WO2019061994A1

WO2019061994A1 - Electronic device, insurance product recommendation method and system, and computer readable storage medium

Info

Publication number: WO2019061994A1
Application number: PCT/CN2018/076186
Authority: WO
Inventors: 李芳�; 王建明; 肖京
Original assignee: 平安科技（深圳）有限公司
Priority date: 2017-09-30
Filing date: 2018-02-10
Publication date: 2019-04-04
Also published as: CN107730389A

Abstract

Disclosed in the present application is an insurance product recommendation method. The method comprises: if a target insurance product needs to be recommended to a customer, acquiring comprehensive data information concerning the customer; analyzing, according to a pre-determined insurance product recommendation model, the acquired comprehensive data information concerning the customer, to obtain a preference probability value of the customer toward the target insurance product; if the obtained preference probability value is greater than a preset probability threshold, determining it is necessary to recommend the target insurance product to the customer, and sending a recommendation instruction regarding the insurance product to a pre-determined terminal. The present application can determine the real needs of customers in a timely and accurate manner, improving customer experience and the success rate of business personnel.

Description

Electronic device, insurance product recommendation method, system and computer readable storage medium

This application claims the priority of the Chinese Patent Application filed on Sep. 30, 2017, the Chinese Patent Application No. 201710916518.3, the name of the invention is the electronic device, the recommended method of the insurance product, and the computer readable storage medium, the entire contents of which are incorporated by reference. Combined in this application.

Technical field

The present application relates to the field of sales of security products, and in particular, to an electronic device, an insurance product recommendation method, a system, and a computer readable storage medium.

Background technique

In recent years, with the improvement of living standards, people have higher and higher service requirements for business personnel when purchasing products. Usually, if the business personnel can grasp the real needs of customers in the first time, and The real needs are the starting point. When the customer introduces and recommends related products, the proportion of customers who continue to pay attention to or purchase the products will be much improved. For example, in the insurance industry, if the customer can multi-dimensional data information, timely and accurate mining With the real needs of different customers, it is possible to specifically recommend the corresponding types of insurance products to different customers, which can greatly improve the probability of successful insurance sales. Currently, the insurance industry's business personnel are usually selectively recommended based on the customer's living conditions (for example, the rise of the business and the structure of the family members) and the customer's basic identity information (eg, age, gender, income, etc.). Related insurance products, this recommendation method has a certain degree of specificity compared to the traditional blind recommendation, and relatively increases the probability of sales. However, in the current era of ever-changing information, the needs of customers are also dynamic and diverse. Therefore, the existing methods for recommending insurance products to customers cannot accurately and accurately uncover the real needs of customers, resulting in a customer experience. Good, the business success rate of business personnel is limited.

Summary of the invention

In view of this, the present application provides an insurance product recommendation method, an electronic device, and a computer readable storage medium, which can accurately mine the real needs of the customer according to the customer's comprehensive data information, improve the customer experience, and further improve the business personnel. Business efficiency.

First, in order to achieve the above object, a first aspect of the present application provides an electronic device including a memory, a processor, and an insurance product recommendation system stored on the memory and operable on the processor. The insurance product recommendation system is implemented by the processor to implement the following steps:

A. If there is a target insurance product that needs to be recommended to the customer with the identification information, the customer's comprehensive data information corresponding to the customer's identification information is obtained from the predetermined database;

B. analyzing the acquired customer comprehensive data information by using a predetermined insurance product recommendation model to obtain a preference probability value corresponding to the target insurance product;

C. If the obtained preference probability value is greater than the preset probability threshold, it is determined that the target insurance product needs to be recommended to the customer, and the recommended instruction for the insurance product is sent to the predetermined terminal.

In addition, in order to achieve the above object, the second aspect of the present application further provides a method for recommending an insurance product, the method comprising the following steps:

Further, in order to achieve the above object, a third aspect of the present application further provides a computer readable storage medium storing an insurance product recommendation system, the insurance product recommendation system being executable by at least one processor So that the at least one processor performs the following steps:

If there is a target insurance product that needs to be recommended to the customer with the identification information, the customer comprehensive data information corresponding to the customer's identification information is obtained from the predetermined database;

B. analyzing the obtained customer comprehensive data information by using a predetermined insurance product recommendation model to obtain a preference probability value corresponding to the target insurance product;

Compared with the prior art, the electronic device, the insurance product recommendation method and the computer readable storage medium proposed by the present application firstly, if a target insurance product needs to be recommended to a customer with identification information, from a predetermined database Obtaining the customer's comprehensive data information corresponding to the customer's identification information; and then using the predetermined insurance product recommendation model to analyze the acquired customer's comprehensive data information to obtain the customer's preference probability corresponding to the target insurance product a value; then, if the derived preference probability value is greater than the preset probability threshold, it is determined that the target insurance product needs to be recommended to the customer, and the recommended instruction for the insurance product is sent to the predetermined terminal. In this way, considering the disadvantages of the customer's contact portrait feature and the static portrait feature to effectively avoid poor communication with the customer, resulting in a bad customer experience, it is also considered to promptly dig out the dynamic demand of the customer according to the dynamic image feature of the customer. To more accurately recommend relevant insurance products to customers and improve the business success rate of business personnel.

DRAWINGS

1 is a schematic diagram of an optional hardware architecture of an electronic device;

2 is a schematic diagram of a program module of a preferred embodiment of the insurance product recommendation system of the present application;

3 is a schematic flow chart of a preferred embodiment of the insurance product recommendation method of the present application.

The implementation, functional features and advantages of the present application will be further described with reference to the accompanying drawings.

Detailed ways

In order to make the objects, technical solutions, and advantages of the present application more comprehensible, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It is understood that the specific embodiments described herein are merely illustrative of the application and are not intended to be limiting. All other embodiments obtained by a person of ordinary skill in the art based on the embodiments of the present application without departing from the inventive scope are the scope of the present application.

It should be noted that the descriptions of "first", "second" and the like in the present application are for the purpose of description only, and are not to be construed as indicating or implying their relative importance or implicitly indicating the number of technical features indicated. . Thus, features defining "first" and "second" may include at least one of the features, either explicitly or implicitly. In addition, the technical solutions between the various embodiments may be combined with each other, but must be based on the realization of those skilled in the art, and when the combination of the technical solutions is contradictory or impossible to implement, it should be considered that the combination of the technical solutions does not exist. Nor is it within the scope of protection required by this application.

Referring to FIG. 1 , it is a schematic diagram of an optional hardware architecture of the electronic device 1 of the present application. In this embodiment, the electronic device 1 may include, but is not limited to, the memory 11, the processor 12, and the network interface 13 being communicably connected to each other through a system bus. It should be noted that FIG. 1 only shows the electronic device 1 having the components 11-13, but it should be understood that not all illustrated components may be implemented, and more or fewer components may be implemented instead.

The memory 11 includes at least one type of readable storage medium including a flash memory, a hard disk, a multimedia card, a card type memory (for example, SD or DX memory, etc.), a random access memory (RAM), and a 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, and the like. In some embodiments, the memory 11 may be an internal storage unit of the electronic device 1, such as a hard disk or a memory of the electronic device 1. In other embodiments, the memory 11 may also be an external storage device of the electronic device 1, such as a plug-in hard disk equipped on the electronic device 1, a smart memory card (SMC), and a secure digital (Secure Digital, SD). ) cards, flash cards, etc. Of course, the memory 11 can also include both an internal storage unit of the electronic device 1 and an external storage device thereof. In the present embodiment, the memory 11 is generally used to store an operating system installed in the electronic device 1 and various types of application software, such as program codes of the insurance product recommendation system 200. Further, the memory 11 can also be used to temporarily store various types of data that have been output or are to be output.

Processor 12 may be a Central Processing Unit (CPU), controller, microcontroller, microprocessor, or other data processing chip in some embodiments. The processor 12 is typically used to control the overall operation of the electronic device 1. In this embodiment, the processor 12 is configured to run program code or processing data stored in the memory 11, such as the running insurance product recommendation system 200 and the like.

The network interface 13 may include a wireless network interface or a wired network interface, and the network interface 13 is generally used to establish a communication connection between the electronic device 1 and other electronic devices.

Figure 1 shows only a social network based user keyword extraction device with components 11-13 and a user keyword extraction program, but it should be understood that not all illustrated components may be implemented, and alternative implementations may be implemented. Or fewer components.

Optionally, the device may further include a user interface, the user interface may include a display, an input unit such as a keyboard, and the optional user interface may further include a standard wired interface and a wireless interface. Optionally, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode) touch sensor, or the like. The display may also be suitably referred to as a display screen or display unit for displaying information processed in the social network based user keyword extraction device and a user interface for displaying visualization.

First, the present application proposes an insurance product recommendation system 200.

Referring to FIG. 3, it is a program block diagram of a preferred embodiment of the insurance product recommendation system 200 of the present application. In this embodiment, the insurance product recommendation system 200 can be divided into one or more modules, one or more modules are stored in the memory 11, and by one or more processors (the processor 12 in this embodiment) Executed to complete the application. For example, in FIG. 3, the insurance product recommendation system 200 can be divided into an acquisition module 201, an analysis module 202, and an instruction transmission module 203. The program module referred to in the present application refers to a series of computer program instruction segments capable of performing a specific function, and is more suitable than the program to describe the execution process of the insurance product recommendation system 200 in the electronic device 2. The function of each program module 201-203 will be described in detail below.

The obtaining module 201 is configured to: if a target insurance product needs to be recommended to the customer with the identification information, obtain the customer comprehensive data information corresponding to the customer identification information from the predetermined database.

The identification information includes an ID number, a passport number, or a mobile phone number, and the predetermined database stores: feedback data of the customer on the marketing mode, for example, feedback data of the customer on the historical sales (for example, whether the user is willing to contact the sales) Feedback data of the customer on the historical online sales (for example, the probability of clicking the online sales link, the duration of the browsing, etc.), the feedback data of the customer on the SMS marketing, and the feedback data of the customer on the APP marketing.

Customer's basic information data, for example, customer's insurance claims information (for example, insurance purchased in the last year, number of claims required, etc.), banking information (for example, mortgage information, credit card information, etc.), customer attribute information (for example, Age, gender, occupation, annual income, birthplace, etc.)

The customer clicks on the data of the website WEB and the mobile app.

Customer's comprehensive data information includes customer contact image features, customer static image features and customer dynamic image features.

Further, customer contact portrait features include customer contact channel preferences (eg, like phone contact communication), customer contact product preferences (eg, frequent attention to pension insurance and critical illness insurance, etc.) and customer contact time preferences (eg, like evenings) The customer's static portrait features include customer identity information (eg, age, gender, occupation, annual income, etc.), customer asset information (eg, property information, deposit information, etc.) Customer loan information (such as online loan product category, online shopping frequency, frequent purchase of consumer goods category, etc.), customer dynamic image features include customer clicks on each product page, Click frequency, number of views, and frequency of views.

The analysis module 202 is configured to analyze the acquired customer comprehensive data information by using a predetermined insurance product recommendation model to obtain a preference probability value corresponding to the target insurance product.

The predetermined insurance product recommendation model is a random forest model, and the establishment process of the insurance product recommendation model includes:

D1. Obtain a preset quantity (for example, 150) of policy information of a customer who purchases insurance, and each policy information sample includes identification information of a corresponding customer who purchases insurance;

D2: Extracting the customer comprehensive data information corresponding to the customer identification information in each policy information sample from the predetermined database, and dividing the extracted customer comprehensive data information set into the first ratio (for example, 70%) a training set and a second set (eg, 30%) of the test set;

D3. The pre-determined insurance product recommendation model is trained by using the comprehensive data information of each customer in the training set to obtain the trained insurance type recommendation model, and the insurance product recommendation model is tested by using the comprehensive data information of each customer in the test set.

D4. If the test passes, the training ends, or if the test fails, the number of policy information samples of the customer who purchased the insurance is increased and the above steps D2 and D3 are re-executed.

Further, the insurance product recommendation model testing process includes:

Using the trained insurance product recommendation model to analyze the comprehensive data information of each customer in the test set to obtain the preference probability value of each customer for different insurance products;

If the customer's preference probability value for the insurance product is greater than a preset probability threshold (for example, 50%), the model is tested for the accuracy of the customer, and the insurance product that the customer has purchased is extracted from the customer's policy data. If the second insurance product corresponding to the preferred probability value of the first insurance product purchased by the customer is the same type of insurance product, it is determined that the model accuracy test result for the customer is correct, or if the customer has purchased If the second insurance product corresponding to the probability value of the preference is a different type of insurance product, it is determined that the model accuracy test result for the customer is an error;

If the correct model accuracy test results account for more than the preset percentage threshold (for example, 60%), then the test for the insurance product recommendation model is determined, or if the correct model accuracy is If the percentage of test results to all model accuracy test results is less than or equal to the preset percentage threshold, then the test for the insurance product recommendation model is determined to fail.

Specifically, in an embodiment, the K samples are randomly selected from the training set, and the N features of the M samples are randomly selected for training to generate a random forest model including K decision trees. Each decision tree includes non-page nodes, leaf nodes, and branches; each non-page node represents a conditional judgment of a feature (eg, age 45, gender male, property, etc.), each leaf node Represents the attributes of the model training classification (eg, preferences, no preferences), and each branch indicates whether the customer prefers a particular type of insurance (eg, whether to prefer pension insurance, whether to favor child education insurance, etc.).

Further, the N features of the m samples randomly selected from the test sample set are substituted into the trained insurance product recommendation model to predict the preference probability value of the insurance type for the customers who have purchased insurance in the m samples.

The instruction sending module 203 is configured to: when the derived preference probability value is greater than the preset probability threshold, send a recommendation instruction for the insurance product to the predetermined terminal, or the used preference probability value is less than or equal to the pre-predetermined value If the probability threshold is set, the target insurance product is saved as the insurance product to be recommended, and a preset proportion of the to-be-recommended insurance product is randomly selected from the insurance products to be recommended, and the selected candidate is sent to the predetermined terminal. Recommended instructions for insurance products.

It can be seen from the above embodiments that the insurance recommendation system of the present application first obtains the comprehensive data information of the customer when a target insurance product needs to be recommended to the customer, and the comprehensive data information includes the characteristics of the customer contact portrait, the characteristics of the customer static portrait and The customer dynamic image feature is then analyzed by the predetermined insurance product recommendation model to obtain the customer's full-scale data information, to obtain the customer's preference probability value for the target insurance product, and if the preference probability value is greater than the preset probability threshold, Then, it is determined that the target insurance product is recommended to the customer, and the recommended instruction for the target insurance product is sent to the predetermined terminal. In this way, compared with the existing recommendation methods, considering the customer's contact portrait features and static portrait features, effectively avoiding the poor communication effect with the customer, resulting in a bad customer experience, but also considering the dynamic image according to the customer. Features timely mining out the dynamic needs of customers, more accurately recommend relevant insurance products to customers, and improve the business success rate of business personnel.

Secondly, the present application proposes an insurance product recommendation method.

Referring to FIG. 4, it is a flowchart of an implementation of a preferred embodiment of the insurance product recommendation method of the present application. In the embodiment, the insurance product recommendation method includes steps S301 to S303.

S301. If a target insurance product needs to be recommended to the customer with the identification information, obtain the customer comprehensive data information corresponding to the customer identification information from the predetermined database.

The identification information includes an ID number, a passport number, or a mobile phone number, and the predetermined database includes: feedback data of the customer on the marketing mode, for example, feedback data of the customer on the historical sales (for example, whether the user is willing to connect to the sales call, Feedback data of the customer's historical online sales (for example, the probability of clicking the online sales link, the duration of the browsing, etc.), the customer's feedback data on the SMS marketing, and the customer's feedback data on the APP marketing.

The customer clicks on the data of the website WEB and the mobile app.

S302. Analyze the acquired customer comprehensive data information by using a predetermined insurance product recommendation model to obtain a preference probability value corresponding to the target insurance product.

Further, the insurance product recommendation model testing process includes:

S303. If the obtained preference probability value is greater than the preset probability threshold, determine that the target insurance product needs to be recommended to the client, send a recommendation instruction for the insurance product to the predetermined terminal, or obtain a preference probability. If the value is less than or equal to the preset probability threshold, it is determined that the target insurance product is not recommended to the customer, and the target insurance product needs to be saved as the insurance product to be recommended, and the preset is randomly selected from the insurance products to be recommended. The proportion of the insurance product to be recommended, and the recommendation instruction for each extracted insurance product to be recommended is sent to the predetermined terminal.

The serial numbers of the embodiments of the present application are merely for the description, and do not represent the advantages and disadvantages of the embodiments.

Through the description of the above embodiments, those skilled in the art can clearly understand that the foregoing embodiment method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be through hardware, but in many cases, the former is better. Implementation. Based on such understanding, the technical solution of the present application, which is essential or contributes to the prior art, may be embodied in the form of a software product stored in a storage medium (such as ROM/RAM, disk, The optical disc includes a number of instructions for causing a terminal device (which may be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to perform the methods described in various embodiments of the present application.

The above is only a preferred embodiment of the present application, and is not intended to limit the scope of the patent application, and the equivalent structure or equivalent process transformations made by the specification and the drawings of the present application, or directly or indirectly applied to other related technical fields. The same is included in the scope of patent protection of this application.

Claims

An electronic device, comprising: a memory, a processor, wherein the memory stores an insurance product recommendation system operable on the processor, the insurance product recommendation system being The following steps are implemented during execution:

A. If there is a target insurance product that needs to be recommended to the customer with the identification information, the customer's comprehensive data information corresponding to the customer's identification information is obtained from the predetermined database;

B. analyzing the acquired customer comprehensive data information by using a predetermined insurance product recommendation model to obtain a preference probability value corresponding to the target insurance product;

C. If the obtained preference probability value is greater than the preset probability threshold, the recommended instruction for the insurance product is sent to the predetermined terminal.
The electronic device according to claim 1, wherein the identification information comprises an ID number, a passport number, or a mobile phone number, wherein the predetermined database stores feedback data of the customer on the marketing mode, and the basis of the customer. The information data and the customer's click browsing data of the website WEB and the mobile APP, the customer comprehensive data information includes customer contact portrait features, customer static portrait features and customer dynamic portrait features.
The electronic device of claim 2, wherein the customer contact image feature comprises a customer contact channel preference, a customer contact product preference, and a customer contact time preference, the customer static image feature including customer identity information, customer asset information And customer consumption information, the customer dynamic image feature includes the number of clicks, click frequency, number of views, and browsing frequency of the customer for each product page.
The electronic device according to claim 3, wherein the predetermined insurance product recommendation model is a random forest model, and the establishment process of the insurance product recommendation model includes:

D1. Obtain a preset number of policy information samples of a customer who purchases insurance, and each of the policy information samples includes corresponding identification information of a customer who purchases insurance;

D2: Extracting the customer comprehensive data information corresponding to the customer identification information in each policy information sample from the predetermined database, and dividing the extracted customer comprehensive data information set into the first proportion training set and the second Proportional test set;

D3. Training a predetermined insurance product recommendation model by using various customer data of the training set to obtain a trained insurance type recommendation model, and using the comprehensive data information of each customer in the test set to use the insurance product Recommend the model to test;

D4. If the test passes, the training ends, or if the test fails, the number of policy information samples of the customer who purchased the insurance is increased, and the above steps D2 and D3 are re-executed.
The electronic device according to claim 4, wherein the step of testing the insurance product recommendation model by using various customer comprehensive data information in the test set comprises:

Using the trained insurance product recommendation model to analyze the comprehensive data information of each customer in the test set to obtain a preference probability value of each customer for different insurance products;

If the customer's preference probability value for the insurance product is greater than a preset probability threshold, the model is tested for the accuracy of the customer, and the insurance product that the customer has purchased is extracted from the customer's policy data;

If the second insurance product corresponding to the preferred probability value of the first insurance product purchased by the customer is the same type of insurance product, it is determined that the model accuracy test result for the customer is correct, or if the customer has purchased If the second insurance product corresponding to the probability value of the preference is a different type of insurance product, it is determined that the model accuracy test result for the customer is an error;

If the correct model accuracy test results account for more than the preset percentage threshold for all model accuracy test results, then the test for the insurance product recommendation model is determined, or if the correct model accuracy test results account for all If the percentage of the model accuracy test result is less than or equal to the preset percentage threshold, it is determined that the test of the insurance product recommendation model fails.
An insurance product recommendation method, characterized in that the method comprises the following steps:

A. If there is a target insurance product that needs to be recommended to the customer with the identification information, the customer's comprehensive data information corresponding to the customer's identification information is obtained from the predetermined database;

B. analyzing the acquired customer comprehensive data information by using a predetermined insurance product recommendation model to obtain a preference probability value corresponding to the target insurance product;

C. If the obtained preference probability value is greater than the preset probability threshold, it is determined that the target insurance product needs to be recommended to the customer, and the recommended instruction for the insurance product is sent to the predetermined terminal.
The insurance product recommendation method according to claim 6, wherein the identification information comprises an ID number, a passport number, or a mobile phone number, wherein the predetermined database stores feedback data of the customer on the marketing mode, and the customer The basic information data and the customer's click browsing data of the website WEB and the mobile APP, the customer's all-round data information includes customer contact portrait features, customer static portrait features and customer dynamic portrait features.
The insurance product recommendation method according to claim 7, wherein the customer contact image feature comprises a customer contact channel preference, a customer contact product preference, and a customer contact time preference, the customer static image feature including customer identity information, a customer The asset information and the customer consumption information, the customer dynamic image feature includes the number of clicks, click frequency, number of views, and browsing frequency of the customer for each product page.
The insurance product recommendation method according to claim 8, wherein the predetermined insurance product recommendation model is a random forest model, and the predetermined insurance product recommendation model establishment process comprises:

D1. Obtain a preset number of policy information samples of a customer who purchases insurance, and each of the policy information samples includes corresponding identification information of a customer who purchases insurance;

D2: Extracting the customer comprehensive data information corresponding to the customer identification information in each policy information sample from the predetermined database, and dividing the extracted customer comprehensive data information set into the first proportion training set and the second Proportional test set;

D3. Training a predetermined insurance product recommendation model by using various customer data of the training set to obtain a trained insurance type recommendation model, and using the comprehensive data information of each customer in the test set to use the insurance product Recommend the model to test;

D4. If the test passes, the training ends, or if the test fails, the number of policy information samples of the customer who purchased the insurance is increased and the above steps D2 and D3 are re-executed.
The insurance product recommendation method according to claim 9, wherein the step of testing the insurance product recommendation model by using the customer comprehensive data information in the test set comprises:

Using the trained insurance product recommendation model to analyze the comprehensive data information of each customer in the test set to obtain a preference probability value of each customer for different insurance products;

If the customer's preference probability value for the insurance product is greater than a preset probability threshold, the model is tested for the accuracy of the customer, and the insurance product that the customer has purchased is extracted from the customer's policy data;

If the second insurance product corresponding to the preferred probability value of the first insurance product purchased by the customer is the same type of insurance product, it is determined that the model accuracy test result for the customer is correct, or if the customer has purchased If the second insurance product corresponding to the probability value of the preference is a different type of insurance product, it is determined that the model accuracy test result for the customer is an error;

If the correct model accuracy test results account for more than the preset percentage threshold for all model accuracy test results, then the test for the insurance product recommendation model is determined, or if the correct model accuracy test results account for all If the percentage of the model accuracy test result is less than or equal to the preset percentage threshold, it is determined that the test of the insurance product recommendation model fails.
An insurance product recommendation system, wherein the insurance product recommendation system comprises an acquisition module, an analysis module, and an instruction sending module;

The obtaining module is configured to: if a target insurance product needs to be recommended to the customer with the identification information, obtain the customer comprehensive data information corresponding to the customer identification information from the predetermined database;

The analysis module is configured to analyze the acquired customer comprehensive data information by using a predetermined insurance product recommendation model to obtain a preference probability value corresponding to the target insurance product by the customer;

The command sending module is configured to: if the derived preference probability value is greater than the preset probability threshold, send a recommended instruction for the insurance product to the predetermined terminal.
The insurance product recommendation system according to claim 11, wherein the identification information comprises an ID number, a passport number, or a mobile phone number, wherein the predetermined database stores feedback data of the customer on the marketing mode, and the customer The basic information data and the customer's click browsing data of the website WEB and the mobile APP, the customer's all-round data information includes customer contact portrait features, customer static portrait features and customer dynamic portrait features.
The insurance product recommendation system of claim 12, wherein the customer contact portrait feature comprises customer contact channel preferences, customer contact product preferences, and customer contact time preferences, the customer static image features including customer identity information, customers The asset information and the customer consumption information, the customer dynamic image feature includes the number of clicks, click frequency, number of views, and browsing frequency of the customer for each product page.
The insurance product recommendation system according to claim 13, wherein the predetermined insurance product recommendation model is a random forest model, and the insurance product recommendation model establishment process comprises:

D1. Obtain a preset number of policy information samples of a customer who purchases insurance, and each of the policy information samples includes corresponding identification information of a customer who purchases insurance;

D2: Extracting the customer comprehensive data information corresponding to the customer identification information in each policy information sample from the predetermined database, and dividing the extracted customer comprehensive data information set into the first proportion training set and the second Proportional test set;

D3. Training a predetermined insurance product recommendation model by using various customer data of the training set to obtain a trained insurance type recommendation model, and using the comprehensive data information of each customer in the test set to use the insurance product Recommend the model to test;

D4. If the test passes, the training ends, or if the test fails, the number of policy information samples of the customer who purchased the insurance is increased, and the above steps D2 and D3 are re-executed.
The insurance product recommendation system according to claim 14, wherein the step of testing the insurance product recommendation model by using various customer comprehensive data information in the test set comprises:

Using the trained insurance product recommendation model to analyze the comprehensive data information of each customer in the test set to obtain a preference probability value of each customer for different insurance products;

If the customer's preference probability value for the insurance product is greater than a preset probability threshold, the model is tested for the accuracy of the customer, and the insurance product that the customer has purchased is extracted from the customer's policy data;

If the second insurance product corresponding to the preferred probability value of the first insurance product purchased by the customer is the same type of insurance product, it is determined that the model accuracy test result for the customer is correct, or if the customer has purchased If the second insurance product corresponding to the probability value of the preference is a different type of insurance product, it is determined that the model accuracy test result for the customer is an error;

If the correct model accuracy test results account for more than the preset percentage threshold for all model accuracy test results, then the test for the insurance product recommendation model is determined, or if the correct model accuracy test results account for all If the percentage of the model accuracy test result is less than or equal to the preset percentage threshold, it is determined that the test of the insurance product recommendation model fails.
A computer readable storage medium storing an insurance product recommendation system executable by at least one processor to cause the at least one processor to perform the following steps:

If there is a target insurance product that needs to be recommended to the customer with the identification information, the customer comprehensive data information corresponding to the customer's identification information is obtained from the predetermined database;

Using the predetermined insurance product recommendation model, the acquired customer comprehensive data information is analyzed to obtain a preference probability value corresponding to the target insurance product by the customer;

If the derived preference probability value is greater than the preset probability threshold, it is determined that the target insurance product needs to be recommended to the customer, and the recommended instruction for the insurance product is sent to the predetermined terminal.
The storage medium according to claim 16, wherein the identification information comprises an ID number, a passport number, or a mobile phone number, wherein the predetermined database stores feedback data of the customer on the marketing mode, and the basis of the customer. The information data and the customer's click browsing data of the website WEB and the mobile APP, the customer comprehensive data information includes customer contact portrait features, customer static portrait features and customer dynamic portrait features.
The storage medium of claim 17, wherein the customer contact image features comprise customer contact channel preferences, customer contact product preferences, and customer contact time preferences, the customer static image features including customer identity information, customer asset information And customer consumption information, the customer dynamic image feature includes the number of clicks, click frequency, number of views, and browsing frequency of the customer for each product page.
The storage medium according to claim 18, wherein the predetermined insurance product recommendation model is a random forest model, and the predetermined insurance product recommendation model establishment process comprises:

D1. Obtain a preset number of policy information samples of a customer who purchases insurance, and each of the policy information samples includes corresponding identification information of a customer who purchases insurance;

D2: Extracting the customer comprehensive data information corresponding to the customer identification information in each policy information sample from the predetermined database, and dividing the extracted customer comprehensive data information set into the first proportion training set and the second Proportional test set;

D3. Training a predetermined insurance product recommendation model by using various customer data of the training set to obtain a trained insurance type recommendation model, and using the comprehensive data information of each customer in the test set to use the insurance product Recommend the model to test;

D4. If the test passes, the training ends, or if the test fails, the number of policy information samples of the customer who purchased the insurance is increased and the above steps D2 and D3 are re-executed.
The storage medium according to claim 19, wherein said step of testing said insurance product recommendation model by using respective customer omnidirectional data information in said test set comprises:

Using the trained insurance product recommendation model to analyze the comprehensive data information of each customer in the test set to obtain a preference probability value of each customer for different insurance products;

If the customer's preference probability value for the insurance product is greater than a preset probability threshold, the model is tested for the accuracy of the customer, and the insurance product that the customer has purchased is extracted from the customer's policy data;

If the second insurance product corresponding to the preferred probability value of the first insurance product purchased by the customer is the same type of insurance product, it is determined that the model accuracy test result for the customer is correct, or if the customer has purchased If the second insurance product corresponding to the probability value of the preference is a different type of insurance product, it is determined that the model accuracy test result for the customer is an error;

If the correct model accuracy test results account for more than the preset percentage threshold for all model accuracy test results, then the test for the insurance product recommendation model is determined, or if the correct model accuracy test results account for all If the percentage of the model accuracy test result is less than or equal to the preset percentage threshold, it is determined that the test of the insurance product recommendation model fails.