Method for recommending a financial product to a reinsurance client and related device

Abstract

This application, belonging to the fields of artificial intelligence and fintech, is applied to the field of auto insurance renewal risk assessment. It relates to a method and related equipment for recommending financial products to reinsurance clients. The method includes using a random forest tree algorithm to classify and train preprocessed target client attribute information and first business information to obtain potential target clients; collecting corresponding second business information based on the identification information of the potential target clients; using the second business information as risk prediction data, predicting the risk level corresponding to the potential target clients through a preset prediction model; determining whether to offer the first business quote to the potential target clients based on preset judgment conditions and the risk level; and recommending the first business to the potential clients based on the judgment result. This application, by conducting business quote risk assessment for auto insurance reinsurance clients, facilitates business companies in accurately recommending reinsurance-related information to target clients based on sufficient relevant data.

Description

Technical Field

[0001] This application relates to the fields of artificial intelligence and financial technology, and in particular to a method and related equipment for recommending financial products to reinsurance clients. Background Technology

[0002] Due to market competition, there are certain risks involved when business service providers quote prices to clients across all industries. To mitigate these risks, business service providers need to prepare a large amount of attribute information about target clients related to the type of business in advance. This allows them to accurately combine the risk pre-assessment results with business recommendations to the target customer group when quoting prices to those clients.

[0003] Taking auto insurance reinsurance as an example, it is common for customers to frequently change their auto insurance companies. Customers who were once our clients but later lapsed their coverage and switched to other insurance companies are called reinsurance clients. During the lapse period, because these clients have limited historical underwriting and claims data with our company, it is difficult to accurately recommend reinsurance services to them based on sufficient relevant data. Summary of the Invention

[0004] The purpose of this application is to provide a method, apparatus, computer equipment, and storage medium for recommending financial products to reinsurance clients, so that business quoting companies can accurately recommend reinsurance-related information to target clients by combining sufficient relevant data.

[0005] To address the aforementioned technical problems, this application provides a method for recommending financial products to reinsurance clients, employing the following technical solution:

[0006] A method for recommending financial products to reinsurance clients includes the following steps:

[0007] Based on the identification information of the target customers, the attribute information of the target customers is collected from the company's internal platform. The target customers are customers who have a first business history with the company but have not continued the first business in this year.

[0008] Based on the identification information, collect information from the industry sharing platform on the first business information corresponding to the first business continued by the target customer in other companies in the same industry;

[0009] The attribute information and the first business information are preprocessed, and the preprocessed attribute information and the first business information are classified and trained using the random forest tree algorithm to obtain the target potential customers after classification and training.

[0010] Obtain the identification information of the target potential customer, and collect the second business information corresponding to the target potential customer from the industry sharing platform or authoritative platform corresponding to the second business based on the identification information, wherein the second business is a follow-up business of the first business;

[0011] Using the second business information as risk prediction data, a preset prediction model is used to predict the risk level corresponding to the target potential customer.

[0012] Based on preset judgment conditions and the risk level, determine whether to quote the first service to the target potential customer, and recommend the first service to the final potential customer based on the judgment result.

[0013] Furthermore, the step of collecting information from the industry sharing platform, based on the identification information, regarding the first business information corresponding to the target customer continuing the first business with other companies in the same industry, specifically includes:

[0014] A distinguishing status code is pre-set based on whether the first service information corresponding to the first service can be retrieved. The distinguishing status code includes a first status code and a second status code. The first status code indicates that the first service information cannot be retrieved, and the second status code indicates that the first service information can be retrieved.

[0015] Using the identification information as a retrieval field, it is determined whether the first business information corresponding to the first business during the period when the target customer was not continuing the first business with the company can be retrieved from an industry sharing platform or an authoritative platform.

[0016] If the first service information cannot be retrieved, the distinguishing status code corresponding to the first service is set to the first status code;

[0017] If the first service information can be retrieved, then the distinguishing status code corresponding to the first service is set to the second status code.

[0018] Furthermore, the step of preprocessing the attribute information and the first business information specifically includes:

[0019] A data set with the same number of categories as the attribute information is constructed in advance, and a set differentiation identifier is set according to the different categories of the attribute information.

[0020] Taking each individual customer in the customer group as a unit, obtain the number of calls received, total call connection time, number of proactive inquiries, duration of proactive inquiries, and total number of other similar services purchased by the company for each customer in the customer group during the preset period.

[0021] Based on the set differentiation identifier, the number of calls received, total call connection time, number of proactive inquiries, duration of proactive inquiries, and total number of purchases of other similar services of the company during the preset period are cached into their respective corresponding data sets;

[0022] The data within each of the data sets is normalized, mapping the data within each data set to a range of 0 to 1;

[0023] A data set is pre-constructed for caching the distinguishing status codes corresponding to the first service information;

[0024] Obtain the distinctive status code corresponding to the first business information for each customer in the customer group;

[0025] The distinguishing status code is cached in the data set, and the elements in the data set are divided according to the different code values ​​of the distinguishing status code, and the data in the data set is replaced with 0 or 1.

[0026] Furthermore, the step of using a random forest tree algorithm to classify and train the preprocessed attribute information and the first business information to obtain the target potential customers after classification training specifically includes:

[0027] Pre-set intention thresholds for different categories of the attribute information;

[0028] The data sets corresponding to the different categories of the preprocessed attribute information are input into the classification model based on the random forest tree algorithm, and it is determined in turn whether each data in the data set is greater than its corresponding intention threshold.

[0029] If it is greater than, then the intention category corresponding to the current data in the dataset is high intention;

[0030] If it is less than, then the intention category corresponding to the current data in the dataset is low intention;

[0031] Obtain the data set corresponding to the preprocessed first business information;

[0032] Traverse the elements in the dataset. If the current element is 0, the intention category corresponding to the element is low intention. If the current element is 1, the intention category corresponding to the element is high intention.

[0033] Based on the identification information corresponding to each customer in the customer group, the intention category corresponding to each customer in different data sets is obtained, and probability statistics are performed based on a preset probability algorithm to obtain statistical results;

[0034] Based on the statistical results and the preset probability threshold, the business intention category corresponding to each customer is determined, and the business intention category corresponding to each customer is used as the output result of the classification model for output.

[0035] According to the preset filtering rules, target potential customers are selected from the output results, wherein the filtering rules are to select customers with high interest in the business intention category as target potential customers.

[0036] Furthermore, the step of using the second business information as risk prediction data and predicting the risk level corresponding to the target potential customer through a preset prediction model specifically includes:

[0037] Pre-set corresponding risk threshold ranges for different risk levels, and use the risk threshold ranges as configuration parameters for the prediction model;

[0038] The risk prediction data is input into the initialized prediction model, and the risk threshold range corresponding to the risk prediction data is determined by comparison.

[0039] The risk level of the target potential customer is determined based on the risk threshold range corresponding to the risk prediction data.

[0040] Furthermore, in the step of using the second business information as risk prediction data and predicting the risk level corresponding to the target potential customer through a preset prediction model, the method further includes:

[0041] Based on the preset configuration file, identify whether there have been any changes to the number of risk levels.

[0042] If the number of risk levels is changed, the corresponding risk threshold intervals are adjusted using the LightGBM model so that the number of risk threshold intervals is the same as the number of risk levels, thereby completing the fitting and optimization of the configuration parameters of the prediction model.

[0043] Furthermore, the step of determining whether to offer the first service quote to the target potential customer based on preset judgment conditions and the risk level specifically includes:

[0044] Pre-set the price level threshold;

[0045] Based on the quoted price level threshold and the risk level, determine whether to offer the first service quote to the target potential customer;

[0046] If the risk level does not meet the pricing level threshold, then the first service quote will not be made to the target potential customer;

[0047] If the risk level meets the price level threshold, a price request is sent to the preset price platform to obtain the price result for the first service corresponding to the target potential customer.

[0048] Furthermore, the step of issuing a quote request to a preset quote platform and obtaining the quote result for the first service corresponding to the target potential customer if the risk level meets the quote level threshold specifically includes:

[0049] Different risk coefficients are pre-set for different risk levels on the quotation platform;

[0050] Upon receiving the quotation request, the risk level is obtained through parsing, and the corresponding risk coefficient is obtained through the risk level.

[0051] The risk coefficient is substituted into the preset pricing algorithm formula as an algorithm factor to calculate the pricing result. The pricing result is then returned to the pricing requester as the pricing result for the first service corresponding to the target potential customer.

[0052] Compared with the prior art, the embodiments of this application have the following main advantages:

[0053] The method for recommending financial products to reinsurance clients as described in this application involves: collecting target client attribute information from an internal company platform based on the target client's identification information; collecting first business information corresponding to the first business during the period when the target client was not continuing the first business with the company, based on the identification information, from an industry-shared platform or authoritative platform; using a random forest tree algorithm to classify and train the preprocessed target client attribute information and first business information to obtain potential target clients; obtaining the identification information of the potential target clients, and collecting second business information corresponding to the potential target clients from an industry-shared platform or authoritative platform corresponding to the second business based on the identification information; using the second business information as risk prediction data, predicting the risk level corresponding to the potential target clients through a preset prediction model; and determining whether to quote the first business to the potential target clients based on preset judgment conditions and the risk level. This application, by conducting a business quotation risk assessment for auto insurance reinsurance clients, facilitates business quotation companies to accurately recommend reinsurance-related information to target clients by combining sufficient relevant data. Attached Figure Description

[0054] To more clearly illustrate the solutions in this application, the accompanying drawings used in the description of the embodiments of this application will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0055] Figure 1 This is an exemplary system architecture diagram to which this application can be applied;

[0056] Figure 2 A flowchart of an embodiment of a method for recommending financial products to reinsurance clients according to this application;

[0057] Figure 3 yes Figure 2 A flowchart of a specific embodiment of step 202 shown;

[0058] Figure 4 yes Figure 2 A flowchart of a specific embodiment of preprocessing the attribute information and the first business information in step 203 is shown;

[0059] Figure 5 yes Figure 2 A flowchart of a specific embodiment of obtaining the target intention customers after classification training in step 203 is shown;

[0060] Figure 6 yes Figure 2 A flowchart of a specific implementation of step 205 shown;

[0061] Figure 7 yes Figure 2 A flowchart of a specific implementation of step 206 shown;

[0062] Figure 8 yes Figure 7 A flowchart of a specific implementation of step 704 shown;

[0063] Figure 9 A schematic diagram of a structure of an embodiment of an apparatus for recommending financial products to reinsurance clients according to this application;

[0064] Figure 10 A schematic diagram of the structure of an embodiment of the computer device according to this application. Detailed Implementation

[0065] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains; the terminology used herein in the specification of the application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application; the terms "comprising" and "having," and any variations thereof, in the specification, claims, and foregoing drawings of this application, are intended to cover non-exclusive inclusion. The terms "first," "second," etc., in the specification, claims, or foregoing drawings of this application are used to distinguish different objects, not to describe a particular order.

[0066] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0067] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings.

[0068] like Figure 1 As shown, system architecture 100 may include terminal devices 101, 102, and 103, a network 104, and a server 105. Network 104 serves as the medium for providing communication links between terminal devices 101, 102, and 103 and server 105. Network 104 may include various connection types, such as wired or wireless communication links, or fiber optic cables, etc.

[0069] Continue to refer to Figure 2 The diagram illustrates a flowchart of an embodiment of a method for recommending financial products to reinsurance clients according to this application. The method for recommending financial products to reinsurance clients includes the following steps:

[0070] Step 201: Based on the target customer's identification information, collect the target customer's attribute information from the company's internal platform. The attribute information includes: the number of calls received during a preset period, the total call connection duration, the number of proactive inquiries, the duration of proactive inquiries, and the total number of other similar services purchased from the company. The target customer is a group of customers who have a first service history with the company but did not continue the first service this year. The identification information is the target customer's ID card information or the unique customer number corresponding to the target customer. The unique customer number is unique in both the preset industry sharing platform and the company's internal platform.

[0071] In this embodiment, the target customer refers to a group of customers who have a first business history with the company but did not continue the first business this year, rather than a single customer.

[0072] Assuming a risk assessment is conducted on auto insurance reinsurance business, the total number of purchases of other similar businesses from the company can be understood as the total number of non-auto insurance policies purchased from the company. The target customers are those who previously had auto insurance underwritten by the company but did not renew their auto insurance with the company this year, resulting in lapsed insurance coverage. The preset period's number of calls received, total call connection time, number of proactive inquiries, proactive inquiry time, and total number of purchases of other similar businesses from the company are, in other words, the number of calls received, total call connection time, number of proactive inquiries, proactive inquiry time, and total number of purchases of other similar businesses from the company during the lapsed insurance period.

[0073] By acquiring attribute information of lapsed customers from the company's internal platform, including the number of calls made during the lapse period, total call connection time, number of proactive inquiries, duration of proactive inquiries, and the total number of other similar services purchased from the company, this attribute information is used as classification feature information to identify whether the corresponding lapsed customer is a potential target customer. This helps insurance companies screen potential reinsurance customers based on the behavior of lapsed customers, changing the screening criteria from random screening to screening based on customer behavior, making it more scientific.

[0074] Step 202: Based on the identification information, collect the first business information corresponding to the target customer continuing the first business in other companies in the same industry from the industry sharing platform, wherein the industry sharing platform is a platform in which multiple companies in the same industry can share the business information.

[0075] Continue to refer to Figure 3 , Figure 3 for Figure 2 A flowchart of a specific embodiment of step 202 shown includes the following steps:

[0076] Step 301: Set a distinguishing status code in advance based on whether the first service information corresponding to the first service can be retrieved. The distinguishing status code includes a first status code and a second status code. The first status code indicates that the first service information cannot be retrieved, and the second status code indicates that the first service information can be retrieved.

[0077] Step 302: Using the identification information as a retrieval field, determine whether it is possible to retrieve the first business information corresponding to the first business during the period when the target customer was not continuing the first business with the company from an industry sharing platform or authoritative platform;

[0078] Step 303: If the first service information cannot be retrieved, then set the distinguishing status code corresponding to the first service to the first status code;

[0079] Step 304: If the first service information can be retrieved, then set the distinguishing status code corresponding to the first service to the second status code.

[0080] Continuing with the example of auto insurance, the business information corresponding to the target customer's business during the period when the customer was not continuing the business with the company is collected from an industry-shared platform. Specifically, the auto insurance renewal information of the target customer during the period when their insurance coverage was lapsed with the company is collected from the industry-shared platform. This industry-shared platform may also include a platform provided by a national authoritative department for insurance companies to query auto insurance information.

[0081] The business status identifier code indicates whether the business is being continued with another company, and is represented by the distinguishing codes "1" and "0". Taking auto insurance as an example, the business status identifier code is "1" if the business is reinsured with another company during the period of lapse in coverage with the company, and "0" if the business is not reinsured with another company during the period of lapse in coverage with the company.

[0082] By checking whether the target customer had reinsured with other companies in the same industry during the period of lapse in coverage with our company, we set a different status code for them. If they did not reinsurance, it means that they no longer intend to purchase auto insurance, have voluntarily lapsed in coverage, and do not want to purchase auto insurance business again. They are low-intention reinsurance customers and do not need to be pursued. We set a different status code for them. If they reinsured, it means that they are willing to continue to purchase auto insurance. They are high-intention reinsurance customers and can still be pursued. We set a different status code for them. By identifying whether the customer has reinsured with other companies during the period of lapse in coverage, we can help the company identify the customer's intention to renew their insurance.

[0083] Step 203: Preprocess the attribute information and the first business information, and use the random forest tree algorithm to classify and train the preprocessed attribute information and the first business information to obtain the target potential customers after classification and training, wherein the target potential customers are customers who meet the preset intention criteria.

[0084] Continue to refer to Figure 4 , Figure 4 yes Figure 2 The flowchart shown is a specific embodiment of the preprocessing of the attribute information and the first business information in step 203, including the following steps:

[0085] Step 401: Construct a data set in advance, with the number of data sets equal to the number of categories of the attribute information, and set a set differentiation identifier according to the different categories of the attribute information;

[0086] Step 402: Taking an individual customer in the customer group as a unit, obtain the number of calls received, total call connection time, number of proactive inquiries, duration of proactive inquiries, and total number of other similar services purchased by the company for each customer in the customer group during the preset period.

[0087] Step 403: Based on the set differentiation identifier, cache the number of calls received, total call connection time, number of proactive inquiries, proactive inquiry time, and total number of purchases of other similar services of the company during the preset period into their respective corresponding data sets;

[0088] Step 404: Normalize the data in each data set, mapping the data in each data set to a range of 0 to 1;

[0089] Step 405: Pre-construct a data set for caching the distinguishing status codes corresponding to the first service information;

[0090] Step 406: Obtain the distinctive status code corresponding to the first business information for each customer in the customer group;

[0091] Step 407: Cache the distinguishing status code into the data set, and perform binary processing on the elements in the data set according to the different code values ​​of the distinguishing status code, replacing the data in the data set with 0 or 1.

[0092] Taking auto insurance as an example, the target potential customers are those with a high intention to renew their auto insurance policies.

[0093] By normalizing the discrete data corresponding to the attribute information, and dividing the first business information into two categories based on whether the corresponding Boolean value can be obtained, the random forest algorithm classification model can be used to classify and obtain target users.

[0094] Continue to refer to Figure 5 , Figure 5 yes Figure 2 The flowchart shown in step 203 illustrates a specific embodiment of obtaining the target potential customer after classification training, including the following steps:

[0095] Step 501: Pre-set intention thresholds for different categories of the attribute information;

[0096] Step 502: Input the data sets corresponding to the different categories of the preprocessed attribute information into the classification model based on the random forest tree algorithm, and sequentially determine whether each data in the data set is greater than its corresponding intention threshold;

[0097] Step 503: If it is greater than, then the intention category corresponding to the current data in the dataset is high intention;

[0098] Step 504: If it is less than, then the intention category corresponding to the current data in the dataset is low intention;

[0099] Step 505: Obtain the data set corresponding to the preprocessed first business information;

[0100] Step 506: Traverse the elements in the data set. If the current element is 0, the intention category corresponding to the element is low intention. If the current element is 1, the intention category corresponding to the element is high intention.

[0101] Step 507: Based on the identification information corresponding to each customer in the customer group, obtain the intention category corresponding to each customer in different data sets, and perform probability statistics based on a preset probability algorithm to obtain statistical results;

[0102] Step 508: Based on the statistical results and the preset probability threshold, determine the business intention category corresponding to each customer, and output the business intention category corresponding to each customer as the output result of the classification model;

[0103] Step 509: According to the preset filtering rules, select target potential customers from the output results, wherein the filtering rules are to select customers with high interest in the business intention category as target potential customers.

[0104] By employing a random forest tree algorithm to classify and train the preprocessed attribute information and the first business information, the selection of classification features is ensured to avoid being too singular or overly fitted, thus ensuring the scientific nature of the classification.

[0105] Step 204: Obtain the identification information of the target potential customer, and collect the second business information corresponding to the target potential customer from the industry sharing platform or authoritative platform corresponding to the second business based on the identification information, wherein the second business is a follow-up business of the first business.

[0106] Taking auto insurance as an example, the second business information corresponding to the target potential customer is collected from the industry sharing platform or authoritative platform corresponding to the second business. That is, the claim information corresponding to the high-intent auto insurance renewal customer is collected from the industry sharing platform or authoritative platform corresponding to the subsequent claims business of the auto insurance business. At this time, the first business is the auto insurance business, and the second business is the related claims business generated by the prior agreement of the auto insurance business.

[0107] By collecting claims information from highly motivated auto insurance renewal customers, the drawback of these customers having no claims information during their lapse period with the company is compensated for. Combined with claims information from external platforms, risk assessment and pricing are conducted to ensure the reasonableness of the assessment and pricing.

[0108] Step 205: Using the second business information as risk prediction data, predict the risk level corresponding to the target potential customer through a preset prediction model.

[0109] Continue to refer to Figure 6 , Figure 6 yes Figure 2 A flowchart of a specific embodiment of step 205 shown includes the following steps:

[0110] Step 601: Pre-set corresponding risk threshold ranges for different risk levels, and use the risk threshold ranges as configuration parameters for the prediction model;

[0111] Step 602: Input the risk prediction data into the initialized prediction model, and determine the risk threshold range corresponding to the risk prediction data by comparison.

[0112] Step 603: Determine the risk level of the target potential customer based on the risk threshold range corresponding to the risk prediction data.

[0113] In this embodiment, during the step of using the second business information as risk prediction data and predicting the risk level corresponding to the target potential customer through a preset prediction model, the method further includes: identifying whether the number of risk level levels has been changed according to a preset configuration file; if the number of risk level levels has been changed, adjusting the corresponding risk threshold intervals through the LightGBM model so that the number of risk threshold intervals is the same as the number of risk level levels, thereby completing the fitting and optimization of the configuration parameters of the prediction model.

[0114] Taking auto insurance as an example, the second business information is used as risk prediction data, and a preset prediction model is used to predict the risk level of the target potential customer. In other words, the claims information is used as risk prediction data, and a preset prediction model is used to predict the risk level of the high-potential customer.

[0115] By setting corresponding risk threshold ranges for different risk levels, it is easy to determine the risk level corresponding to the target potential customer. At the same time, the configuration parameters of the prediction model are fitted and optimized using the LightGBM model so that the risk threshold range can be changed in a timely manner according to the change of risk level when the risk level setting changes.

[0116] Step 206: Based on the preset judgment conditions and the risk level, determine whether to quote the first service to the target potential customer, and recommend the first service to the final potential customer based on the judgment result.

[0117] In this embodiment, the final potential customer is the target potential customer for whom the first service quotation needs to be quoted after the judgment is based on preset judgment conditions and the risk level.

[0118] Continue to refer to Figure 7 , Figure 7 yes Figure 2 A flowchart of a specific embodiment of step 206 shown includes the following steps:

[0119] Step 701: Pre-set the price level threshold;

[0120] Step 702: Based on the quoted price level threshold and the risk level, determine whether to offer the first service quote to the target potential customer;

[0121] Step 703: If the risk level does not meet the quotation level threshold, then the first service quotation will not be made to the target potential customer;

[0122] Step 704: If the risk level meets the price level threshold, then send a price request to the preset price platform to obtain the price result of the first business corresponding to the target potential customer.

[0123] Continue to refer to Figure 8 , Figure 8 yes Figure 7 A flowchart of a specific embodiment of step 704 shown includes the following steps:

[0124] Step 801: Set different risk coefficients for different risk levels in advance on the quotation platform;

[0125] Step 802: After receiving the quotation request, obtain the risk level by parsing, and obtain the corresponding risk coefficient through the risk level;

[0126] Step 803: Substitute the risk coefficient as an algorithm factor into the preset pricing algorithm formula to calculate the pricing result, and return the pricing result as the pricing result for the first business corresponding to the target potential customer to the pricing requester.

[0127] Taking auto insurance as an example, the process of determining whether to quote the first business to the target potential customer based on preset judgment conditions and the risk level involves determining whether the risk level corresponding to the high-potential customer meets the condition of no longer quoting. If it does, then no auto insurance quote will be quoted to the high-potential customer; if it does not, then an auto insurance quote will be quoted to the high-potential customer.

[0128] By setting risk levels and risk coefficients, the quoting companies are further guaranteed to mitigate business quotation risks to a certain extent and make reasonable business quotations.

[0129] This application collects the target customer's attribute information from an internal company platform based on the target customer's identification information; based on the identification information, it collects the first business information corresponding to the first business during the period when the target customer is not continuing the first business with the company from an industry-shared platform or authoritative platform; it uses a random forest tree algorithm to classify and train the preprocessed target customer's attribute information and the first business information to obtain target potential customers; it obtains the target potential customer's identification information, and based on the identification information, collects the second business information corresponding to the target potential customer from an industry-shared platform or authoritative platform corresponding to the second business; it uses the second business information as risk prediction data, and predicts the risk level corresponding to the target potential customer through a preset prediction model; based on preset judgment conditions and the risk level, it determines whether to quote the first business to the target potential customer. This application, by conducting a business quotation risk assessment for auto insurance reinsurance customers, facilitates business quotation companies to accurately combine sufficient relevant data to recommend reinsurance-related information to target customers.

[0130] The embodiments of this application can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence (AI) refers to the theories, methods, technologies, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0131] Foundational technologies for artificial intelligence generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing, operating / interactive systems, and mechatronics. AI software technologies mainly encompass computer vision, robotics, biometrics, speech processing, natural language processing, and machine learning / deep learning.

[0132] In this embodiment, attribute information, first business information, and second business information can be obtained from the aforementioned different platforms through big data processing technology. High-intent customers can be automatically identified through the classification model, and pricing risk assessment can be automatically performed through the prediction model, making it more scientific and intelligent.

[0133] Further reference Figure 9 As a response to the above Figure 2 The implementation of the method shown in this application provides an embodiment of an apparatus for recommending financial products to reinsurance clients. This apparatus embodiment is similar to... Figure 2 Corresponding to the method embodiments shown, this device can be specifically applied to various electronic devices.

[0134] like Figure 9 As shown, the device 900 for recommending financial products to reinsurance clients in this embodiment includes: a client attribute acquisition module 901, a first business information acquisition module 902, a potential client screening module 903, a second business information acquisition module 904, a risk prediction module 905, and a price judgment module 906. Wherein:

[0135] The customer attribute acquisition module 901 is used to collect the attribute information of the target customer from the company's internal platform based on the identification information of the target customer. The attribute information includes: the number of calls received during a preset period, the total duration of call connection, the number of proactive inquiries, the duration of proactive inquiries, and the total number of other similar services purchased from the company. The target customer is a group of customers who have a first service history with the company but have not continued the first service this year. The identification information is the target customer's ID card information or the unique customer number corresponding to the target customer. The unique customer number is unique in both the preset industry sharing platform and the company's internal platform.

[0136] The first business information acquisition module 902 is used to collect, based on the identification information, the first business information corresponding to the target customer continuing the first business in other companies in the same industry from the industry sharing platform, wherein the industry sharing platform is a platform in which multiple companies in the same industry can share the business information;

[0137] The potential customer screening module 903 is used to preprocess the attribute information and the first business information, and use the random forest tree algorithm to classify and train the preprocessed attribute information and the first business information to obtain the target potential customers after classification and training, wherein the target potential customers are customers who meet the preset intention criteria.

[0138] The second business information acquisition module 904 is used to acquire the identification information of the target potential customer, and collect the second business information corresponding to the target potential customer from the industry sharing platform or authoritative platform corresponding to the second business based on the identification information, wherein the second business is a follow-up business of the first business;

[0139] The risk prediction module 905 is used to use the second business information as risk prediction data and predict the risk level corresponding to the target potential customer through a preset prediction model.

[0140] The quotation judgment module 906 is used to determine whether to quote the first service to the target potential customer based on preset judgment conditions and the risk level, and to recommend the first service to the final potential customer based on the judgment result.

[0141] This application collects the target customer's attribute information from an internal company platform based on the target customer's identification information; based on the identification information, it collects the first business information corresponding to the first business during the period when the target customer is not continuing the first business with the company from an industry-shared platform or authoritative platform; it uses a random forest tree algorithm to classify and train the preprocessed target customer's attribute information and the first business information to obtain target potential customers; it obtains the target potential customer's identification information, and based on the identification information, collects the second business information corresponding to the target potential customer from an industry-shared platform or authoritative platform corresponding to the second business; it uses the second business information as risk prediction data, and predicts the risk level corresponding to the target potential customer through a preset prediction model; based on preset judgment conditions and the risk level, it determines whether to quote the first business to the target potential customer. This application, by conducting a business quotation risk assessment for auto insurance reinsurance customers, facilitates business quotation companies to accurately combine sufficient relevant data to recommend reinsurance-related information to target customers.

[0142] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by instructing related hardware through computer-readable instructions. These computer-readable instructions can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the methods described above. The aforementioned storage medium can be a non-volatile storage medium such as a magnetic disk, optical disk, or read-only memory (ROM), or random access memory (RAM).

[0143] It should be understood that although the steps in the flowcharts of the accompanying figures are shown sequentially as indicated by the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the accompanying figures may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the sub-steps or stages of other steps.

[0144] To address the aforementioned technical problems, embodiments of this application also provide a computer device. Please refer to [link / reference needed]. Figure 10 , Figure 10 This is a basic structural block diagram of the computer device in this embodiment.

[0145] The computer device 10 includes a memory 10a, a processor 10b, and a network interface 10c that are interconnected via a system bus. It should be noted that only the computer device 10 with components 10a-10c is shown in the figure; however, it should be understood that it is not required to implement all the shown components, and more or fewer components can be implemented alternatively. Those skilled in the art will understand that the computer device described here is a device capable of automatically performing numerical calculations and / or information processing according to pre-set or stored instructions, and its hardware includes, but is not limited to, microprocessors, application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), digital signal processors (DSPs), embedded devices, etc.

[0146] The computer device can be a desktop computer, laptop, handheld computer, or cloud server, etc. The computer device can interact with the user via a keyboard, mouse, remote control, touchpad, or voice control.

[0147] The memory 10a includes at least one type of readable storage medium, including flash memory, hard disk, multimedia card, card-type memory (e.g., SD or DX memory), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, magnetic disk, optical disk, etc. In some embodiments, the memory 10a may be an internal storage unit of the computer device 10, such as the hard disk or memory of the computer device 10. In other embodiments, the memory 10a may also be an external storage device of the computer device 10, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc., equipped on the computer device 10. Of course, the memory 10a may also include both the internal storage unit and the external storage device of the computer device 10. In this embodiment, the memory 10a is typically used to store the operating system and various application software installed on the computer device 10, such as computer-readable instructions for recommending financial products to reinsurance clients. Furthermore, the memory 10a can also be used to temporarily store various types of data that have been output or will be output.

[0148] In some embodiments, the processor 10b may be a central processing unit (CPU), controller, microcontroller, microprocessor, or other data processing chip. The processor 10b is typically used to control the overall operation of the computer device 10. In this embodiment, the processor 10b is used to execute computer-readable instructions stored in the memory 10a or to process data, for example, to execute computer-readable instructions for the method of recommending financial products to reinsurance customers.

[0149] The network interface 10c may include a wireless network interface or a wired network interface, which is typically used to establish communication connections between the computer device 10 and other electronic devices.

[0150] The computer equipment proposed in this embodiment belongs to the field of financial technology. This application collects the target customer's attribute information from an internal company platform based on the target customer's identification information; based on the identification information, it collects the first business information corresponding to the first business during the period when the target customer is not continuing the first business with the company from an industry-shared platform or authoritative platform; it uses a random forest tree algorithm to classify and train the preprocessed target customer's attribute information and the first business information to obtain potential target customers; it obtains the target potential customer's identification information, and based on the identification information, collects the second business information corresponding to the target potential customer from an industry-shared platform or authoritative platform corresponding to the second business; it uses the second business information as risk prediction data, and predicts the risk level corresponding to the target potential customer through a preset prediction model; based on preset judgment conditions and the risk level, it determines whether to quote the first business to the target potential customer. This application, by conducting a business quotation risk assessment for auto insurance reinsurance customers, facilitates business quotation companies to accurately recommend reinsurance-related information to target customers by combining sufficient relevant data.

[0151] This application also provides another embodiment, namely, a computer-readable storage medium storing computer-readable instructions that can be executed by a processor to cause the processor to perform the steps of the method for recommending financial products to reinsurance customers as described above.

[0152] The computer-readable storage medium proposed in this embodiment belongs to the field of financial technology. This application collects the target customer's attribute information from an internal company platform based on the target customer's identification information; based on the identification information, it collects the first business information corresponding to the first business during the period when the target customer is not continuing the first business with the company from an industry-shared platform or authoritative platform; it uses a random forest tree algorithm to classify and train the preprocessed target customer's attribute information and the first business information to obtain potential target customers; it obtains the target potential customer's identification information, and based on the identification information, collects the second business information corresponding to the target potential customer from an industry-shared platform or authoritative platform corresponding to the second business; it uses the second business information as risk prediction data, and predicts the risk level corresponding to the target potential customer through a preset prediction model; based on preset judgment conditions and the risk level, it determines whether to quote the first business to the target potential customer. This application, by conducting a business quotation risk assessment for auto insurance reinsurance customers, facilitates business quotation companies to accurately recommend reinsurance-related information to target customers by combining sufficient relevant data.

[0153] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0154] Obviously, the embodiments described above are only some embodiments of this application, not all embodiments. The accompanying drawings show preferred embodiments of this application, but do not limit the patent scope of this application. This application can be implemented in many different forms; rather, the purpose of providing these embodiments is to provide a more thorough and comprehensive understanding of the disclosure of this application. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing specific embodiments, or make equivalent substitutions for some of the technical features. Any equivalent structures made using the content of this application's specification and drawings, directly or indirectly applied to other related technical fields, are similarly within the scope of patent protection of this application.

Claims

1. A method for recommending financial products to reinsurance clients, characterized in that, Includes the following steps: Based on the identification information of the target customers, the attribute information of the target customers is collected from the company's internal platform. The target customers are customers who have a first business history with the company but have not continued the first business in the current year. The identification information is the ID card information of the target customer or the unique customer number corresponding to the target customer. The attribute information includes: the number of calls received during the preset period, the total call connection time, the number of proactive inquiries, the duration of proactive inquiries, and the total number of other similar services purchased from the company. Based on the identification information, collect information from the industry sharing platform on the first business information corresponding to the first business continued by the target customer in other companies in the same industry; The attribute information and the first business information are preprocessed, and the preprocessed attribute information and the first business information are classified and trained using the random forest tree algorithm to obtain the target potential customers after classification and training. Obtain the identification information of the target potential customer, and collect the second business information corresponding to the target potential customer from the industry sharing platform or authoritative platform corresponding to the second business based on the identification information. The first business is auto insurance business, and the second business is related claims business generated by the auto insurance business as a prior agreement. Using the second business information as risk prediction data, a preset prediction model is used to predict the risk level corresponding to the target potential customer. Based on preset judgment conditions and the risk level, determine whether to quote the first service to the target potential customer, and recommend the first service to the final potential customer based on the judgment result.

2. The method for recommending financial products to reinsurance clients according to claim 1, characterized in that, The step of collecting information from the industry sharing platform, based on the identification information, regarding the first business information of the target customer continuing the first business with other companies in the same industry, specifically includes: A distinguishing status code is pre-set based on whether the first service information corresponding to the first service can be retrieved. The distinguishing status code includes a first status code and a second status code. The first status code indicates that the first service information cannot be retrieved, and the second status code indicates that the first service information can be retrieved. Using the identification information as a retrieval field, it is determined whether the first business information corresponding to the first business during the period when the target customer was not continuing the first business with the company can be retrieved from an industry sharing platform or an authoritative platform. If the first service information cannot be retrieved, the distinguishing status code corresponding to the first service is set to the first status code; If the first service information can be retrieved, then the distinguishing status code corresponding to the first service is set to the second status code.

3. The method for recommending financial products to reinsurance clients according to claim 2, characterized in that, The step of preprocessing the attribute information and the first business information specifically includes: A data set with the same number of categories as the attribute information is constructed in advance, and a set differentiation identifier is set according to the different categories of the attribute information. Taking each individual customer in the customer group as a unit, obtain the number of calls received, total call connection time, number of proactive inquiries, duration of proactive inquiries, and total number of other similar services purchased by the company for each customer in the customer group during the preset period. Based on the set differentiation identifier, the number of calls received, total call connection time, number of proactive inquiries, duration of proactive inquiries, and total number of purchases of other similar services of the company during the preset period are cached into their respective corresponding data sets; The data within each of the data sets is normalized, mapping the data within each data set to a range of 0 to 1; A data set is pre-constructed for caching the distinguishing status codes corresponding to the first service information; Obtain the distinctive status code corresponding to the first business information for each customer in the customer group; The distinguishing status code is cached in the data set, and the elements in the data set are divided according to the different code values ​​of the distinguishing status code, and the data in the data set is replaced with 0 or 1.

4. The method for recommending financial products to reinsurance clients according to claim 3, characterized in that, The step of using a random forest tree algorithm to classify and train the preprocessed attribute information and the first business information to obtain the target potential customers after classification training specifically includes: Pre-set intention thresholds for different categories of the attribute information; The data sets corresponding to the different categories of the preprocessed attribute information are input into the classification model based on the random forest tree algorithm, and it is determined in turn whether each data in the data set is greater than its corresponding intention threshold. If it is greater than, then the intention category corresponding to the current data in the dataset is high intention; If it is less than, then the intention category corresponding to the current data in the dataset is low intention; Obtain the data set corresponding to the preprocessed first business information; Traverse the elements in the dataset. If the current element is 0, the intention category corresponding to the element is low intention. If the current element is 1, the intention category corresponding to the element is high intention. Based on the identification information corresponding to each customer in the customer group, the intention category corresponding to each customer in different data sets is obtained, and probability statistics are performed based on a preset probability algorithm to obtain statistical results; Based on the statistical results and the preset probability threshold, the business intention category corresponding to each customer is determined, and the business intention category corresponding to each customer is used as the output result of the classification model for output. According to the preset filtering rules, target potential customers are selected from the output results, wherein the filtering rules are to select customers with high interest in the business intention category as target potential customers.

5. The method for recommending financial products to reinsurance clients according to claim 1, characterized in that, The step of using the second business information as risk prediction data and predicting the risk level of the target potential customer through a preset prediction model specifically includes: Pre-set corresponding risk threshold ranges for different risk levels, and use the risk threshold ranges as configuration parameters for the prediction model; The risk prediction data is input into the initialized prediction model, and the risk threshold range corresponding to the risk prediction data is determined by comparison. The risk level of the target potential customer is determined based on the risk threshold range corresponding to the risk prediction data.

6. The method for recommending financial products to reinsurance clients according to claim 5, characterized in that, In the process of using the second business information as risk prediction data and predicting the risk level of the target potential customer through a preset prediction model, the method further includes: Based on the preset configuration file, identify whether there have been any changes to the number of risk levels. If the number of risk levels is changed, the corresponding risk threshold intervals are adjusted using the LightGBM model so that the number of risk threshold intervals is the same as the number of risk levels, thereby completing the fitting and optimization of the configuration parameters of the prediction model.

7. The method for recommending financial products to reinsurance clients according to claim 1, characterized in that, The step of determining whether to offer the first service quote to the target potential customer based on preset judgment conditions and the risk level specifically includes: Pre-set the price level threshold; Based on the quoted price level threshold and the risk level, determine whether to offer the first service quote to the target potential customer; If the risk level does not meet the pricing level threshold, then the first service quote will not be made to the target potential customer; If the risk level meets the price level threshold, a price request is sent to the preset price platform to obtain the price result for the first service corresponding to the target potential customer.

8. The method for recommending financial products to reinsurance clients according to claim 7, characterized in that, The step of issuing a quote request to a preset quote platform and obtaining the quote result for the first service corresponding to the target potential customer if the risk level meets the quote level threshold specifically includes: Different risk coefficients are pre-set for different risk levels on the quotation platform; Upon receiving the quotation request, the risk level is obtained through parsing, and the corresponding risk coefficient is obtained through the risk level. The risk coefficient is substituted into the preset pricing algorithm formula as an algorithm factor to calculate the pricing result. The pricing result is then returned to the pricing requester as the pricing result for the first service corresponding to the target potential customer.

9. An apparatus for recommending financial products to reinsurance clients, characterized in that, include: The customer attribute acquisition module is used to collect the attribute information of the target customer from the company's internal platform based on the target customer's identification information. The target customer is a customer group that has a first business history with the company but has not continued the first business in the current year. The identification information is the target customer's ID card information or the unique customer number corresponding to the target customer. The attribute information includes: the number of calls received during the preset period, the total call connection time, the number of proactive inquiries, the duration of proactive inquiries, and the total number of other similar services purchased from the company. The first business information acquisition module is used to collect, based on the identification information, the first business information corresponding to the target customer continuing the first business in other companies in the same industry from the industry sharing platform; The potential customer screening module is used to preprocess the attribute information and the first business information, and use the random forest tree algorithm to classify and train the preprocessed attribute information and the first business information to obtain the target potential customers after classification and training. The second business information acquisition module is used to acquire the identification information of the target potential customer, and collect the second business information corresponding to the target potential customer from the industry sharing platform or authoritative platform corresponding to the second business based on the identification information. The first business is auto insurance business, and the second business is related claims business generated by the auto insurance business as a prior agreement. The risk prediction module is used to use the second business information as risk prediction data and predict the risk level corresponding to the target potential customer through a preset prediction model. The pricing judgment module is used to determine whether to quote the first service to the target potential customer based on preset judgment conditions and the risk level, and to recommend the first service to the final potential customer based on the judgment result.

10. A computer device comprising a memory and a processor, the memory storing computer-readable instructions, wherein the processor, when executing the computer-readable instructions, implements the steps of the method for recommending financial products to reinsurance customers as claimed in any one of claims 1 to 8.

11. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-readable instructions, which, when executed by a processor, implement the steps of the method for recommending financial products to reinsurance clients as described in any one of claims 1 to 8.

Images