Online insurance product sales system and method based on artificial intelligence

Abstract

The invention relates to an online insurance product sales system based on artificial intelligence, and relates to the field of electrical digital data processing, and the system comprises a multi-time training device which carries out the multi-time training of a Hough neural network to obtain an AI sales prediction model; and the sales prediction device is used for intelligently predicting the insurance identifier of the current user relative to the target insurance product within the preset time duration after the current moment by adopting an AI sales prediction model according to the multiple items of basic information. The invention further relates to an online insurance product selling method based on artificial intelligence. According to the invention, the technical problems that future sales data of online insurance products are difficult to count in advance and users who tend to place orders in the future are difficult to ensure ordering are solved; an artificial intelligence mode can be adopted to intelligently predict whether each user will complete insurance buying of the target insurance product within the preset time duration after the current moment according to the multiple items of basic information screened in a targeted mode, and therefore the technical problem is solved.

Description

Technical Field

[0001] The present invention relates to the field of electronic digital data processing, and in particular to an online insurance product sales system and method based on artificial intelligence. Background Art

[0002] Insurance, originally meaning reliable guarantee, has been extended into a guarantee mechanism. It is a tool for financial planning in life, a basic means of risk management under market economy conditions, and an important pillar of the financial system and social security system. Specifically, it refers to a commercial insurance act in which the applicant pays insurance premiums to the insurer according to the contract agreement, and the insurer undertakes the liability of compensating the insurance money for the property losses caused by the possible accidents stipulated in the contract, or undertakes the liability of paying the insurance money when the insured dies, is disabled, suffers from diseases, or reaches the age, term and other conditions stipulated in the contract. With the rapid development of electronic technology and network technology, the related behaviors such as the sales, display and order placement of insurance have gradually shifted from offline to online.

[0003] Exemplarily, an online insurance product sales system proposed in the Chinese invention patent publication text CN113888203A includes: an identity recognition module for collecting and recognizing the user information of insurance customers; an analysis module for automatically analyzing the user information of insurance customers to obtain an analysis verification code; an enumeration module for displaying all relevant insurance products to insurance customers according to the analysis verification code; a selection module for insurance customers to select the most suitable insurance product in combination with their own needs and the suggestions of insurance practitioners; an interaction module for establishing a communication platform and a data transmission channel between insurance customers and insurance practitioners to facilitate the communication between insurance customers and insurance practitioners; and an order contract module for insurance customers to sign an order contract with the insurance company to complete the online sales of insurance products.

[0004] Exemplarily, a method and a system for generating a long-term online insurance policy proposed in the Chinese invention patent publication text CN114049229A. The generation method includes: establishing a policy database; collecting the basic information of the applicant and performing verification to finally obtain the identity information of the applicant; collecting the basic information of the insured and performing verification to finally obtain the insurance information of the insured; determining whether the insured has the qualification to purchase insurance; presenting the insurance products that the applicant can choose; calculating the price of the selected insurance product and the price of the combined product after combining the selected insurance product with other insurance products according to the selected insurance product of the applicant and the insurance information of the insured; and generating the corresponding insurance policy. Through the above content, it can quickly and efficiently issue an insurance policy according to the information data of the applicant and the insured, and give a personalized policy price through a neural network model, achieving a win-win situation for the insurance company and the insured.

[0005] Obviously, the above-mentioned prior art is only limited to the specific implementation method of the online insurance product sales process, providing the ways and entrances for the sales, display and order placement of online insurance products. Only artificial intelligence processing is used to provide personalized policy prices and is not used for attracting customers for online insurance products. Obviously, for various insurance products, whether online or offline, attracting customers is a difficult point. How to judge the order placement tendency of each user through data analysis, and then concentrate limited sales resources to attract customers and guide order placement for users with order placement tendencies, so as to avoid the loss of users with order placement tendencies, is one of the main problems that the prior art needs to solve. Summary of the Invention

[0006] To solve the technical problems in the prior art, the present invention provides an online insurance product sales system and method based on artificial intelligence, which can use the artificial intelligence mode to intelligently predict whether the current user will complete the purchase of the target insurance product within a preset time length after the current moment according to a number of targeted basic information, and pour more sales resources into attracting customers and guiding order placement for the current users who are predicted to complete the purchase of the target insurance product within a preset time length after the current moment, so as to avoid the loss of users with order placement tendencies and improve the statistical efficiency of future sales data of online insurance products.

[0007] According to one aspect of the present invention, there is provided an online insurance product sales system based on artificial intelligence. The system includes:

[0008] A first capture device, configured to obtain the number of times that the current user browses the target insurance product marketing web page of a set insurance product dealer within a preset time length before the current moment, and the duration that the current user reads the target insurance product introduction document in the target insurance product marketing web page of the set insurance product dealer, and use them as multiple pieces of attention data of the current user;

[0009] A second capture device, configured to obtain the number of types of multiple insurance products simultaneously sold by a set insurance product dealer, the number of insured users corresponding to multiple insurance products respectively, the duration of obtaining multiple sales licenses corresponding to multiple insurance products respectively, and the usage duration of the longest using user of the target insurance product, so as to use them as each piece of configuration information corresponding to the set insurance product dealer;

[0010] A multiple training device, configured to perform multiple trainings on a Hoffit neural network to obtain the Hoffit neural network after multiple trainings, and output the Hoffit neural network after multiple trainings as an AI sales prediction model;

[0011] A sales prediction device, connected to the first capture device, the second capture device, and the multiple training device respectively, configured to use the AI sales prediction model to intelligently predict the insurance purchase identifier of the current user for the target insurance product within a preset time length after the current moment according to the past insurance purchase identifier of the target insurance product of the current user, the age information of the current user, the gender information of the current user, the preset time length, the multiple pieces of attention data of the current user, and each piece of configuration information corresponding to the set insurance product dealer;

[0012] Wherein, performing multiple trainings on the Hoffit neural network to obtain the Hoffit neural network after multiple trainings, and outputting the Hoffit neural network after multiple trainings as an AI sales prediction model includes: the number of trainings performed on the Hoffit neural network is positively correlated with the number of insured users of the target insurance product and positively correlated with the usage duration of the longest using user of the target insurance product.

[0013] According to another aspect of the present invention, there is provided an online insurance product sales method based on artificial intelligence, the method comprising:

[0014] Obtaining the number of times that the current user browses the target insurance product marketing web page of a set insurance product dealer within a preset time length before the current moment, and the duration that the current user reads the target insurance product introduction document in the target insurance product marketing web page of the set insurance product dealer, and using them as multiple pieces of attention data of the current user;

[0015] Obtain the number of types of multiple insurance products sold by a designated insurance product dealer at the same time, the number of insured users corresponding to multiple insurance products respectively, the acquisition duration of multiple sales licenses corresponding to multiple insurance products respectively, and the usage duration of the longest-serving user of the target insurance product, so as to serve as each piece of configuration information corresponding to the designated insurance product dealer;

[0016] Perform multiple trainings on the Hoffit neural network to obtain the Hoffit neural network after multiple trainings, and output the Hoffit neural network after multiple trainings as an AI sales prediction model;

[0017] Use the AI sales prediction model to intelligently predict the insurance purchase identifier of the current user for the target insurance product within a preset time length after the current moment based on the past insurance purchase identifier of the target insurance product of the current user, the age information of the current user, the gender information of the current user, the preset time length, multiple pieces of concern data of the current user, and each piece of configuration information corresponding to the designated insurance product dealer;

[0018] Among them, performing multiple trainings on the Hoffit neural network to obtain the Hoffit neural network after multiple trainings, and outputting the Hoffit neural network after multiple trainings as an AI sales prediction model includes: the number of trainings performed on the Hoffit neural network is positively correlated with the number of insured users of the target insurance product and positively correlated with the usage duration of the longest-serving user of the target insurance product at the same time.

[0019] Thus, it can be seen that the present invention has at least the following four prominent substantive features:

[0020] Substantive feature one: Provide an artificial intelligence solution for the intelligent prediction of whether the current user of a designated insurance product dealer will complete the insurance purchase of the target insurance product within a preset time length after the current moment, so as to facilitate the statistics of the sales data of the target insurance product and ensure that the intelligent prediction focuses on the current users with an insurance purchase tendency to place an order for insurance purchase. While effectively attracting customers for online insurance products, improve the statistical efficiency of future sales data of online insurance products;

[0021] Substantive feature two: The intelligent prediction of the insurance purchase of the target insurance product is based on an artificial intelligence model with a customized structure. The artificial intelligence model is an AI sales prediction model. Specifically, the AI sales prediction model is the Hoffit neural network after multiple trainings, and the number of trainings performed on the Hoffit neural network is positively correlated with the number of insured users of the target insurance product and positively correlated with the usage duration of the longest-serving user of the target insurance product at the same time, so as to construct AI sales prediction models with different structures for different insurance products;

[0022] Substantive Feature Three: The intelligent prediction of the target insurance product application is based on multiple pieces of basic information selected specifically. The multiple pieces of basic information include the past application identification of the target insurance product of the current user, the age information of the current user, the gender information of the current user, the preset time length, multiple pieces of concern data of the current user, and each piece of configuration information corresponding to the set insurance product dealer. Specifically, the multiple pieces of concern data of the current user include the number of views of the target insurance product marketing web page of the set insurance product dealer by the current user within the preset time length before the current moment and the duration of the current user reading the target insurance product introduction document in the target insurance product marketing web page of the set insurance product dealer. Each piece of configuration information corresponding to the set insurance product dealer includes the number of types of multiple insurance products sold simultaneously by the set insurance product dealer, the number of insured users corresponding to multiple insurance products respectively, the duration of obtaining sales licenses corresponding to multiple insurance products respectively, and the usage duration of the longest-using user of the target insurance product. The full and comprehensive screening of the above multiple pieces of basic information ensures the stability and reliability of the intelligent prediction result;

[0023] Substantive Feature Four: In each training of the Hoffit neural network, the application identification of a certain user relative to the target insurance product within the preset time length after a known historical moment is used as the single output content of the Hoffit neural network, and the past application identification of the target insurance product of the certain user, the age information of the certain user, the gender information of the certain user, the preset time length, multiple pieces of concern data of the certain user, and each piece of configuration information corresponding to the set insurance product dealer are used as the multiple input contents of the Hoffit neural network to complete this training, thereby ensuring the training effect of each training of the Hoffit neural network. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The embodiments of the present invention will be described below in conjunction with the drawings, where:

[0025] Figure 1 FIG. is an internal structure diagram of an online insurance product sales system based on artificial intelligence shown according to the first embodiment of the present invention.

[0026] Figure 2 FIG. is an internal structure diagram of an online insurance product sales system based on artificial intelligence shown according to the second embodiment of the present invention.

[0027] Figure 3 FIG. is an internal structure diagram of an online insurance product sales system based on artificial intelligence shown according to the third embodiment of the present invention.

[0028] Figure 4 FIG. is an internal structure diagram of an online insurance product sales system based on artificial intelligence shown according to the fourth embodiment of the present invention.

[0029] Figure 5 Internal structure diagram of an online insurance product sales system based on artificial intelligence shown in the fifth embodiment of the present invention.

[0030] Figure 6 Step flowchart of an online insurance product sales method based on artificial intelligence shown in the sixth embodiment of the present invention. Detailed implementation manner

[0031] The specific technical process of the present invention is as follows:

[0032] Technical process A: Design an artificial intelligence model with a customized structure for intelligent prediction of whether the current user of an insurance product dealer will complete the insurance application for the target insurance product within a preset time length after the current moment. The artificial intelligence model is an AI sales prediction model;

[0033] Specifically, the customization of the structure of the AI sales prediction model is mainly manifested in the following aspects:

[0034] First: The AI sales prediction model is a Hopfield neural network after multiple trainings;

[0035] Second: The number of times of training performed on the Hopfield neural network is positively correlated with the number of insured users of the target insurance product and positively correlated with the usage duration of the longest - using user of the target insurance product, so as to construct AI sales prediction models with different structures for different insurance products;

[0036] Finally: In each training performed on the Hopfield neural network, the insurance application identifier of a certain user within a preset time length after a known historical moment with respect to the target insurance product is used as the single - item output content of the Hopfield neural network, and the past insurance application identifier of the target insurance product of the certain user, the age information of the certain user, the gender information of the certain user, the preset time length, the multiple pieces of attention data of the certain user, and each piece of configuration information corresponding to the set insurance product dealer are used as the multiple - item input content of the Hopfield neural network to complete this training, thus ensuring the training effect of each training of the Hopfield neural network;

[0037] Specifically, the preset time length is used to represent the length of a time period. Taking the current moment as the boundary, the past time interval corresponding to the preset time length before the current moment is a past time period, and the future time interval corresponding to the preset time length after the current moment is a future time period. In this way, dividing the time axis into each time period can facilitate the intelligent prediction of sequential insurance application identifiers;

[0038] Technical Process B: To intelligently predict whether the current user of an insurance product dealer will complete the purchase of a target insurance product within a preset time length after the current moment, and to selectively screen a number of basic information items;

[0039] Exemplarily, the number of basic information items includes the past insurance purchase identifier of the target insurance product of the current user, the age information of the current user, the gender information of the current user, the preset time length, multiple pieces of concern data of the current user, and each piece of configuration information corresponding to the set insurance product dealer;

[0040] Further exemplarily, multiple pieces of concern data of the current user include the number of views of the marketing page of the target insurance product of the set insurance product dealer by the current user within a preset time length before the current moment and the duration of the current user reading the target insurance product introduction document on the marketing page of the target insurance product of the set insurance product dealer. Each piece of configuration information corresponding to the set insurance product dealer includes the number of types of multiple insurance products sold simultaneously by the set insurance product dealer, the number of insured users corresponding to multiple insurance products respectively, the duration of obtaining multiple sales licenses corresponding to multiple insurance products respectively, and the usage duration of the longest - using user of the target insurance product;

[0041] For example, an insurance product dealer has three link icons on its homepage, namely "auto insurance", "life insurance", and "property insurance". Each of "auto insurance", "life insurance", and "property insurance" has its own marketing page. Any one of "auto insurance", "life insurance", and "property insurance" can be used as the target insurance product;

[0042] Here, through the full and comprehensive screening of the above - mentioned multiple basic information items, the stability and reliability of the intelligent prediction results are ensured;

[0043] Technical Process C: The AI sales prediction model designed with the customized structure of Technical Process A intelligently predicts whether the current user of the set insurance product dealer will complete the purchase of the target insurance product within a preset time length after the current moment based on the multiple basic information items fully and comprehensively screened by Technical Process C;

[0044] It should be noted that even if it is intelligently predicted that the current user will complete the purchase of the target insurance product within a preset time length after the current moment, this is only an intelligent prediction result and not an actual fact. It only indicates that the current user is a key target user with an order - placing tendency in the future time interval and has the value of key order - placing guidance;

[0045] Technical Process D: When the intelligent prediction in Technical Process C determines that the current user will complete the purchase of the target insurance product within a preset time length after the current moment, the insurance product dealer who sells the target insurance product will regard the current user as a key target user with a tendency to place an order, and pour more sales resources on the current user within a future time interval to ensure that the current user actually places an order and purchases the insurance product within the future time interval;

[0046] Specifically, the insurance product dealer who sells the target insurance product may concurrently sell multiple insurance products including the target insurance product, and all of the multiple insurance products are sold through online channels;

[0047] In this way, through the sequential execution of the above-mentioned technical processes, an intelligent prediction can be made in an artificial intelligence mode as to whether each user of the insurance product dealer will complete the purchase of the target insurance product within a preset time length after the current moment. This provides convenience for the statistics of the sales data of the target insurance product and for ensuring that the key target users with a predicted tendency to purchase insurance actually place orders by concentrating sales resources subsequently. While effectively attracting customers for online insurance products, it improves the statistical efficiency of future sales data of online insurance products.

[0048] The key points of the present invention are: introducing multiple basic information for targeted screening for the intelligent prediction of whether each user of the insurance product dealer will complete the purchase of the target insurance product within a future time interval, custom-designing AI sales prediction models with different structures for different insurance products, and focusing on pouring sales resources on key target users with a tendency to purchase insurance.

[0049] Next, a sales system and method for online insurance products based on artificial intelligence according to the present invention will be specifically described by way of embodiments.

[0050] First Embodiment

[0051] Figure 1 The internal structure diagram of an online insurance product sales system based on artificial intelligence shown according to the first embodiment of the present invention.

[0052] As Figure 1 shown, the online insurance product sales system based on artificial intelligence includes the following components:

[0053] A first capture device, which is used to obtain the number of times the current user browses the marketing web page of the target insurance product of the set insurance product dealer within a preset time length before the current moment and the duration of the current user reading the target insurance product introduction document in the marketing web page of the target insurance product of the set insurance product dealer, and use them as multiple pieces of attention data of the current user;

[0054] Exemplarily, obtain the number of views of the target insurance product marketing web page of the set insurance product dealer by the current user within a preset time length before the current moment, and the duration of the current user reading the target insurance product introduction document in the target insurance product marketing web page of the set insurance product dealer, and use them as multiple pieces of attention data of the current user, including: the target insurance product introduction document in the target insurance product marketing web page can be an electronic document embedded in the target insurance product marketing web page and needs to be opened and displayed under the click of the current user;

[0055] A second capture device, configured to obtain the number of types of multiple insurance products sold by the set insurance product dealer at the same time, multiple pieces of in-force user numbers respectively corresponding to the multiple insurance products, multiple pieces of sales license acquisition durations respectively corresponding to the multiple insurance products, and the usage duration of the longest-using user of the target insurance product, so as to use them as respective pieces of configuration information corresponding to the set insurance product dealer;

[0056] Specifically, obtaining the number of types of multiple insurance products sold by the set insurance product dealer at the same time, multiple pieces of in-force user numbers respectively corresponding to the multiple insurance products, multiple pieces of sales license acquisition durations respectively corresponding to the multiple insurance products, and the usage duration of the longest-using user of the target insurance product, so as to use them as respective pieces of configuration information corresponding to the set insurance product dealer includes: multiple information capture components can be used to respectively obtain the number of types of multiple insurance products sold by the set insurance product dealer at the same time, multiple pieces of in-force user numbers respectively corresponding to the multiple insurance products, multiple pieces of sales license acquisition durations respectively corresponding to the multiple insurance products, and the usage duration of the longest-using user of the target insurance product;

[0057] Exemplarily, multiple information capture components can be used to respectively obtain the number of types of multiple insurance products sold by the set insurance product dealer at the same time, multiple pieces of in-force user numbers respectively corresponding to the multiple insurance products, multiple pieces of sales license acquisition durations respectively corresponding to the multiple insurance products, and the usage duration of the longest-using user of the target insurance product, including: both the sales license acquisition duration and the usage duration of the longest-using user can be represented in the mode of days;

[0058] A multiple training device, configured to perform multiple trainings on the Hoffit neural network to obtain the Hoffit neural network after multiple trainings, and output the Hoffit neural network after multiple trainings as an AI sales prediction model;

[0059] Exemplarily, the Hofit neural network is trained multiple times to obtain the Hofit neural network after multiple trainings, and the Hofit neural network after multiple trainings is used as the AI sales prediction model to output, including: The process of testing and simulating the model construction that can be completed by using the numerical simulation mode to train the Hofit neural network multiple times to obtain the Hofit neural network after multiple trainings, and using the Hofit neural network after multiple trainings as the AI sales prediction model to output;

[0060] The sales prediction device is respectively connected to the first capture device, the second capture device and the multiple training devices, and is used to use the AI sales prediction model to intelligently predict the insurance purchase identifier of the current user for the target insurance product within a preset time length after the current moment according to the past insurance purchase identifier of the target insurance product of the current user, the age information of the current user, the gender information of the current user, the preset time length, multiple pieces of concern data of the current user, and each piece of configuration information corresponding to the set insurance product dealer;

[0061] Specifically, the preset time length is used to represent the length of a time period. Taking the current moment as the boundary, the past time interval corresponding to the preset time length before the current moment is a past time period, and the future time interval corresponding to the preset time length after the current moment is a future time period. In this way, dividing the time axis into each time period can facilitate the intelligent prediction of the sequential insurance purchase identifier.

[0062] Among them, training the Hofit neural network multiple times to obtain the Hofit neural network after multiple trainings, and using the Hofit neural network after multiple trainings as the AI sales prediction model to output, including: The number of times of training the Hofit neural network is positively correlated with the number of insured users of the target insurance product and is positively correlated with the usage duration of the longest used user of the target insurance product;

[0063] Exemplarily, the number of times of training the Hofit neural network is positively correlated with the number of insured users of the target insurance product and is positively correlated with the usage duration of the longest used user of the target insurance product, including: When the number of insured users of the target insurance product is in the millions and the usage duration of the longest used user of the target insurance product is 5 years, the number of times of training the Hofit neural network is 500; when the number of insured users of the target insurance product is in the millions and the usage duration of the longest used user of the target insurance product is 6 years, the number of times of training the Hofit neural network is 550; when the number of insured users of the target insurance product is in the half millions and the usage duration of the longest used user of the target insurance product is 5 years, the number of times of training the Hofit neural network is 600, and so on;

[0064] Among them, obtaining the number of views of the target insurance product marketing web page of the set insurance product dealer by the current user within a preset time length before the current moment and the duration of the current user reading the target insurance product introduction document in the target insurance product marketing web page of the set insurance product dealer, and taking them as multiple pieces of concern data of the current user includes: The set insurance product dealer sells multiple insurance products including the target insurance product at the same time, and each of the multiple insurance products has its own marketing web page;

[0065] Specifically, the set insurance product dealer sells multiple insurance products including the target insurance product at the same time, and each of the multiple insurance products having its own marketing web page includes: Each of the marketing web pages of the multiple insurance products has its own link icon on the home page of the set insurance product dealer;

[0066] Exemplarily, the set insurance product dealer has three link icons on its home page, namely "auto insurance", "life insurance" and "property insurance", and "auto insurance", "life insurance" and "property insurance" each have their own marketing web pages;

[0067] Among them, using the AI sales prediction model to intelligently predict the insurance purchase identifier of the current user for the target insurance product within a preset time length after the current moment based on the past insurance purchase identifier of the target insurance product of the current user, the age information of the current user, the gender information of the current user, the preset time length, the multiple pieces of concern data of the current user, and each piece of configuration information corresponding to the set insurance product dealer includes: The insurance purchase identifier of the current user for the target insurance product within a preset time length after the current moment is used to indicate whether the current user has placed an order to purchase the target insurance product through the target insurance product marketing web page within a preset time length before the current moment;

[0068] Among them, performing multiple trainings on the Hofit neural network to obtain the Hofit neural network after multiple trainings, and taking the Hofit neural network after multiple trainings as the output of the AI sales prediction model further includes: In each training performed on the Hofit neural network, taking the insurance purchase identifier of a certain user for the target insurance product within a preset time length after a known historical moment as the single output content of the Hofit neural network, and taking the past insurance purchase identifier of the target insurance product of the certain user, the age information of the certain user, the gender information of the certain user, the preset time length, the multiple pieces of concern data of the certain user, and each piece of configuration information corresponding to the set insurance product dealer as the multiple input contents of the Hofit neural network to complete this training;

[0069] And among them, in each training of the Hoffit neural network, the insurance purchase identifier of a certain user relative to the target insurance product within a preset time length after a certain known historical moment is used as the single output content of the Hoffit neural network. The past insurance purchase identifiers of the target insurance product of the certain user, the age information of the certain user, the gender information of the certain user, the preset time length, the multiple pieces of concern data of the certain user, and each piece of configuration information corresponding to the set insurance product dealer are used as the multiple input contents of the Hoffit neural network. Completing this training includes: the certain user is a user of the set insurance product dealer, and the past insurance purchase identifier of the target insurance product of the certain user is used to indicate whether the certain user placed an order to purchase the target insurance product through the target insurance product marketing webpage within a preset time length before a certain historical moment.

[0070] Second Embodiment

[0071] Figure 2 It is an internal structure diagram of an online insurance product sales system based on artificial intelligence shown in the second embodiment of the present invention.

[0072] As Figure 2 shown, compared with Figure 1 , the online insurance product sales system based on artificial intelligence further includes:

[0073] A resource allocation device, connected to the sales prediction device, is configured to allocate more sales resources to the current user within a preset time length after the current moment when the insurance purchase identifier of the current user relative to the target insurance product within a preset time length after the received current moment indicates that the current user will place an order to purchase the target insurance product through the target insurance product marketing webpage within a preset time length after the current moment;

[0074] Among them, when the insurance purchase identifier of the current user relative to the target insurance product within a preset time length after the received current moment indicates that the current user will place an order to purchase the target insurance product through the target insurance product marketing webpage within a preset time length after the current moment, allocating more sales resources to the current user within a preset time length after the current moment includes: the number of sales resources allocated to the current user within a preset time length after the current moment is greater than the default number of sales resources allocated by the set insurance product dealer to each user;

[0075] Among them, when the insurance purchase identifier of the current user with respect to the target insurance product within a preset time length after the received current moment indicates that the current user will place an order to purchase the target insurance product through the target insurance product marketing webpage within the preset time length after the current moment, allocating more sales resources to the current user within the preset time length after the current moment further includes: The sales resources are the advertising placement frequency of the target insurance product and / or the placement quantity of the salespersons of the target insurance product.

[0076] The third embodiment

[0077] Figure 3 It is an internal structure diagram of an online insurance product sales system based on artificial intelligence shown in the third embodiment of the present invention.

[0078] As Figure 3 shown, compared with Figure 2 , the online insurance product sales system based on artificial intelligence further includes:

[0079] A model processing device, connected to the multiple training devices, for receiving the AI sales prediction model and completing the model storage of the AI sales prediction model through various model data of the AI sales prediction model.

[0080] The fourth embodiment

[0081] Figure 4 It is an internal structure diagram of an online insurance product sales system based on artificial intelligence shown in the fourth embodiment of the present invention.

[0082] As Figure 4 shown, compared with Figure 3 , the online insurance product sales system based on artificial intelligence further includes:

[0083] A content display device, connected to the sales prediction device, for receiving the insurance purchase identifier of the current user with respect to the target insurance product within a preset time length after the current moment and displaying in real time the insurance purchase identifier of the current user with respect to the target insurance product within a preset time length after the current moment.

[0084] The fifth embodiment

[0085] Figure 5 It is an internal structure diagram of an online insurance product sales system based on artificial intelligence shown in the fifth embodiment of the present invention.

[0086] As Figure 5 shown, compared with Figure 4 , the online insurance product sales system based on artificial intelligence further includes:

[0087] A wireless transmission device, connected to a sales prediction device, is used to receive the insurance application identification of the current user relative to the target insurance product within a preset time length after the current moment, and wirelessly transmit it to the big data storage network element of the designated insurance product dealer through a wireless transmission link.

[0088] Next, further descriptions will be continued for various embodiments of the present invention.

[0089] Optionally, in the above-mentioned various embodiments, in the online insurance product sales system based on artificial intelligence:

[0090] Obtaining the type quantity of multiple insurance products simultaneously sold by the designated insurance product dealer, the quantity of multiple insured users respectively corresponding to the multiple insurance products, the acquisition duration of multiple sales licenses respectively corresponding to the multiple insurance products, and the usage duration of the longest-using user of the target insurance product, to be used as each piece of configuration information corresponding to the designated insurance product dealer includes: the acquisition durations of multiple sales licenses respectively corresponding to the multiple insurance products simultaneously sold by the designated insurance product dealer are all numerically represented in days;

[0091] Among them, using the AI sales prediction model to intelligently predict the insurance application identification of the current user relative to the target insurance product within a preset time length after the current moment according to the past insurance application identification of the target insurance product of the current user, the age information of the current user, the gender information of the current user, the preset time length, multiple pieces of concern data of the current user, and each piece of configuration information corresponding to the designated insurance product dealer further includes: parallelly inputting the past insurance application identification of the target insurance product of the current user, the age information of the current user, the gender information of the current user, the preset time length, multiple pieces of concern data of the current user, and each piece of configuration information corresponding to the designated insurance product dealer into the AI sales prediction model;

[0092] Among them, parallelly inputting the past insurance application identification of the target insurance product of the current user, the age information of the current user, the gender information of the current user, the preset time length, multiple pieces of concern data of the current user, and each piece of configuration information corresponding to the designated insurance product dealer into the AI sales prediction model includes: respectively performing numerical normalization processing on the past insurance application identification of the target insurance product of the current user, the age information of the current user, the gender information of the current user, the preset time length, multiple pieces of concern data of the current user, and each piece of configuration information corresponding to the designated insurance product dealer, and then parallelly inputting them into the AI sales prediction model;

[0093] Among them, using the AI sales prediction model to intelligently predict the insurance purchase identifier of the current user for the target insurance product within a preset time length after the current moment based on the past insurance purchase identifiers of the target insurance product of the current user, the age information of the current user, the gender information of the current user, the preset time length, multiple pieces of the current user's concerned data, and each piece of configuration information corresponding to the set insurance product dealer further includes: running the AI sales prediction model to obtain the insurance purchase identifier of the current user for the target insurance product within a preset time length after the current moment output by the AI sales prediction model;

[0094] And among them, running the AI sales prediction model to obtain the insurance purchase identifier of the current user for the target insurance product within a preset time length after the current moment includes: the insurance purchase identifier of the current user for the target insurance product within a preset time length after the current moment is in a numerically normalized numerical representation form.

[0095] And in each of the above embodiments, optionally, in the online insurance product sales system based on artificial intelligence:

[0096] The number of times of training performed on the Hoffit neural network is positively correlated with the number of insured users of the target insurance product and positively correlated with the usage duration of the longest - using user of the target insurance product, which includes: using a double - input single - output numerical mapping function to represent the numerical mapping relationship between the number of insured users of the target insurance product and the usage duration of the longest - using user of the target insurance product together with the number of times of training performed on the Hoffit neural network;

[0097] Among them, using a double - input single - output numerical mapping function to represent the numerical mapping relationship between the number of insured users of the target insurance product and the usage duration of the longest - using user of the target insurance product together with the number of times of training performed on the Hoffit neural network includes: taking the number of insured users of the target insurance product and the usage duration of the longest - using user of the target insurance product as the double - input numerical values of the numerical mapping function;

[0098] And among them, using a double - input single - output numerical mapping function to represent the numerical mapping relationship between the number of insured users of the target insurance product and the usage duration of the longest - using user of the target insurance product together with the number of times of training performed on the Hoffit neural network further includes: taking the number of times of training performed on the Hoffit neural network as the single - output numerical value of the numerical mapping function.

[0099] Sixth Embodiment

[0100] Figure 6 It is a step - flow chart of an online insurance product sales method based on artificial intelligence shown in the sixth embodiment of the present invention.

[0101] As Figure 6As shown, the method for selling online insurance products based on artificial intelligence includes the following steps:

[0102] Obtain the number of views of the target insurance product marketing page of the set insurance product dealer by the current user within a preset time length before the current moment, and the duration of the current user reading the target insurance product introduction document in the target insurance product marketing page of the set insurance product dealer, and use them as multiple pieces of concern data of the current user;

[0103] Exemplarily, obtaining the number of views of the target insurance product marketing page of the set insurance product dealer by the current user within a preset time length before the current moment, and the duration of the current user reading the target insurance product introduction document in the target insurance product marketing page of the set insurance product dealer, and using them as multiple pieces of concern data of the current user includes: the target insurance product introduction document in the target insurance product marketing page can be an electronic document embedded in the target insurance product marketing page, and needs to be opened and displayed under the click of the current user;

[0104] Obtain the number of types of multiple insurance products sold by the set insurance product dealer at the same time, the number of multiple in-force users corresponding to the multiple insurance products respectively, the duration of obtaining multiple sales licenses corresponding to the multiple insurance products respectively, and the usage duration of the longest user of the target insurance product, so as to use them as each piece of configuration information corresponding to the set insurance product dealer;

[0105] Specifically, obtaining the number of types of multiple insurance products sold by the set insurance product dealer at the same time, the number of multiple in-force users corresponding to the multiple insurance products respectively, the duration of obtaining multiple sales licenses corresponding to the multiple insurance products respectively, and the usage duration of the longest user of the target insurance product, so as to use them as each piece of configuration information corresponding to the set insurance product dealer includes: multiple information capture components can be used to respectively obtain the number of types of multiple insurance products sold by the set insurance product dealer at the same time, the number of multiple in-force users corresponding to the multiple insurance products respectively, the duration of obtaining multiple sales licenses corresponding to the multiple insurance products respectively, and the usage duration of the longest user of the target insurance product;

[0106] Exemplarily, multiple information capture components can be used to respectively obtain the number of types of multiple insurance products sold by the set insurance product dealer at the same time, the number of multiple in-force users corresponding to the multiple insurance products respectively, the duration of obtaining multiple sales licenses corresponding to the multiple insurance products respectively, and the usage duration of the longest user of the target insurance product includes: both the duration of obtaining the sales license and the usage duration of the longest user can be represented in the mode of days;

[0107] Perform multiple trainings on the Hoffit neural network to obtain the Hoffit neural network after multiple trainings, and output the Hoffit neural network after multiple trainings as an AI sales prediction model;

[0108] Exemplarily, performing multiple trainings on the Hoffit neural network to obtain the Hoffit neural network after multiple trainings, and outputting the Hoffit neural network after multiple trainings as an AI sales prediction model includes: The process of testing and simulating the model construction of performing multiple trainings on the Hoffit neural network to obtain the Hoffit neural network after multiple trainings, and outputting the Hoffit neural network after multiple trainings as an AI sales prediction model can be completed in a numerical simulation mode;

[0109] Use the AI sales prediction model to intelligently predict the insurance purchase identifier of the current user for the target insurance product within a preset time length after the current moment based on the past insurance purchase identifiers of the target insurance product of the current user, the age information of the current user, the gender information of the current user, the preset time length, multiple pieces of concern data of the current user, and each piece of configuration information corresponding to the set insurance product dealer;

[0110] Specifically, the preset time length is used to represent the length of a time period. Taking the current moment as the boundary, the past time interval corresponding to the preset time length before the current moment is a past time period, and the future time interval corresponding to the preset time length after the current moment is a future time period. In this way, dividing the time axis into each time period can facilitate the intelligent prediction of the sequential insurance purchase identifiers;

[0111] Among them, performing multiple trainings on the Hoffit neural network to obtain the Hoffit neural network after multiple trainings, and outputting the Hoffit neural network after multiple trainings as an AI sales prediction model includes: The number of trainings performed on the Hoffit neural network is positively correlated with the number of insured users of the target insurance product and positively correlated with the usage duration of the longest - using user of the target insurance product;

[0112] Exemplarily, the number of trainings performed on the Hoffit neural network is positively correlated with the number of insured users of the target insurance product and positively correlated with the usage duration of the longest - using user of the target insurance product includes: When the number of insured users of the target insurance product is in the millions and the usage duration of the longest - using user of the target insurance product is 5 years, the number of trainings performed on the Hoffit neural network is 500; when the number of insured users of the target insurance product is in the millions and the usage duration of the longest - using user of the target insurance product is 6 years, the number of trainings performed on the Hoffit neural network is 550; when the number of insured users of the target insurance product is in the half - millions and the usage duration of the longest - using user of the target insurance product is 5 years, the number of trainings performed on the Hoffit neural network is 600, and so on;

[0113] Among them, obtaining the number of views of the target insurance product marketing web page of the set insurance product dealer by the current user within a preset time length before the current moment and the duration of the current user reading the target insurance product introduction document in the target insurance product marketing web page of the set insurance product dealer, and taking them as multiple pieces of concern data of the current user includes: the set insurance product dealer sells multiple insurance products including the target insurance product at the same time, and the multiple insurance products each have their own marketing web pages;

[0114] Specifically, the set insurance product dealer sells multiple insurance products including the target insurance product at the same time, and the multiple insurance products each have their own marketing web pages includes: each of the marketing web pages of the multiple insurance products has its own link icon on the home page of the set insurance product dealer;

[0115] Exemplarily, the set insurance product dealer has three link icons on its home page, namely "auto insurance", "life insurance" and "property insurance", and "auto insurance", "life insurance" and "property insurance" each have their own marketing web pages;

[0116] Among them, using the AI sales prediction model to intelligently predict the insurance purchase identifier of the current user for the target insurance product within a preset time length after the current moment based on the past insurance purchase identifier of the target insurance product of the current user, the age information of the current user, the gender information of the current user, the preset time length, the multiple pieces of concern data of the current user, and each piece of configuration information corresponding to the set insurance product dealer includes: the insurance purchase identifier of the current user for the target insurance product within a preset time length after the current moment is used to indicate whether the current user placed an order to purchase the target insurance product through the target insurance product marketing web page within a preset time length before the current moment;

[0117] Among them, performing multiple trainings on the Hoffit neural network to obtain the Hoffit neural network after multiple trainings, and taking the Hoffit neural network after multiple trainings as the output of the AI sales prediction model further includes: in each training performed on the Hoffit neural network, taking the insurance purchase identifier of a certain user for the target insurance product within a preset time length after a known historical moment as the single output content of the Hoffit neural network, and taking the past insurance purchase identifier of the target insurance product of the certain user, the age information of the certain user, the gender information of the certain user, the preset time length, the multiple pieces of concern data of the certain user, and each piece of configuration information corresponding to the set insurance product dealer as the multiple input contents of the Hoffit neural network to complete this training;

[0118] And among them, in each training performed on the Hoffit neural network, the insurance purchase identifier of a certain user with respect to the target insurance product within a preset time length after a certain known historical moment is used as the single output content of the Hoffit neural network. The past insurance purchase identifiers of the target insurance product of the certain user, the age information of the certain user, the gender information of the certain user, the preset time length, multiple pieces of concern data of the certain user, and each piece of configuration information corresponding to the set insurance product dealer are used as the multiple input contents of the Hoffit neural network. Completing this training includes: the certain user is a user of the set insurance product dealer, and the past insurance purchase identifier of the target insurance product of the certain user is used to indicate whether the certain user placed an order to purchase the target insurance product through the target insurance product marketing webpage within a preset time length before a certain historical moment.

[0119] In addition, in an online insurance product sales system and method based on artificial intelligence according to the present invention:

[0120] Performing numerical normalization processing on the past insurance purchase identifier of the target insurance product of the current user, the age information of the current user, the gender information of the current user, the preset time length, multiple pieces of concern data of the current user, and each piece of configuration information corresponding to the set insurance product dealer respectively and then parallelly inputting them into the AI sales prediction model includes: performing binary numerical conversion processing on the past insurance purchase identifier of the target insurance product of the current user, the age information of the current user, the gender information of the current user, the preset time length, multiple pieces of concern data of the current user, and each piece of configuration information corresponding to the set insurance product dealer respectively and then parallelly inputting them into the AI sales prediction model;

[0121] Among them, performing numerical normalization processing on the past insurance purchase identifier of the target insurance product of the current user, the age information of the current user, the gender information of the current user, the preset time length, multiple pieces of concern data of the current user, and each piece of configuration information corresponding to the set insurance product dealer respectively and then parallelly inputting them into the AI sales prediction model further includes: using a parallel control device to complete the parallel input of the past insurance purchase identifier of the target insurance product of the current user, the age information of the current user, the gender information of the current user, the preset time length, multiple pieces of concern data of the current user, and each piece of configuration information corresponding to the set insurance product dealer that have respectively performed numerical normalization processing into the AI sales prediction model;

[0122] Exemplarily, the parallel input of the past insurance purchase identifier of the target insurance product of the current user, the age information of the current user, the gender information of the current user, the preset time length, multiple pieces of concern data of the current user, and each piece of configuration information corresponding to the set insurance product dealer to the AI sales prediction model completed by using a parallel control device includes: the parallel control device is a programmable logic device designed by using the VHDL language;

[0123] And wherein, the insurance purchase identifier of the current user relative to the target insurance product within the preset time length after the current moment in the form of a numerically normalized numerical representation includes: the insurance purchase identifier of the current user relative to the target insurance product within the preset time length after the current moment in the form of a binary numerical representation.

[0124] It should be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "including one..." does not exclude the existence of additional identical elements in the process, method, article or device including the said element.

[0125] Each embodiment in this specification is described in a related manner. The same or similar parts between the embodiments can be referred to each other, and the differences between each embodiment and other embodiments are emphasized. In particular, for the device / electronic device / computer-readable storage medium / computer program product embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can refer to the partial description of the method embodiments. The above is only the preferred embodiment of the present invention and is not used to limit the protection scope of the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention are included in the protection scope of the present invention.

Claims

1. An online insurance product sales system based on artificial intelligence, characterized in that: The system comprises: The first capturing device is used to obtain the number of views of the target insurance product marketing webpage of the set insurance product dealer by the current user within a preset time length before the current moment and the length of time the current user reads the target insurance product introduction document on the target insurance product marketing webpage of the set insurance product dealer, and use them as multiple sets of attention data of the current user; The second capture device is used to obtain the number of types of multiple insurance products sold simultaneously by the set insurance product dealer, the number of multiple insured users corresponding to the multiple insurance products, the acquisition time of multiple sales licenses corresponding to the multiple insurance products, and the usage time of the longest user of the target insurance product, so as to set the configuration information corresponding to the insurance product dealer; A multiple training device is used to perform multiple training on the Hoffet neural network to obtain the Hoffet neural network after the multiple training, and output the Hoffet neural network after the multiple training as an AI sales prediction model; The sales prediction device is connected to the first capture device, the second capture device and the multiple training devices respectively, and is used to use an AI sales prediction model to intelligently predict the insurance identification of the current user relative to the target insurance product within a preset time length after the current moment based on the past insurance identification of the target insurance product of the current user, the age information of the current user, the gender information of the current user, the preset time length, multiple pieces of attention data of the current user, and the configuration information corresponding to the insurance product dealer; Among them, performing multiple training on the Hoffitt neural network to obtain the Hoffitt neural network after multiple trainings, and outputting the Hoffitt neural network after multiple trainings as the AI ​​sales prediction model includes: the number of trainings performed on the Hoffitt neural network is positively correlated with the number of insured users of the target insurance product and is positively correlated with the usage time of the longest users of the target insurance product.

2. The artificial intelligence-based online insurance product sales system according to claim 1, characterized in that: Obtaining the number of views of the target insurance product marketing webpage of the set insurance product dealer by the current user within a preset time length before the current moment and the time length for the current user to read the target insurance product introduction document in the target insurance product marketing webpage of the set insurance product dealer, and including as the multiple pieces of attention data of the current user: the set insurance product dealer sells multiple insurance products including the target insurance product at the same time, and the multiple insurance products respectively have their own marketing webpages; Among them, an AI sales prediction model is used to intelligently predict the insurance identification of the current user relative to the target insurance product within a preset time length after the current moment based on the past insurance identification of the current user's target insurance product, the current user's age information, the current user's gender information, the preset time length, multiple attention data of the current user, and the configuration information corresponding to the insurance product dealer. The insurance identification includes: the insurance identification of the current user relative to the target insurance product within the preset time length after the current moment is used to indicate whether the current user has placed an order to insure the target insurance product through the target insurance product marketing webpage within the preset time length before the current moment.

3. The online insurance product sales system based on artificial intelligence as claimed in claim 2, characterized in that: Performing multiple training on the Hoffitt neural network to obtain the Hoffitt neural network after multiple training, and outputting the Hoffitt neural network after multiple training as the AI ​​sales prediction model also includes: in each training performed on the Hoffitt neural network, using the insurance identification of a certain user with respect to the target insurance product within a preset time length after a certain historical moment as a single output content of the Hoffitt neural network, using the past insurance identification of the target insurance product of the certain user, the age information of the certain user, the gender information of the certain user, the preset time length, multiple pieces of attention data of the certain user, and each piece of configuration information corresponding to the insurance product dealer as multiple input contents of the Hoffitt neural network to complete this training; Among them, in each training performed on the Hoffitt neural network, the insurance identification of a certain user with respect to the target insurance product within a preset time length after a known historical moment is used as a single output content of the Hoffitt neural network, and the past insurance identification of the target insurance product of the certain user, the age information of the certain user, the gender information of the certain user, the preset time length, multiple pieces of attention data of the certain user, and the configuration information corresponding to the set insurance product dealer are used as multiple input contents of the Hoffitt neural network. Completing this training includes: the certain user is a user who sets the insurance product dealer, and the past insurance identification of the target insurance product of the certain user is used to indicate whether the certain user has placed an order for the target insurance product through the target insurance product marketing webpage within the preset time length before a certain historical moment.

4. The online insurance product sales system based on artificial intelligence as claimed in claim 3, characterized in that: The system further comprises: a resource allocation device connected to the sales forecasting device, and configured to allocate more sales resources to the current user within a preset time length after the current moment when the insurance identification of the current user with respect to the target insurance product within a preset time length after the current moment indicates that the current user will place an order for the target insurance product through the marketing webpage of the target insurance product within a preset time length after the current moment; Wherein, when the insurance identification of the current user with respect to the target insurance product within the preset time length after the current moment received indicates that the current user will place an order to insure the target insurance product through the target insurance product marketing webpage within the preset time length after the current moment, allocating more sales resources to the current user within the preset time length after the current moment includes: the number of sales resources allocated to the current user within the preset time length after the current moment is greater than the default number of sales resources allocated by the insurance product dealer to each user; Among them, when the insurance identification of the current user with respect to the target insurance product within the preset time length after the current moment received indicates that the current user will place an order to insure the target insurance product through the target insurance product marketing webpage within the preset time length after the current moment, allocating more sales resources to the current user within the preset time length after the current moment also includes: the sales resources are the advertising frequency of the target insurance product and / or the number of sales personnel of the target insurance product.

5. The online insurance product sales system based on artificial intelligence as claimed in claim 3, characterized in that: The system further comprises: The model processing device is connected to multiple training devices, and is used to receive the AI ​​sales prediction model and complete the model storage of the AI ​​sales prediction model through various model data of the AI ​​sales prediction model.

6. The online insurance product sales system based on artificial intelligence as claimed in claim 3, characterized in that: The system further comprises: The content display device is connected to the sales forecasting device, and is used to receive the insurance identification of the current user relative to the target insurance product within a preset time length after the current moment, and display the insurance identification of the current user relative to the target insurance product within the preset time length after the current moment in real time.

7. The online insurance product sales system based on artificial intelligence as claimed in claim 3, characterized in that: The system further comprises: The wireless transmission device is connected to the sales forecasting device, and is used to receive the insurance identification of the current user relative to the target insurance product within a preset time length after the current moment, and wirelessly transmit it to the big data storage network element of the set insurance product dealer through a wireless transmission link.

8. The artificial intelligence-based online insurance product sales system according to any one of claims 3 to 7, characterized in that: The number of types of multiple insurance products sold simultaneously by the set insurance product dealer, the number of insured users corresponding to the multiple insurance products, the acquisition time of multiple sales licenses corresponding to the multiple insurance products, and the usage time of the longest user of the target insurance product are obtained as the configuration information corresponding to the set insurance product dealer, including: the acquisition time of multiple sales licenses corresponding to the multiple insurance products sold simultaneously by the set insurance product dealer are all expressed in days; Among them, using the AI ​​sales prediction model to intelligently predict the insurance identification of the current user relative to the target insurance product within a preset time length after the current moment based on the past insurance identification of the target insurance product of the current user, the age information of the current user, the gender information of the current user, the preset time length, multiple pieces of attention data of the current user, and each piece of configuration information corresponding to the set insurance product dealer also includes: inputting the past insurance identification of the target insurance product of the current user, the age information of the current user, the gender information of the current user, the preset time length, multiple pieces of attention data of the current user, and each piece of configuration information corresponding to the set insurance product dealer into the AI ​​sales prediction model in parallel; The step of inputting the past insurance identification of the target insurance product of the current user, the age information of the current user, the gender information of the current user, the preset time length, the multiple pieces of attention data of the current user, and the configuration information corresponding to the set insurance product dealer into the AI ​​sales prediction model in parallel includes: performing numerical normalization processing on the past insurance identification of the target insurance product of the current user, the age information of the current user, the gender information of the current user, the preset time length, the multiple pieces of attention data of the current user, and the configuration information corresponding to the set insurance product dealer, and then inputting them into the AI ​​sales prediction model in parallel; Among them, using the AI ​​sales prediction model to intelligently predict the insurance identification of the current user relative to the target insurance product within a preset time length after the current moment based on the past insurance identification of the target insurance product of the current user, the age information of the current user, the gender information of the current user, the preset time length, multiple pieces of attention data of the current user, and the configuration information corresponding to the insurance product dealer also includes: running the AI ​​sales prediction model to obtain the insurance identification of the current user relative to the target insurance product within the preset time length after the current moment output by the AI ​​sales prediction model; Among them, running the AI ​​sales prediction model to obtain the insurance identification of the current user relative to the target insurance product within a preset time length after the current moment output by the AI ​​sales prediction model includes: the insurance identification of the current user relative to the target insurance product within the preset time length after the current moment is a numerical representation of a normalized value.

9. The artificial intelligence-based online insurance product sales system according to any one of claims 3 to 7, characterized in that: The number of trainings performed on the Hoffet neural network is positively correlated with the number of insured users of the target insurance product and the usage time of the longest user of the target insurance product, including: using a double-input single-output numerical mapping function to represent the numerical mapping relationship between the number of insured users of the target insurance product and the usage time of the longest user of the target insurance product and the number of trainings performed on the Hoffet neural network; The use of a double-input single-output numerical mapping function to represent the numerical mapping relationship between the number of insured users of the target insurance product and the length of time the longest user of the target insurance product is used and the number of times the Hoffitt neural network is trained includes: using the number of insured users of the target insurance product and the length of time the longest user of the target insurance product is used as double input values ​​of the numerical mapping function; Among them, the use of a dual-input single-output numerical mapping function to represent the numerical mapping relationship between the number of insured users of the target insurance product and the usage time of the longest-user of the target insurance product and the number of training times performed on the Hoffitt neural network also includes: using the number of training times performed on the Hoffitt neural network as the single output value of the numerical mapping function.

10. An online insurance product sales method based on artificial intelligence, characterized in that: The method comprises: Obtain the number of views of the target insurance product marketing webpage of the set insurance product dealer by the current user within a preset time length before the current moment and the duration of the current user reading the target insurance product introduction document on the target insurance product marketing webpage of the set insurance product dealer, and use them as multiple sets of attention data of the current user; Obtain the number of types of multiple insurance products sold simultaneously by the set insurance product dealer, the number of multiple insured users corresponding to the multiple insurance products, the acquisition time of multiple sales licenses corresponding to the multiple insurance products, and the usage time of the longest user of the target insurance product, as the configuration information corresponding to the set insurance product dealer; Performing multiple training on the Hoffet neural network to obtain the Hoffet neural network after the multiple trainings, and outputting the Hoffet neural network after the multiple trainings as an AI sales prediction model; The AI ​​sales prediction model is used to intelligently predict the insurance identification of the current user relative to the target insurance product within a preset time length after the current moment based on the previous insurance identification of the current user's target insurance product, the current user's age information, the current user's gender information, the preset time length, multiple pieces of attention data of the current user, and the corresponding configuration information of the insurance product dealer; Among them, performing multiple training on the Hoffitt neural network to obtain the Hoffitt neural network after multiple trainings, and outputting the Hoffitt neural network after multiple trainings as the AI ​​sales prediction model includes: the number of trainings performed on the Hoffitt neural network is positively correlated with the number of insured users of the target insurance product and is positively correlated with the usage time of the longest users of the target insurance product.

Images