customer has different monthly incomes during the analysis period, and there are unstable factors in his work or income, indicating that the repayment is relatively unstable;

[0166] Get the number of elements in the collection, which is defined as the collection quantity;

[0167] Calculate the average value of the elements, the average value of the elements = the sum of all elements in the collection / the number of collections;

[0168] If the repayment data is ≤ the element average value × 50%, the target customer is judged to meet the second stable data, that is, the target customer can ensure that half of the monthly income can cover the repayment amount during the analysis period, and his income and life are relatively stable, indicating that the repayment is relatively stable;

[0169] If the repayment data is greater than the element average value × 50%, it is determined that the target customer does not meet the second stable data, that is, the target customer cannot guarantee that his monthly income can cover the repayment amount during the analysis period, and there are unstable factors in his income or life, indicating that the repayment is relatively unstable;

[0170] Implement a stable analysis strategy.

[0171] The stability analysis strategy is specifically as follows:

[0172] Obtaining safeguard data;

[0173] Get the analysis period;

[0174] Get the collection quantity;

[0175] Obtain all security data within the analysis period to form a security collection set;

[0176] Get the number of elements in the guaranteed collection set and set it as the analysis number;

[0177] If the number of analysis = the number of collection, then the target customer is determined to meet the third stable data, that is, the target customer pays social security on time every month during the analysis period, and his work and life are relatively stable, indicating that the repayment is relatively stable;

[0178] If the analysis quantity ≠ the collection quantity, it is determined that the target customer does not meet the third stable data, that is, the target customer did not pay social security on time every month during the analysis period, and there may be unstable factors in his work or life, indicating that the repayment is relatively unstable;

[0179] Obtain the unit information of each element in the security collection set and define it as the analysis unit. The unit information is the payment unit of the social security, so as to judge whether the customer has a stable job. If the customer's social security payment unit is not unique during the analysis period, it means that the customer may have an unstable job;

[0180] If all the analysis units in the security collection set are the same unit, then the target customer is judged to meet the fourth stable data, that is, the target customer has only one social security payment unit during the analysis period, which means that the customer's job is relatively stable, so that the customer's repayment is relatively stable;

[0181] If there are different analysis units in all the security collection sets, the target customer is judged as not meeting the fourth stable data, that is, the target customer's social security payment unit during the analysis period is not unique, indicating that the customer may have an unstable job, which may lead to unstable repayments by the customer.

[0182] Embodiment 3: This embodiment is an improvement made on the basis of Embodiment 2. In this embodiment, the data collection strategy is specifically as follows:

[0183] Obtain personal information, including historical application data;

[0184] Obtain the applicant enterprise of each historical application data, where the applicant enterprise is the company that the target customer has applied for a loan;

[0185] If there is an applicant enterprise, the interactive database is obtained. The interactive database is a database for the enterprise to store the relevant loan information of customers obtained through the People's Bank of China Credit Information Center, which is used for internal and external data interaction, so that the enterprise can refer to the borrower's loan information and post-loan performance in financial institutions while reviewing the information submitted by the borrower, so as to conduct a more accurate and comprehensive risk assessment of the borrower;

[0186] Extracting application information of personal information in the interactive database to analyze application information, wherein the application information is the historical application record of the target customer for loans from other consumer finance companies in the interactive database;

[0187] Get the number of historical application data and set it as the first application number;

[0188] The number of analysis application information is obtained and determined as the second application number;

[0189] If the first application quantity ≥ the second application quantity, the customer is determined to meet the fifth stable data, and all historical application data are obtained to form an application analysis set, which means that the customer has a historical loan record, and the customer actively registers the historical loan record that is not in the interactive database, which means that the customer is relatively honest, but it is still necessary to analyze its historical loan situation to determine whether there is a problem with its credit, and the current assessment result cannot be directly used as the basis for borrowing;

[0190] If the first application quantity is less than the second application quantity, the customer is determined to be not in compliance with the fifth stable data, and all analysis application information is obtained to form an application analysis set, which means that the customer has a historical loan record, and the customer may have concealed the historical loan record. It is necessary to further determine whether the concealment of the historical loan record is intentional and whether there is a credit problem. The current assessment result cannot be directly used as the basis for borrowing;

[0191] If the applicant enterprise does not exist, obtain the interactive database;

[0192] Extract the application information of personal information in the interactive database and define it as target application information;

[0193] If the target application information exists, the customer is judged not to meet the fifth stable data, and all target application information is obtained to form an application analysis set, which means that the customer has a historical loan record. It is necessary to further judge whether he has concealed his historical loan record and whether there are any credit problems. The evaluation results cannot be directly used as the basis for borrowing;

[0194] If the target application information does not exist, the customer is judged to meet the fifth stable data, which means that the customer has no historical borrowing record and has not concealed any historical borrowing record, and has a good credit. The result of this assessment can be directly used as the basis for borrowing.

[0195] Among them, in the generation of the interactive database, a data security analysis strategy is implemented to ensure data security as much as possible;

[0196] The data security analysis strategy specifically includes:

[0197] Network policy: Generate an IP whitelist to control the IPs that can communicate with the internal;

[0198] Interface standard: set the format and content of interface messages;

[0199] Communication forwarding: All internal and external communications are forwarded through the open platform. External organizations first interact with the open platform for data. The open platform performs preset security checks and rule verification on the communication content before forwarding it to the internal target system.

[0200] Data cleaning: obtain the target communication interface;

[0201] Setting a determined communication interface, wherein the determined communication interface is a communication interface set according to an internal standard;

[0202] If the target communication interface is consistent with the determined communication interface, the data can be directly transmitted to the target system of the communication;

[0203] If the target communication interface is inconsistent with the determination communication interface, the target communication interface is standardized through the open platform and the data is transmitted to the target system of communication.

[0204] Message encryption: Set secure communication protocols and message encryption to make communication with external interfaces more secure;

[0205] Dynamic real-time monitoring: obtain the communication traffic between all external communication parties and our company, and locate the target communication traffic;

[0206] Set the warning flow value;

[0207] If the target communication browsing volume ≥ the warning flow value, a traffic abnormality warning will be issued;

[0208] If the target communication flow is less than the warning flow value, monitoring will continue.

[0209] Embodiment 4: This embodiment is an improvement made on the basis of Embodiment 3. In this embodiment, the data determination strategy is specifically as follows:

[0210] Get the application analysis collection;

[0211] Identify each element in the application analysis set as the target element in turn;

[0212] Obtain the evaluation data of the target element and set it as the initial evaluation data for analysis;

[0213] The analysis of the preliminary evaluation data includes a first risk and a second risk, wherein the first risk is that the customer has a low repayment risk, and the second risk is that the customer has a high repayment risk;

[0214] Execute element evaluation strategy on target element;

[0215] Extract secondary evaluation data of target elements;

[0216] If the target element's initial analysis data is the first risk and the target element's secondary analysis data is the first risk, the historical analysis data is determined to be unchanged;

[0217] If the target element's initial analysis data is the first risk but the target element's secondary evaluation data is the second risk, determine the change in historical evaluation data;

[0218] If the target element's initial analysis data is the second risk but the target element's secondary evaluation data is the first risk, determine the change in historical evaluation data;

[0219] If the target element's initial analysis data is the second risk and the target element's secondary evaluation data is the second risk, the historical evaluation data is judged to be unchanged;

[0220] If there are elements of historical evaluation data changes in the application analysis set, the judgment data is judged to be incorrect customer data;

[0221] If all elements in the application analysis set are historical assessment data and remain unchanged, it is determined that the data is customer data and is correct.

[0222] The element evaluation strategy is specifically:

[0223] Obtain the scheduled repayment time and actual repayment time of the target element, wherein the scheduled repayment time is the repayment time promised by the borrower when applying for the loan, and the actual repayment time is the actual repayment time of the borrower;

[0224] Obtain the scheduled repayment data and actual repayment data of the target element, wherein the scheduled repayment data is the single repayment amount calculated by the borrower based on the total loan amount and the total repayment period promised by the borrower when applying for a loan, and the actual repayment data is the actual single repayment amount of the borrower;

[0225] If the actual repayment time ≤ the scheduled repayment time and the actual repayment data = the scheduled repayment data, the customer is considered to be actively repaying;

[0226] If the actual repayment time is greater than the scheduled repayment time or the actual repayment data is less than the scheduled repayment data, the customer analysis strategy is executed to determine whether the customer's failure to repay on time or the failure to repay the corresponding amount of the loan is due to deliberate default by the customer or improper behavior by the company from which the customer has borrowed money, such as maliciously preventing the user from submitting the corresponding amount of repayment within the corresponding time, etc.;

[0227] Obtain the number of abnormal repayment data that indicates that the customer is not actively repaying the loan, and define this as the number of abnormal repayments;

[0228] Withdraw the predetermined amount;

[0229] If the number of abnormal repayments is ≤ the predetermined number × 15%, the secondary assessment data of the target element is determined to be the first risk;

[0230] If the abnormal repayment amount>predetermined amount×15%, the secondary assessment data of the target element is determined to be the second risk.

[0231] The customer analysis strategy is specifically as follows:

[0232] Extract the target customer's actual repayment time under the target element > the scheduled repayment time or the actual repayment data < the scheduled repayment data, and define it as abnormal repayment data;

[0233] Get the number of all scheduled repayment times of the target customer under the target element and set it as the scheduled number;

[0234] The number of abnormal repayment data is obtained and defined as the abnormal number;

[0235] If the abnormal amount is ≤ the predetermined amount × 10%, the customer is judged to be not active in repayment;

[0236] If the abnormal amount is greater than the predetermined amount × 10%, the collection data of the target customer under the abnormal repayment data is obtained, and the collection data is the collection notice issued by the company corresponding to the target element to the target customer when the target customer fails to submit the corresponding repayment in time;

[0237] If there is collection data, it is determined that the customer is not active in repayment;

[0238] If there is no collection data, the repayment submission data of the target customer under the abnormal repayment data is obtained, and the repayment submission data is the number of times the target customer submits repayments in the month where the abnormal repayment data is located;

[0239] If the repayment submission data is greater than 1, the repayment submission time of each repayment submission data is obtained;

[0240] If the repayment submission time is ≤ the scheduled repayment time, it is considered as on-time data;

[0241] The number of on-time data is obtained and is defined as the on-time number;

[0242] Get the number of repayment submission data, which is defined as the repayment submission number;

[0243] If the number of on-time payments is greater than 0 and the number of on-time payments is less than or equal to the number of repayment submissions - 1, the customer is considered to be actively repaying, that is, the customer has submitted at least one repayment before the end of the day of the promised repayment date, but it was unsuccessful, indicating that the customer has taken the initiative to repay.

[0244] If the number of on-time payments = 0, the customer is judged to be not active in repayment, that is, the customer has not submitted repayment before the end of the day of the promised repayment date, but the customer has submitted repayment at least twice in that month, indicating that the customer may have delayed repayment;

[0245] If the repayment submission data = 1, it is determined that the customer is not active in repayment, that is, the customer has not submitted the repayment before the end of the day of the repayment date he promised, and the customer has only one repayment behavior in that month, which means that the customer himself may have delayed repayment behavior.

[0246] Embodiment 5: This embodiment is an improvement made on the basis of Embodiment 4. In this embodiment, the data correction strategy is specifically as follows:

[0247] Get the number of elements with historical evaluation data changes in the application analysis set, which is defined as the number of changes;

[0248] Get the number of elements in the application analysis set and set it as the number of historical applications;

[0249] If the number of changes is ≥ the number of historical applications × 30%, the element corresponding to the number of changes in the application analysis set is extracted and defined as the changed element;

[0250] The application analysis set with the changed elements removed is identified as the unchanged set;

[0251] Extract the secondary assessment data as the number of changing elements of the first risk, which is defined as the second variable quantity

[0252] Extract the number of elements in the invariant set that are the first risk in the initial assessment data and define it as the initial assessment number;

[0253] If the number of secondary changes + the number of initial evaluations ≥ the number of historical applications × 80%, the customer is judged to meet the repayment stability data;

[0254] If the number of secondary changes + the number of initial evaluations is less than the number of historical applications × 80%, the customer is judged to be ineligible for stable repayment data;

[0255] If the number of changes is less than the number of historical applications × 30%, the number of elements with the first risk in the initial evaluation data of the application analysis set is extracted and determined as the stable number;

[0256] If the stable number ≥ the historical application number × 80%, the customer is judged to meet the repayment stability data;

[0257] If the stable quantity is less than the historical application quantity × 80%, the customer is judged as not meeting the repayment stability data.

[0258] This embodiment also provides that the data analysis strategy is specifically:

[0259] Setting an evaluation score, wherein the evaluation score is 100;

[0260] If the customer meets the first stable data, the first score is calculated, the first score = [(assessment score × 60%) × 10%] × 80%;

[0261] If the customer does not meet the first stable data, the first score is calculated, the first score = [(assessment score × 60%) × 10%] × 20%;

[0262] If the customer meets the second stable data, the second score is calculated, the second score = [(assessment score × 60%) × 30%] × 80%;

[0263] If the customer does not meet the second stable data, calculate the second score, the second score = [(assessment score × 60%) × 30%] × 20%;

[0264] If the customer meets the third stable data, the third score is calculated, the third score = [(assessment score × 60%) × 20%] × 80%;

[0265] If the customer does not meet the third stable data, the third score is calculated, the third score = [(assessment score × 60%) × 20%] × 20%;

[0266] If the customer meets the fourth stable data, the fourth score is calculated, the fourth score = [(assessment score × 60%) × 10%] × 80%;

[0267] If the customer does not meet the fourth stable data, the fourth score is calculated, the fourth score = [(assessment score × 60%) × 10%] × 20%;

[0268] If the customer meets the fifth stable data, the fifth score is calculated, the fifth score = [(assessment score × 60%) × 30%] × 80%;

[0269] If the customer does not meet the fifth stable data, the fifth score is calculated, the fifth score = [(assessment score × 60%) × 30%] × 20%;

[0270] If the customer meets the repayment stability data, the sixth score is calculated, the sixth score = (assessment score × 40%) × 80%;

[0271] If the customer does not meet the repayment stability data, the sixth score is calculated, the sixth score = (assessment score × 40%) × 20%;

[0272] Calculate the analysis data, analysis data = first score + second score + third score + fourth score + fifth score + sixth score, where the larger the value of the analysis data, the lower the customer's repayment assessment risk, and the smaller the value of the analysis data, the higher the customer's repayment assessment risk.

[0273] In this embodiment, an evaluation model is automatically formed for the borrower based on the borrower's historical borrowing information to determine whether the repayment is stable. Based on the borrower's historical borrowing record and the data interaction with the partner, it is determined whether the borrower has concealed the historical borrowing record. When the borrower's historical borrowing record contains the behavior of not submitting the corresponding repayment in time, it is determined whether the behavior is caused by the borrower's active and intentional default, thereby reducing misjudgment or wrong judgment of the customer and avoiding the customer's inability to borrow money reasonably. The interactive database of interbank credit and post-loan status established based on the data of the People's Bank of China Credit Information Center is used for internal and external data interaction, so as to diversify the data sources and make the evaluation data accurate and comprehensive.

Claims

1. A method for secure interaction of internal and external data based on data analysis, characterized in that: include: Obtain target submission information from target customers; The target submitted information includes loan information and personal information; Submit information based on the target and form a preliminary evaluation model; Submit information based on the target and form collected data through data collection strategy; According to the collected data, the judgment data is formed through the data judgment strategy; The judgment data includes whether the customer data is correct or incorrect; If the data is judged to be customer data, the collected data will be identified as reference data; If the data is judged to be incorrect customer data, a data correction strategy is executed to form reference data; Based on the preliminary evaluation model and reference data, analytical data is generated through data analysis strategies; Generate evaluation results of target customers based on the analyzed data.

2. The method for secure interaction of internal and external data based on data analysis according to claim 1, characterized in that: The formation of the preliminary evaluation model is specifically as follows: Obtaining loan information, wherein the loan information includes loan data and loan duration; Calculate repayment data, repayment data = loan data ÷ loan duration; Access personal information, including income and security data; Obtain target submission information, wherein the target submission information includes submission time; Set the collection duration; Calculate the collection time, collection time = submission time - collection duration; The analysis period is formed by taking the collection time as the starting time point and the submission time as the ending time point; Obtain the target customer's income data during the analysis period to form an income set; Sequentially extract two adjacent elements in the income set and define them as the first and last elements respectively; If the latter element is ≥ the former element × 90%, the target customer is determined to meet the first stable data; If the latter element is less than the former element × 90%, it is determined that the target customer does not meet the first stable data; Get the number of elements in the collection, which is defined as the collection quantity; Calculate the average value of the elements, the average value of the elements = the sum of all elements in the collection / the number of collections; If the repayment data is ≤ the element average value × 50%, the target customer is determined to meet the second stable data; If the repayment data > element average value × 50%, it is determined that the target customer does not meet the second stable data; Implement a stable analysis strategy.

3. The method for secure interaction of internal and external data based on data analysis according to claim 2 is characterized in that: The stability analysis strategy is specifically: Obtaining safeguarding data; Get the analysis period; Get the collection quantity; Obtain all security data within the analysis period to form a security collection set; Get the number of elements in the guaranteed collection set and set it as the analysis number; If the analysis quantity = the collection quantity, the target customer is determined to meet the third stable data; If the analysis quantity ≠ the collection quantity, it is determined that the target customer does not meet the third stable data; Obtain the unit information of each element in the collection set and define it as the analysis unit; If all analysis units in the collection set are guaranteed to be the same unit, the target customer is determined to meet the fourth stable data; If all analysis units in the guaranteed collection set have different units, it is determined that the target customer does not meet the fourth stable data.

4. The method for secure interaction of internal and external data based on data analysis according to claim 1 is characterized in that: The data collection strategy is specifically as follows: Obtain personal information, including historical application data; Obtain the applicant enterprise of each historical application data; If there is an applicant enterprise, obtain the interactive database; Extracting the application information of personal information in the interactive database to analyze the application information; Get the number of historical application data and set it as the first application number; The number of analysis application information is obtained and determined as the second application number; If the first application quantity ≥ the second application quantity, the customer is determined to meet the fifth stable data, and all historical application data are obtained to form an application analysis set; If the first application quantity is less than the second application quantity, the customer is determined not to meet the fifth stable data, and all analysis application information is obtained to form an application analysis set; If the applicant enterprise does not exist, obtain the interactive database; Extract the application information of personal information in the interactive database and define it as target application information; If target application information exists, the customer is determined not to meet the fifth stable data, and all target application information is obtained to form an application analysis set; If the target application information does not exist, it is determined that the customer meets the fifth stable data.

5. The method for secure interaction of internal and external data based on data analysis according to claim 1 is characterized in that: The data determination strategy is specifically as follows: Get the application analysis collection; Identify each element in the application analysis set as the target element in turn; Obtain the evaluation data of the target element and set it as the initial evaluation data for analysis; The analysis of the preliminary assessment data includes the first risk and the second risk; Execute element evaluation strategy on target element; Extract secondary evaluation data of target elements; If the target element's initial analysis data is the first risk and the target element's secondary analysis data is the first risk, the historical analysis data is determined to be unchanged; If the target element's initial analysis data is the first risk but the target element's secondary evaluation data is the second risk, determine the change in historical evaluation data; If the target element's initial analysis data is the second risk but the target element's secondary evaluation data is the first risk, determine the change in historical evaluation data; If the target element's initial analysis data is the second risk and the target element's secondary evaluation data is the second risk, the historical evaluation data is judged to be unchanged; If there are elements of historical evaluation data changes in the application analysis set, the judgment data is judged to be incorrect customer data; If all elements in the application analysis set are historical assessment data and remain unchanged, it is determined that the data is customer data and is correct.

6. The method for secure interaction of internal and external data based on data analysis according to claim 5 is characterized in that: The element evaluation strategy is specifically: Get the scheduled repayment time and actual repayment time of the target element; Get the scheduled repayment data and actual repayment data of the target element; If the actual repayment time ≤ the scheduled repayment time and the actual repayment data = the scheduled repayment data, the customer is considered to be actively repaying; If the actual repayment time is greater than the scheduled repayment time or the actual repayment data is less than the scheduled repayment data, the customer analysis strategy is executed; Obtain the number of abnormal repayment data that indicates that the customer is not actively repaying the loan, and define this as the number of abnormal repayments; Withdraw the predetermined amount; If the number of abnormal repayments is ≤ the predetermined number × 15%, the secondary assessment data of the target element is determined to be the first risk; If the abnormal repayment amount>predetermined amount×15%, the secondary assessment data of the target element is determined to be the second risk.

7. The method for secure interaction of internal and external data based on data analysis according to claim 6 is characterized in that: The customer analysis strategy is specifically: Extract the target customer's actual repayment time under the target element > the scheduled repayment time or the actual repayment data < the scheduled repayment data, and define it as abnormal repayment data; Get the number of all scheduled repayment times of the target customer under the target element and set it as the scheduled number; The number of abnormal repayment data is obtained and defined as the abnormal number; If the abnormal amount is ≤ the predetermined amount × 10%, the customer is judged to be not active in repayment; If the abnormal amount > the predetermined amount × 10%, then the collection data of the target customer under the abnormal repayment data is obtained; If there is collection data, it is determined that the customer is not active in repayment; If there is no collection data, the repayment submission data of the target customer under the abnormal repayment data is obtained; If the repayment submission data is greater than 1, the repayment submission time of each repayment submission data is obtained; If the repayment submission time is ≤ the scheduled repayment time, it is considered as on-time data; The number of on-time data is obtained and is defined as the on-time number; Get the number of repayment submission data, which is defined as the repayment submission number; If the on-time quantity is greater than 0 and the on-time quantity is less than or equal to the repayment submission quantity - 1, the customer is considered to be actively repaying; If the on-time quantity = 0, the customer is judged to be not active in repayment; If the repayment submission data = 1, it is determined that the customer is not active in repayment.

8. The method for secure interaction of internal and external data based on data analysis according to claim 1, characterized in that: The data correction strategy is specifically as follows: Get the number of elements with historical evaluation data changes in the application analysis set, which is defined as the number of changes; Get the number of elements in the application analysis set and set it as the number of historical applications; If the number of changes is ≥ the number of historical applications × 30%, the element corresponding to the number of changes in the application analysis set is extracted and defined as the changed element; The application analysis set with the changed elements removed is identified as the unchanged set; Extract the secondary assessment data as the number of changing elements of the first risk, which is defined as the second variable quantity Extract the number of elements in the invariant set that are the first risk in the initial assessment data and define it as the initial assessment number; If the number of secondary changes + the number of initial evaluations ≥ the number of historical applications × 80%, the customer is judged to meet the repayment stability data; If the number of secondary changes + the number of initial evaluations is less than the number of historical applications × 80%, the customer is judged to be ineligible for stable repayment data; If the number of changes is less than the number of historical applications × 30%, the number of elements with the first risk in the initial evaluation data of the application analysis set is extracted and determined as the stable number; If the stable number ≥ the historical application number × 80%, the customer is judged to meet the repayment stability data; If the stable quantity is less than the historical application quantity × 80%, the customer is judged as not meeting the repayment stability data.

9. The method for secure interaction of internal and external data based on data analysis according to claim 1, characterized in that: The data analysis strategy is specifically as follows: Setting assessment scores; If the customer meets the first stable data, the first score is calculated, the first score = [(assessment score × 60%) × 10%] × 80%; If the customer does not meet the first stable data, the first score is calculated, the first score = [(assessment score × 60%) × 10%] × 20%; If the customer meets the second stable data, the second score is calculated, the second score = [(assessment score × 60%) × 30%] × 80%; If the customer does not meet the second stable data, calculate the second score, the second score = [(assessment score × 60%) × 30%] × 20%; If the customer meets the third stable data, the third score is calculated, the third score = [(assessment score × 60%) × 20%] × 80%; If the customer does not meet the third stable data, the third score is calculated, the third score = [(assessment score × 60%) × 20%] × 20%; If the customer meets the fourth stable data, the fourth score is calculated, the fourth score = [(assessment score × 60%) × 10%] × 80%; If the customer does not meet the fourth stable data, the fourth score is calculated, the fourth score = [(assessment score × 60%) × 10%] × 20%; If the customer meets the fifth stable data, the fifth score is calculated, the fifth score = [(assessment score × 60%) × 30%] × 80%; If the customer does not meet the fifth stable data, the fifth score is calculated, the fifth score = [(assessment score × 60%) × 30%] × 20%; If the customer meets the repayment stability data, the sixth score is calculated, the sixth score = (assessment score × 40%) × 80%; If the customer does not meet the repayment stability data, the sixth score is calculated, the sixth score = (assessment score × 40%) × 20%; Calculate and analyze the data, analysis data = first score + second score + third score + fourth score + fifth score + sixth score.

Images