Automatic risk portrait generation method

Abstract

The invention relates to the technical field of risk portrait generation, and discloses an automatic risk portrait generation method, which comprises the steps of data aggregation, data processing, label generation, label analysis and risk portrait generation, and is characterized in that the repayment positive degree of a customer is judged according to historical data provided by the customer, and the repayment positive degree of the customer is judged according to the historical data of the customer. According to the method, the credit degree of the customer with delayed repayment is judged, whether the customer with delayed repayment knows the passive repayment service or not is judged, misjudgment and misjudgment are reduced, the situation that an inaccurate risk assessment result is generated for the customer due to the fact that the customer has the unknown passive repayment service is avoided, and the risk assessment efficiency is improved. And therefore, certain economic loss is caused between the customer and the company, the experience feeling of the customer for handling business is improved, and the contradiction between the customer and the company caused by undefined business is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of risk profile generation, and in particular to an automated risk profile generation method. Background Art

[0002] Automated risk profile generation methods are usually used to assess and predict various risks. Risk profiles are generated by comprehensively analyzing multiple data sources and indicators, including: data collection, data preprocessing, data analysis, risk profile generation, result interpretation and application, and feedback and optimization. Advanced data analysis technology and machine learning algorithms are combined with actual application scenarios to automatically generate useful risk information.

[0003] Collect relevant data from various data sources, including structured data, unstructured data, and semi-structured data, handle missing values, outliers, and erroneous data, ensure data accuracy and consistency, integrate data from different sources, build a unified data set, extract and select features that are important for risk assessment, use statistical methods to analyze basic features such as data distribution and correlation, apply machine learning algorithms to predict and identify potential risks, establish risk prediction models, use training data to adjust model parameters, optimize prediction accuracy, convert the risk level or risk score predicted by the model into a visual risk portrait, classify risks into different categories, and mark different categories of risk areas in the portrait, regularly update risk portraits to reflect the latest data and model prediction results, provide explanations for risk portraits to help users understand the sources and possible impacts of risks, apply risk portraits to actual decision-making, evaluate the performance of the model through actual results, identify and improve the shortcomings of the model, collect user feedback on risk portraits, and adjust and optimize the generation method to improve accuracy and user satisfaction;

[0004] The existing automated risk profile generation method cannot judge the customer's repayment enthusiasm based on the historical data provided by the customer, cannot judge the creditworthiness of customers with delayed repayments based on the customer's historical data, and cannot judge whether customers with delayed repayments are aware of the passive repayment business, which can easily lead to misjudgment and inaccurate risk assessment results for customers due to the customer's uninformed passive repayment business, which in turn causes certain economic losses to both the customer and the company. Its practicality has certain limitations. Summary of the invention

[0005] The present invention provides an automated risk profile generation method for facilitating the resolution of the problems raised in the background technology.

[0006] The present invention provides the following technical solution: an automated risk profile generation method, comprising:

[0007] Step S1: Data aggregation, collecting basic information provided by target customers in various business systems within the company and credit reports authorized for inquiry;

[0008] Obtain internal and external model scores;

[0009] Collect credit information of target customers authorized to query from third parties;

[0010] Collect historical transaction information and historical repayment information of target customers in target companies;

[0011] Build data warehouses and data marts and model data;

[0012] Step S2: Data processing: pre-processing the collected data, performing data cleaning, format unification and deduplication processing, extracting key features through feature engineering, and converting them into numerical representations;

[0013] Step S3: label generation, generating a number of customer labels based on the preprocessed data, and segmenting the customer labels;

[0014] Step S4: label analysis, forming analysis data through data analysis strategy;

[0015] Step S5: Risk profile generation, integrating various customer labels and analysis data to generate a customer risk profile.

[0016] As an optional solution of the automated risk profile generation method of the present invention, the data analysis strategy includes a user habit analysis strategy, specifically:

[0017] Obtain historical business data of target customers in target companies;

[0018] The historical business data includes first historical business data and second historical business data;

[0019] The total business amount of all first historical business data is defined as the first total amount;

[0020] The total business amount of all the second historical business data is defined as the second total amount;

[0021] If the target customer does not have the first historical business data, the other party analysis strategy is executed;

[0022] If the target customer has the first historical business data, the target customer is identified as a historical customer, and the first total amount of the target customer in the target company is obtained;

[0023] Get the second total amount of the target customer in the target company;

[0024] If the target customer’s first total amount in the target company = the target customer’s second total amount in the target company, then the target customer is determined to be the first customer;

[0025] If the first total amount of the target customer in the target company ≠ the second total amount of the target customer in the target company, the target customer is determined to be the second customer.

[0026] As an optional solution of the automated risk profile generation method of the present invention, the other-party analysis strategy is specifically:

[0027] Obtain all business companies of target customers and define them as analysis companies;

[0028] If the target customer does not exist in the analysis company, the target customer will be identified as a customer with no history;

[0029] If the target customer exists in an analysis company, the target customer is identified as a historical customer, and the first total amount of the target customer in each analysis company is obtained respectively;

[0030] Get the second total amount of target customers in each analyzed company respectively;

[0031] If the target customer's first total amount in each analysis company = the target customer's second total amount in each analysis company, then it is determined to be the third customer;

[0032] If the first total amount of the target customer in each analyzed company ≠ the second total amount of the target customer in each analyzed company, the target customer is determined to be the fourth customer.

[0033] As an optional solution of the method for generating an automated risk profile according to the present invention, the user habit analysis strategy further includes:

[0034] Get all the first historical business data of historical customers;

[0035] Obtain the business-limited time of each first historical business data of the historical customer, and set it as the first business time;

[0036] Obtain the business end time of each second historical business data of the historical customer, and set it as the second business time;

[0037] Set the extension period;

[0038] If the second business time is greater than the first business time + the extension time, the customer is determined to be the fifth customer and the customer analysis strategy is executed;

[0039] If the second business time ≤ the first business time + extension time, the customer is determined to be the sixth customer.

[0040] As an optional solution of the method for generating an automated risk profile according to the present invention, the customer analysis strategy is specifically:

[0041] Acquire historical business data of historical customers, wherein the historical business data includes operation methods;

[0042] The operation mode includes active operation and passive operation;

[0043] Get all the second historical business data of historical customers;

[0044] Obtain the operation method corresponding to each second historical business data of the historical customer;

[0045] If the operation modes corresponding to all the second historical business data are passive operations, the operation analysis strategy is executed;

[0046] If the operation mode corresponding to the second historical business data is an active operation, the customer with history is identified as a delayed customer.

[0047] As an optional solution of the automated risk profile generation method of the present invention, the operation analysis strategy is specifically:

[0048] Obtain all operation information of historical customers;

[0049] Get all the second historical business data of historical customers;

[0050] Obtaining operation information corresponding to each second historical business data of the historical customer;

[0051] If the operation information corresponding to all the second historical business data is consistent, executing the first information analysis strategy;

[0052] If the operation information corresponding to the second historical business data is inconsistent with other operation information, the second information analysis strategy is executed.

[0053] As an optional solution of the method for generating an automated risk profile according to the present invention, the first information analysis strategy is specifically:

[0054] Get the prompt operation corresponding to the operation information;

[0055] Get all the second historical business data of historical customers;

[0056] Obtain the prompt operation corresponding to each second historical business data of the historical customer;

[0057] If there is second historical business data with a corresponding prompt operation, the customer with history is identified as a delayed customer;

[0058] If there is no corresponding prompt operation for all the second historical business data, then all the second historical business data are identified as target analysis data in sequence;

[0059] Obtain the signing information of target analysis data;

[0060] Obtain the browsing time of the contract information by the historical customers and set it as the judgment time;

[0061] Set the analysis duration;

[0062] If the determination time is less than the analysis time, the target analysis data is determined to be unknown data;

[0063] If the determination time is ≥ the analysis time, the target analysis data is determined to be known data;

[0064] If all the second historical business data are unknown data, the historical customer is identified as an unknown customer.

[0065] As an optional solution of the method for generating an automated risk profile according to the present invention, the second information analysis strategy is specifically:

[0066] Obtaining operation information corresponding to each second historical business data of the historical customer;

[0067] Get the number of each operation information, which is defined as the number of operations;

[0068] The operation information with operation number greater than 1 is defined as majority information;

[0069] Define the operation information with the operation number of 1 as minority information;

[0070] Execute the first information analysis strategy for all second historical business data corresponding to the majority information;

[0071] Obtaining second historical business data corresponding to the minority information, and defining it as the minority business data;

[0072] Get prompt operations corresponding to a small amount of business data;

[0073] If a small number of business data contain prompt operations, the customer with history will be identified as a delayed customer;

[0074] If there is no prompt operation for the minority business data, then the signing information of the minority business data is obtained;

[0075] Obtain the browsing time of the contract information by the historical customers and set it as the judgment time;

[0076] Set the analysis duration;

[0077] If the determination time is less than the analysis time, the small amount of business data is determined to be unknown data;

[0078] If the determination time is ≥ the analysis time, the small amount of business data is determined to be known data.

[0079] As an optional solution of the automated risk profile generation method of the present invention, the data analysis strategy also includes a scoring strategy, specifically:

[0080] Set target scores;

[0081] If the target customer is the first customer, calculate the first score, first score = target score × 35%;

[0082] If the target customer is the second customer, calculate the first score, first score = target score × 10%;

[0083] If the target customer is the third customer, calculate the second score, the second score = target score × 25%;

[0084] If the target customer is the fourth customer, calculate the second score, the second score = target score × 10%;

[0085] If the target customer is the fifth customer and the target customer is an unknown customer, calculate the third score, the third score = target score × 35%;

[0086] If the target customer is the fifth customer and the target customer is a delayed customer, the quantities of unknown data and known data are obtained and determined as the unknown quantity and known quantity respectively;

[0087] If unknown quantity ÷ (unknown quantity + known quantity) ≥ 60%, calculate the third score, the third score = target score × 25%;

[0088] If unknown quantity ÷ (unknown quantity + known quantity) < 60%, calculate the third score, the third score = target score × 15%;

[0089] If the target customer is the sixth customer, calculate the third score, the third score = target score × 40%;

[0090] If the target customer is a customer with no history, calculate the target score, target score = target score;

[0091] If the target customer is a historical customer, calculate the target score, target score = first score + second score + third score.

[0092] The present invention has the following beneficial effects:

[0093] 1. This automated risk profile generation method obtains the customer's historical data to determine whether the customer has corresponding borrowing business in the company and other companies, whether the customer has timely repayment behavior for historical borrowing business, and whether the customer has delayed repayment, so as to determine the customer's enthusiasm for repayment and prevent the customer from repaying passively, causing certain economic losses to the company.

[0094] 2. The automated risk profile generation method obtains the customer's historical data on delayed repayments to determine whether the customer has actively repaid the corresponding loan businesses for delayed repayments. If all the corresponding loan businesses for delayed repayments are actively repaid, it means that the customer fully understands his or her own loan business, and then determines that the customer may have passive repayment behavior. If there are loan businesses corresponding to delayed repayments that are passive repayments, it means that the customer may have loan businesses that he or she does not understand, resulting in delayed repayment behavior. Therefore, it is necessary to further confirm the customer's enthusiasm for repayment, reduce misjudgments, and avoid inaccurate risk assessment results for customers due to uninformed passive repayment businesses of customers, which in turn causes certain economic losses to both the customer and the company, improve the customer's business experience, and reduce conflicts between the two parties due to unclear business.

[0095] 3. This automated risk profile generation method, in real life, due to the complexity of software and web pages and the customer's lack of understanding of electronic products, the customer may not pay attention to the payment interface when browsing the page or shopping, which may lead to the customer signing a loan business that he is not aware of or does not understand. Therefore, by obtaining the customer's repayment operation for each loan business, automatic deduction binding information and signing information when the business is signed, it is judged whether the customer is aware of the passive repayment business, and then the customer's repayment enthusiasm for the loan business is determined, reducing misjudgments and misjudgments, avoiding the generation of inaccurate risk assessment results for customers due to the existence of passive repayment businesses that the customer is not aware of, which in turn causes certain economic losses to both the customer and the company, improves the customer's business experience, and reduces conflicts between the two parties due to unclear business. BRIEF DESCRIPTION OF THE DRAWINGS

[0096] Figure 1 This is a block diagram of the automated risk profile generation method of the present invention. DETAILED DESCRIPTION

[0097] 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.

[0098] Example 1, a method for automatically generating risk profiles, see Figure 1 ,include:

[0099] Step S1: Data aggregation, collecting basic information provided by target customers in various business systems within the company and credit reports authorized for inquiry;

[0100] Obtain internal and external model scores;

[0101] Collect the target customer's credit information in a third party authorized for inquiry, where the third party's credit information is the target customer's credit information in other companies other than the target company;

[0102] Collect historical transaction information and historical repayment information of target customers in target companies;

[0103] Build data warehouses and data marts and model data;

[0104] Step S2: Data processing: pre-processing the collected data, performing data cleaning, format unification and deduplication processing, extracting key features through feature engineering, and converting them into numerical representations to facilitate model application and calculation;

[0105] Step S3: tag generation, generating a number of customer tags based on the preprocessed data, the customer tags including but not limited to credit score, consumption habits, repayment behavior, income level, etc., and segmenting the customer tags to more accurately describe customer characteristics;

[0106] Step S4: Tag analysis: Through data analysis strategies, analytical data is generated, and cluster analysis and association analysis are performed on the generated customer tags to identify customer groups with similar risk characteristics. Intelligent algorithms are then used to analyze customer tags to generate risk scores for each customer.

[0107] Step S5: Risk profile generation: integrating various customer labels and analysis data to generate a customer risk profile, including risk level, risk feature description and corresponding risk management recommendations.

[0108] Through the above method, the customer's repayment enthusiasm can be judged according to the historical data provided by the customer. The creditworthiness of the customer with delayed repayment can be judged according to the customer's historical data. It can also be judged whether the customer with delayed repayment is aware of the passive repayment business, thereby reducing misjudgments and misjudgments, avoiding inaccurate risk assessment results for customers due to passive repayment business that the customer is not aware of, which in turn causes certain economic losses to both the customer and the company, improving the customer's business experience, and reducing conflicts between the two parties due to unclear business.

[0109] Embodiment 2: This embodiment is an improvement made on the basis of embodiment 1. The automated risk profile generation method, the data analysis strategy, includes a user habit analysis strategy, specifically:

[0110] Obtain historical business data of target customers in target companies;

[0111] The historical business data includes first historical business data and second historical business data, wherein the first historical business data is the historical borrowing record, and the second historical business data is the historical repayment record;

[0112] The total business amount of all first historical business data is defined as the first total amount;

[0113] The total business amount of all the second historical business data is defined as the second total amount;

[0114] If the target customer does not have the first historical business data, the other party analysis strategy is executed;

[0115] If the target customer has the first historical business data, the target customer is identified as a historical customer, and the first total amount of the target customer in the target company is obtained;

[0116] Get the second total amount of the target customer in the target company;

[0117] If the target customer’s first total amount in the target company = the target customer’s second total amount in the target company, then the target customer is determined to be the first customer;

[0118] If the first total amount of the target customer in the target company ≠ the second total amount of the target customer in the target company, the target customer is determined to be the second customer.

[0119] The other party analysis strategy is specifically:

[0120] Obtain all business companies of the target customer and define them as analysis companies, where the business companies are companies that carry out loan and repayment business for the target customer;

[0121] If the target customer does not exist in the analysis company, the target customer will be identified as a customer with no history;

[0122] If the target customer exists in an analysis company, the target customer is identified as a historical customer, and the first total amount of the target customer in each analysis company is obtained respectively;

[0123] Get the second total amount of target customers in each analyzed company respectively;

[0124] If the target customer's first total amount in each analysis company = the target customer's second total amount in each analysis company, then it is determined to be the third customer;

[0125] If the first total amount of the target customer in each analyzed company ≠ the second total amount of the target customer in each analyzed company, the target customer is determined to be the fourth customer.

[0126] Embodiment 3: This embodiment is an improvement made on the basis of Embodiment 2. In this embodiment, the user habit analysis strategy further includes:

[0127] Get all the first historical business data of historical customers;

[0128] Obtaining the business limit time of each first historical business data of the historical customer, and setting it as the first business time, wherein the business limit time is the required repayment time;

[0129] Obtain the business termination time of each second historical business data of the historical customer, and set it as the second business time, wherein the business termination time is the repayment time;

[0130] Set the extension period, which is 3 days;

[0131] If the second business time is greater than the first business time + the extension time, the customer is determined to be the fifth customer and the customer analysis strategy is executed;

[0132] If the second business time ≤ the first business time + extension time, the customer is determined to be the sixth customer.

[0133] The customer analysis strategy is specifically as follows:

[0134] Obtaining historical business data of historical customers, wherein the historical business data includes operation methods, and the operation methods are the repayment behaviors of customers, that is, determining whether the customers actively repay the loans;

[0135] The operation mode includes active operation and passive operation, the active operation is active repayment, and the passive operation is automatic deduction;

[0136] Get all the second historical business data of historical customers;

[0137] Obtain the operation method corresponding to each second historical business data of the historical customer;

[0138] If the operation modes corresponding to all the second historical business data are passive operations, the operation analysis strategy is executed;

[0139] If the operation mode corresponding to the second historical business data is an active operation, the customer with history is identified as a delayed customer.

[0140] The operation analysis strategy is specifically:

[0141] Obtain all operation information of historical customers, where the operation information is information on automatic deductions bound by historical customers to the target company or analysis company;

[0142] Get all the second historical business data of historical customers;

[0143] Obtaining operation information corresponding to each second historical business data of the historical customer;

[0144] If the operation information corresponding to all the second historical business data is consistent, executing the first information analysis strategy;

[0145] If the operation information corresponding to the second historical business data is inconsistent with other operation information, the second information analysis strategy is executed.

[0146] This embodiment also provides that the first information analysis strategy is specifically:

[0147] Obtaining a prompt operation corresponding to the operation information, wherein the prompt operation is an operation of reminding the target customer to repay the money by phone or text message before automatic deduction;

[0148] Get all the second historical business data of historical customers;

[0149] Obtain the prompt operation corresponding to each second historical business data of the historical customer;

[0150] If there is second historical business data with a corresponding prompt operation, the customer with history is identified as a delayed customer;

[0151] If there is no corresponding prompt operation for all the second historical business data, then all the second historical business data are identified as target analysis data in sequence;

[0152] Acquire the signing information of the target analysis data, where the signing information is the contract signed by the target customer during the loan business;

[0153] Obtain the browsing time of the contract information by the historical customers and set it as the judgment time;

[0154] Set the analysis time, which is 2 minutes;

[0155] If the determination time is less than the analysis time, the target analysis data is determined to be unknown data;

[0156] If the determination time is greater than or equal to the analysis time, the target analysis data is determined to be known data;

[0157] If all the second historical business data are unknown data, the historical customer is identified as an unknown customer.

[0158] This embodiment also provides that the second information analysis strategy is specifically:

[0159] Obtaining operation information corresponding to each second historical business data of the historical customer;

[0160] Get the number of each operation information, which is defined as the number of operations;

[0161] The operation information with operation number greater than 1 is defined as majority information;

[0162] Define the operation information with the operation number of 1 as minority information;

[0163] Execute the first information analysis strategy for all second historical business data corresponding to the majority information;

[0164] Obtaining second historical business data corresponding to the minority information, and defining it as the minority business data;

[0165] Get prompt operations corresponding to a small amount of business data;

[0166] If a small number of business data contain prompt operations, the historical customer will be identified as a delayed customer;

[0167] If there is no prompt operation for the minority business data, then the signing information of the minority business data is obtained;

[0168] Obtain the browsing time of the contract information by the historical customers and set it as the judgment time;

[0169] Set the analysis time, which is 2 minutes;

[0170] If the determination time is less than the analysis time, the small amount of business data is determined to be unknown data;

[0171] If the determination time ≥ analysis time, the small amount of business data is determined to be known data.

[0172] Embodiment 4: This embodiment is an improvement made on the basis of Embodiment 3. In this embodiment, the data analysis strategy also includes a scoring strategy, which is specifically:

[0173] Setting a target score, wherein the target score is 100;

[0174] If the target customer is the first customer, calculate the first score, first score = target score × 35%;

[0175] If the target customer is the second customer, calculate the first score, first score = target score × 10%;

[0176] If the target customer is the third customer, calculate the second score, the second score = target score × 25%;

[0177] If the target customer is the fourth customer, calculate the second score, the second score = target score × 10%;

[0178] If the target customer is the fifth customer and the target customer is an unknown customer, calculate the third score, the third score = target score × 35%;

[0179] If the target customer is the fifth customer and the target customer is a delayed customer, the quantities of unknown data and known data are obtained and determined as the unknown quantity and known quantity respectively;

[0180] If unknown quantity ÷ (unknown quantity + known quantity) ≥ 60%, calculate the third score, the third score = target score × 25%;

[0181] If unknown quantity ÷ (unknown quantity + known quantity) < 60%, calculate the third score, the third score = target score × 15%;

[0182] If the target customer is the sixth customer, calculate the third score, the third score = target score × 40%;

[0183] If the target customer is a customer with no history, calculate the target score, target score = target score;

[0184] If the target customer is a historical customer, calculate the target score, target score = first score + second score + third score.

[0185] In this embodiment, the customer's repayment enthusiasm is judged based on the historical data provided by the customer. The credit rating of the customer with delayed repayment is judged based on the customer's historical data, and whether the customer with delayed repayment is aware of the passive repayment business is judged, thereby reducing misjudgments and misjudgments, avoiding inaccurate risk assessment results for customers due to passive repayment business that the customer is not aware of, which in turn causes certain economic losses to both the customer and the company, improves the customer's experience in handling business, and reduces conflicts between the two parties due to unclear business.

[0186] The embodiments of the present disclosure also provide a non-transitory computer-readable storage medium, which stores computer instructions, and the computer instructions are used to enable the computer to execute the automated risk profile generation method in the aforementioned method embodiment.

[0187] It should be noted that more specific examples of computer-readable storage media may include portable computer disks, hard disks, erasable programmable read-only memories (EPROMs or flash memories), optical storage devices, magnetic storage devices, or any suitable combination thereof. In the present disclosure, computer-readable storage media may be any tangible medium that contains or stores a program, which may be used by or in conjunction with an instruction execution system, apparatus, or device, and the computer-readable medium may be contained in the electronic device; or it may exist independently without being assembled into the electronic device.

[0188] The computer-readable medium carries one or more programs. When the one or more programs are executed by the electronic device, the electronic device can implement the solution provided by the method embodiment.

[0189] Computer program code for performing operations of the present disclosure may be written in one or more programming languages ​​or a combination thereof, and may be executed entirely on a user's computer, partially on a user's computer, as a stand-alone software package, partially on a user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., through the Internet using an Internet service provider).

[0190] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0191] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the technical principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. An automated risk profile generation method, characterized by: include: Step S1: Data aggregation, collecting basic information provided by target customers in various business systems within the company and credit reports authorized for inquiry; Obtain internal and external model scores; Collect credit information of target customers authorized to query from third parties; Collect historical transaction information and historical repayment information of target customers in target companies; Build data warehouses and data marts and model data; Step S2: Data processing: pre-processing the collected data, performing data cleaning, format unification and deduplication processing, extracting key features through feature engineering, and converting them into numerical representations; Step S3: label generation, generating a number of customer labels based on the preprocessed data, and segmenting the customer labels; Step S4: label analysis, forming analysis data through data analysis strategy; Step S5: Risk profile generation, integrating various customer labels and analysis data to generate a customer risk profile.

2. The automated risk profile generation method according to claim 1, characterized in that: The data analysis strategy includes a user habit analysis strategy, specifically: Obtain historical business data of target customers in target companies; The historical business data includes first historical business data and second historical business data; The total business amount of all first historical business data is defined as the first total amount; The total business amount of all the second historical business data is defined as the second total amount; If the target customer does not have the first historical business data, the other party analysis strategy is executed; If the target customer has the first historical business data, the target customer is identified as a historical customer, and the first total amount of the target customer in the target company is obtained; Get the second total amount of the target customer in the target company; If the target customer’s first total amount in the target company = the target customer’s second total amount in the target company, then the target customer is determined to be the first customer; If the first total amount of the target customer in the target company ≠ the second total amount of the target customer in the target company, the target customer is determined to be the second customer.

3. The automated risk profile generation method according to claim 2, characterized in that: The other-party analysis strategy is specifically as follows: Obtain all business companies of target customers and define them as analysis companies; If the target customer does not exist in the analysis company, the target customer will be identified as a customer with no history; If the target customer exists in an analysis company, the target customer is identified as a historical customer, and the first total amount of the target customer in each analysis company is obtained respectively; Get the second total amount of target customers in each analyzed company respectively; If the target customer's first total amount in each analysis company = the target customer's second total amount in each analysis company, then it is determined to be the third customer; If the first total amount of the target customer in each analyzed company ≠ the second total amount of the target customer in each analyzed company, the target customer is determined to be the fourth customer.

4. The automated risk profile generation method according to claim 2, characterized in that: The user habit analysis strategy also includes: Get all the first historical business data of historical customers; Obtain the business-limited time of each first historical business data of the historical customer, and set it as the first business time; Obtain the business end time of each second historical business data of the historical customer, and set it as the second business time; Set the extension period; If the second business time is greater than the first business time + the extension time, the customer is determined to be the fifth customer and the customer analysis strategy is executed; If the second business time ≤ the first business time + extension time, the customer is determined to be the sixth customer.

5. The automated risk profile generation method according to claim 4, characterized in that: The customer analysis strategy is specifically: Acquire historical business data of historical customers, wherein the historical business data includes operation methods; The operation mode includes active operation and passive operation; Get all the second historical business data of historical customers; Obtain the operation method corresponding to each second historical business data of the historical customer; If the operation modes corresponding to all the second historical business data are passive operations, the operation analysis strategy is executed; If the operation mode corresponding to the second historical business data is an active operation, the customer with history is identified as a delayed customer.

6. The automated risk profile generation method according to claim 5, characterized in that: The operation analysis strategy is specifically: Obtain all operation information of historical customers; Get all the second historical business data of historical customers; Obtaining operation information corresponding to each second historical business data of the historical customer; If the operation information corresponding to all the second historical business data is consistent, executing the first information analysis strategy; If the operation information corresponding to the second historical business data is inconsistent with other operation information, the second information analysis strategy is executed.

7. The automated risk profile generation method according to claim 6, characterized in that: The first information analysis strategy is specifically: Get the prompt operation corresponding to the operation information; Get all the second historical business data of historical customers; Obtain the prompt operation corresponding to each second historical business data of the historical customer; If there is second historical business data with a corresponding prompt operation, the customer with history is identified as a delayed customer; If there is no corresponding prompt operation for all the second historical business data, then all the second historical business data are identified as target analysis data in sequence; Obtain the signing information of target analysis data; Obtain the browsing time of the contract information by the historical customers and set it as the judgment time; Set the analysis duration; If the determination time is less than the analysis time, the target analysis data is determined to be unknown data; If the determination time is greater than or equal to the analysis time, the target analysis data is determined to be known data; If all the second historical business data are unknown data, the historical customer is identified as an unknown customer.

8. The automated risk profile generation method according to claim 6, characterized in that: The second information analysis strategy is specifically: Obtaining operation information corresponding to each second historical business data of the historical customer; Get the number of each operation information, which is defined as the number of operations; The operation information with operation number greater than 1 is defined as majority information; Define the operation information with the operation number of 1 as minority information; Execute the first information analysis strategy for all second historical business data corresponding to the majority information; Obtaining second historical business data corresponding to the minority information, and defining it as the minority business data; Get prompt operations corresponding to a small amount of business data; If a small number of business data contain prompt operations, the historical customer will be identified as a delayed customer; If there is no prompt operation for the minority business data, then the signing information of the minority business data is obtained; Obtain the browsing time of the contract information by the historical customers and set it as the judgment time; Set the analysis duration; If the determination time is less than the analysis time, the small amount of business data is determined to be unknown data; If the determination time ≥ analysis time, the small amount of business data is determined to be known data.

9. The automated risk profile generation method according to claim 1, characterized in that: The data analysis strategy also includes a scoring strategy, specifically: Set target scores; If the target customer is the first customer, calculate the first score, first score = target score × 35%; If the target customer is the second customer, calculate the first score, first score = target score × 10%; If the target customer is the third customer, calculate the second score, the second score = target score × 25%; If the target customer is the fourth customer, calculate the second score, the second score = target score × 10%; If the target customer is the fifth customer and the target customer is an unknown customer, calculate the third score, the third score = target score × 35%; If the target customer is the fifth customer and the target customer is a delayed customer, the quantities of unknown data and known data are obtained and determined as the unknown quantity and known quantity respectively; If unknown quantity ÷ (unknown quantity + known quantity) ≥ 60%, calculate the third score, the third score = target score × 25%; If unknown quantity ÷ (unknown quantity + known quantity) < 60%, calculate the third score, the third score = target score × 15%; If the target customer is the sixth customer, calculate the third score, the third score = target score × 40%; If the target customer is a customer with no history, calculate the target score, target score = target score; If the target customer is a historical customer, calculate the target score, target score = first score + second score + third score.

Images