Insurance proposal generation method and device and readable storage medium

Abstract

The invention provides an insurance proposal generation method and device and a readable storage medium, and relates to the field of insurance, and the scheme comprises the steps: obtaining the customer portrait information of an insurance applicant, and predicting the target insurance product classification corresponding to a target object based on the customer portrait information and a product classification prediction model; determining a target main insurance product corresponding to the target object under the target insurance product classification based on the customer portrait information and a main insurance product prediction model, and determining a target additional product and target product configuration information associated with the target main insurance product based on the customer portrait information, the target main insurance product and preset association rule information, and generating an insurance proposal including the target main insurance product, the target additional insurance product and the target product configuration information. According to the method and the device, the insurance proposal which better meets the requirements of the target object and is more beneficial to transaction is gradually determined according to the portrait information of the insurance applicant, the method and the device are not limited to personal experience of insurance business personnel, the intelligent degree is high, and improvement of insurance product recommendation accuracy is facilitated.

Description

Technical Field

[0001] The present application relates to the field of insurance technology, and in particular to a method, device and readable storage medium for generating an insurance proposal. Background Art

[0002] An insurance proposal is a communication plan commonly used by insurance business personnel to communicate effectively with customers about insurance purchase and insurance management. It reflects the professional level of insurance services in written form. A reasonable and excellent insurance proposal is conducive to strengthening the customer's sense of trust and laying a good foundation for the next project operation.

[0003] Currently, the preparation method of insurance proposals mainly adopts the traditional model, that is, it is completely dependent on the personal experience of insurance sales personnel. However, due to personal experience, insurance sales personnel sometimes do not know which insurance products are more suitable and more likely to be successful, so they cannot recommend products that customers really need. Sometimes insurance sales personnel are also unclear about the reasonable amount of insurance coverage, premium, protection period and payment period for the products recommended to customers, which leads to inaccurate estimates. The above situation is especially common for new insurance sales personnel.

[0004] Therefore, how to design an intelligent insurance proposal generation solution is an urgent problem to be solved. Summary of the invention

[0005] In view of this, the present application provides a method, device and readable storage medium for generating an insurance proposal, which is not limited to the personal experience of insurance business personnel, but obtains an insurance proposal that is more conducive to transaction, has a high degree of intelligence, and is conducive to improving the accuracy of insurance product recommendations.

[0006] In order to solve the above technical problems, the present application provides a method for generating an insurance proposal, comprising:

[0007] Obtain customer profile information of the policyholder;

[0008] Based on the customer profile information and the product classification prediction model, predict the target insurance product classification corresponding to the target object, the target object including the policyholder and / or the policyholder's family members;

[0009] Based on the customer portrait information and the primary insurance product prediction model, determine the target primary insurance product corresponding to the target object under the target insurance product classification;

[0010] Based on the customer profile information, the target primary insurance product and the preset association rule information, determine the target supplementary insurance product and target product configuration information associated with the target primary insurance product;

[0011] An insurance proposal including the target primary insurance product, the target supplementary insurance product and the target product configuration information is generated.

[0012] In order to solve the above technical problems, the present application also provides a device for generating an insurance proposal, comprising:

[0013] The acquisition module is used to obtain the customer profile information of the policyholder;

[0014] A product classification prediction module, for predicting the target insurance product classification corresponding to the target object based on the customer portrait information and the product classification prediction model, wherein the target object includes the policyholder and / or the policyholder's family members;

[0015] A primary insurance prediction module, used to determine the target primary insurance product corresponding to the target object under the target insurance product classification based on the customer portrait information and the primary insurance product prediction model;

[0016] A supplementary insurance and configuration information prediction module, used to determine the target supplementary insurance product and target product configuration information associated with the target main insurance product based on the customer portrait information, the target main insurance product and preset association rule information;

[0017] The proposal generation module is used to generate an insurance proposal including the target main insurance product, the target supplementary insurance product and the target product configuration information.

[0018] In order to solve the above technical problems, the present application also provides a readable storage medium on which a program or instruction is stored. When the program or instruction is executed by a processor, the steps of the method for generating an insurance proposal as described above are implemented.

[0019] The present application provides a method, device and readable storage medium for generating an insurance proposal, the scheme includes: obtaining the customer portrait information of the insured, predicting the target insurance product classification corresponding to the target object based on the customer portrait information and the product classification prediction model, the target object includes the insured and / or the family members of the insured, determining the target main insurance product corresponding to the target object under the target insurance product classification based on the customer portrait information and the main insurance product prediction model, determining the target additional product and target product configuration information associated with the target main insurance product based on the customer portrait information, the target main insurance product and the preset association rule information, and finally generating an insurance proposal including the target main insurance product, the target additional insurance product and the target product configuration information. It can be seen that through the above scheme, according to the insured's portrait information, the target insurance product classification that is more in line with the target object can be gradually determined, and the target main insurance product and the corresponding target additional product and target product configuration information under this classification are not limited to the personal experience of the insurance business personnel, and an insurance proposal that is more conducive to the transaction is obtained, which has a high degree of intelligence and is conducive to improving the accuracy of insurance product recommendations.

[0020] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0022] Figure 1 A flowchart of a method for generating an insurance proposal provided for this application;

[0023] Figure 2 A schematic diagram of a hierarchical architecture of a process for generating an insurance proposal provided for this application;

[0024] Figure 3 A schematic diagram of the structure of a device for generating an insurance proposal provided in this application;

[0025] Figure 4 is a structural schematic diagram of a computer device in one embodiment of the present application;

[0026] Figure 5 It is another structural schematic diagram of a computer device in one embodiment of the present application. DETAILED DESCRIPTION

[0027] The core of this application is to provide a method, device and readable storage medium for generating an insurance proposal, which is not limited to the personal experience of insurance business personnel, but can obtain an insurance proposal that is more conducive to transaction, has a high degree of intelligence, and is conducive to improving the accuracy of insurance product recommendations.

[0028] The following will be combined with the drawings in the embodiments of the present application to clearly describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments in the present application belong to the scope of protection of this application.

[0029] The terms "first", "second", etc. in the specification and claims of the present application are used to distinguish similar objects, and are not used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable under appropriate circumstances, so that the embodiments of the present application can be implemented in an order other than those illustrated or described here, and the objects distinguished by "first", "second", etc. are generally of one type, and the number of objects is not limited. For example, the first object can be one or more. In addition, "and / or" in the specification and claims represents at least one of the connected objects, and the character " / " generally indicates that the objects associated with each other are in an "or" relationship.

[0030] Please refer to Figure 1 , Figure 1 A flowchart of a method for generating an insurance proposal provided for this application.

[0031] The method for generating the insurance proposal includes:

[0032] S11: Obtain the customer profile information of the policyholder;

[0033] In this embodiment, the customer portrait information may include the insured's basic information, such as age, gender, occupation, annual income, life stage, wealth level and family information, where the family information may include the age, gender, occupation, annual income, life stage, wealth level and other information of the insured's family members; the customer portrait information may also include the insured's previous transaction behavior information, active behavior information and demand tendency information that represents his or her purchasing intention, etc., which are not specifically limited here.

[0034] Each customer has a unique identity ID, and the corresponding customer portrait information can be stored in the form of tags and uploaded to the database for regular updates. The customer portrait information tags can include structured tags and unstructured tags. The unstructured tags here refer to customer portrait information data stored in the form of text, voice, video or image. In addition, in practical applications, large model technology can be introduced to process unstructured portrait information data to obtain corresponding structured data tags. For example, by analyzing the customer's voice data or chat record pictures, it is directly possible to obtain whether the customer is interested in insurance to complete the corresponding tag. This processing method can save the loss of manual labeling input tags, which is conducive to improving the prediction accuracy of subsequent product classification prediction models. The large model technology here includes but is not limited to AIGC technology (Artificial Intelligence Generated Content, generative artificial intelligence), etc., use AIGC technology to analyze chat record information with customers or relevant description information of customers about their insurance needs, so as to extract customer portrait information and obtain corresponding multiple tags for storage; it is understandable that for different insurance products, corresponding product tags can also be generated, that is, the key tags corresponding to the insurance are mined according to product liability and product promotion content, such as dividend type, maturity survival benefit and policy loan, etc.; in addition, the preferences and purchasing habits of the people corresponding to the characteristics of different insurance types can be analyzed and stored to achieve effective management of customer information and insurance product information; from the actual application effect, after using the above method to mine more unstructured data value, the accuracy of insurance proposal generation can be improved by 28.33%, which is significant.

[0035] S12: Based on the customer profile information and the product classification prediction model, predict the target insurance product classification corresponding to the target object, where the target object includes the policyholder and / or the policyholder's family members;

[0036] Specifically, considering that the insured's family members may also have insurance needs, the target objects may include the insured and / or the insured's family members, so that not only more suitable insurance products can be recommended to the insured, but also suitable insurance products can be recommended to their family members, enriching the content of the insurance proposal and helping to increase the probability of a deal.

[0037] S13: Based on the customer profile information and the primary insurance product prediction model, determine the target primary insurance product corresponding to the target object under the target insurance product classification;

[0038] Specifically, the target insurance product classification refers to major product categories such as protection and financial management, and the target main insurance product refers to the sub-category of products under the major product categories.

[0039] S14: Determine the target supplementary insurance product and target product configuration information associated with the target main insurance product based on the customer profile information, the target main insurance product and the preset association rule information;

[0040] S15: Generate an insurance proposal including a target primary insurance product, a target supplementary insurance product and target product configuration information.

[0041] Specifically, the target product configuration information here may include premium, insured amount, delivery date and insurance period, thereby intelligently generating an insurance proposal that is more suitable for the target object.

[0042] In addition, when the target objects include the insured and the insured's family members, steps S13 and S14 are performed for the insured and his family members respectively, so that the insurance proposal obtained in step S15 includes insurance proposals for both the insured and his family members.

[0043] It is understandable that the above models can be evaluated and modified in actual use according to the actual use effect of the generated insurance proposals, so as to continuously optimize the model and generate insurance proposals with higher customer satisfaction.

[0044] Please refer to Figure 2 , Figure 2 A hierarchical architecture diagram of the process of generating an insurance proposal provided for this application. It should be noted that the KYC layer here refers to the layer corresponding to the KYC rules (Know Your Customer), at which the customer portrait information of the policyholder is collected; the model layer shows the information flow process when each model is applied to execute the above steps to generate an insurance proposal; the application layer shows the insurance recommendation content that the policyholder can actually see when it is finally facing the policyholder (wherein, AI digital human refers to the exclusive image of the insurance business personnel generated by the digital human technology in the following embodiment, so that he can serve as the main speaker to explain the plan explanation video corresponding to the insurance proposal); in addition, the operation layer can also be set to continuously optimize and correct the above model, in order to generate an insurance proposal that better meets the needs of the policyholder and is conducive to the transaction.

[0045] In summary, the present application provides a method for generating an insurance proposal, which can gradually determine a target insurance product classification that is more suitable for the target object based on the insured's portrait information. The target main insurance product under this classification and the corresponding target additional product and target product configuration information are not limited to the personal experience of the insurance business personnel, and an insurance proposal that is more conducive to the transaction is obtained. It has a high degree of intelligence and is conducive to improving the accuracy of insurance product recommendations.

[0046] Based on the above embodiments:

[0047] In some embodiments, the product classification prediction model includes at least one of a historical customer multi-classification prediction sub-model, a new customer prediction sub-model, a business experience scoring sub-model, and a customer attention review molecular model; based on the customer portrait information and the product classification prediction model, predicting the target insurance product classification corresponding to the target object includes:

[0048] Determine the type of the target object, and predict the target insurance product classification corresponding to the target object according to the sub-model corresponding to the type of the target object.

[0049] In this embodiment, considering that the target object may be a new customer or an old customer, may have concerns or may not have concerns, the type of the target object is determined based on the customer portrait information, and then prediction is performed according to the sub-model corresponding to the type, which helps to improve the accuracy of the prediction results.

[0050] It should also be noted that the historical customer multi-classification prediction sub-model is designed based on the customer's historical trace data; the business experience scoring sub-model is designed based on the consideration that high-performing insurance business personnel primarily consider the customer's life stage and wealth level when recommending insurance product solutions to customers; the customer attention review molecular model is designed based on the consideration of the customer's demand tendency information for purchasing insurance products, that is, the attention points can be mapped to different product categories; the new customer prediction sub-model is obtained by statistically analyzing the product categories that customers may be interested in based on experience or research.

[0051] In some embodiments, predicting the target insurance product classification corresponding to the target object according to the sub-model corresponding to the type of the target object includes:

[0052] If the target object is a historical customer and the target object has focus points, then based on the life stage information and wealth level information in the customer portrait information and the business experience scoring sub-model, the insurance product classification corresponding to the target object's grid customer group demand tendency and the score of each insurance product classification are predicted, and based on the customer portrait information and the historical customer multi-classification prediction sub-model, the insurance product classification corresponding to the target object's similar customer demand tendency and the score of each insurance product classification are predicted, and based on the focus point information in the customer portrait information and the customer focus comment molecular model, the insurance product classification corresponding to the target object's focus point and the score of each insurance product classification are predicted;

[0053] If the target object is a historical customer and the target object has no focus, then based on the life stage information and wealth level information in the customer portrait information and the business experience scoring sub-model, the insurance product classification corresponding to the target object's Gongge customer group demand tendency and the score of each insurance product classification are predicted, and based on the customer portrait information and the historical customer multi-classification prediction sub-model, the insurance product classification corresponding to the target object's similar customer demand tendency and the score of each insurance product classification are predicted;

[0054] If the target object is a new customer and the target object has focus points, then based on the life stage information and wealth level information in the customer portrait information and the business experience scoring sub-model, the insurance product classification corresponding to the target object's Gongge customer group demand tendency and the score of each insurance product classification are predicted, and based on the life stage information in the customer portrait information and the new customer prediction sub-model, the insurance product classification corresponding to the target object's similar customer demand tendency and the score of each insurance product classification are predicted, and based on the focus point information in the customer portrait information and the customer focus comment molecular model, the insurance product classification corresponding to the target object's focus point and the score of each insurance product classification are predicted;

[0055] If the target object is a new customer and the target object has no focus, then based on the life stage information and wealth level information in the customer portrait information and the business experience scoring sub-model, the insurance product classification corresponding to the target object's Gongge customer group demand tendency and the score of each insurance product classification are predicted, and based on the life stage information in the customer portrait information and the new customer prediction sub-model, the insurance product classification corresponding to the target object's similar customer demand tendency and the score of each insurance product classification are predicted;

[0056] For each insurance product classification, a total score of the insurance product classification is calculated; wherein the method of calculating the total score of the insurance product classification includes: multiplying each score of the insurance product classification by a weight coefficient of a corresponding sub-model to obtain a plurality of products, and calculating the sum of the plurality of products to obtain the total score of the insurance product classification;

[0057] The top n insurance product categories ranked from high to low in total scores are taken as the target insurance product categories corresponding to the target objects, where n is a positive integer.

[0058] In this embodiment, the target object having a focus here means that the target object has a corresponding insurance product demand tendency. Different focus points can be mapped to different product categories. For example, if the insured is interested in wealth insurance products, then the highest priority product category obtained by further relying on the customer attention review molecular model is financial insurance products, which reflects the influence of focus points on customer choices.

[0059] Specifically, the customer portrait information includes various label information of the target object, including customer basic information labels (such as age, gender, education and marital status, etc.), customer purchase intention labels (such as number of appointments, life stage, etc.), and customer purchasing ability labels (such as income, whether or not a VIP member customer, and annual premium); more specifically, the data middle platform of the group where the insurance business personnel is located can also store the information of each insurance business personnel in the form of labels, including but not limited to the basic information labels of the insurance business personnel (such as age, gender, education and marital status, etc.), performance information labels (such as premiums, the number of products sold by agents in a year), and customer contact information labels (such as the number of face-to-face interviews, where insurance proposals are made, and whether there is telephone communication with customers), etc., so that the data middle platform can uniformly manage customer information and insurance business personnel information; in this embodiment, the customer portrait information is input as an input item into the historical customer multi-classification prediction sub-model, and the insurance product classification corresponding to the same type of customer demand tendencies of the target object and the score of each insurance product classification can be obtained. It is understandable that the input data can be preprocessed before entering the sub-model, including but not limited to cleaning data, processing missing values ​​and outliers, converting data according to a standard format, etc.; in addition, the historical customer multi-classification prediction sub-model can be specifically established by relying on multinomial logistic regression and multi-classification prediction models.

[0060] The customer profile information includes life stage information and wealth level information, so that it can be used as an input item to predict the insurance product classification corresponding to the demand tendency of the target object of the grid customer group and the score of each insurance product classification; it can be understood that the business experience scoring sub-model can be a regular grid model designed based on the business experience of high-performing insurance business personnel, so that it can be extracted according to the customer profile information. More specifically, please refer to Table 1, which shows the insurance product classification preferred by different grid customer groups at different life stages and wealth levels and the appeal of each insurance product classification to customers. Here, the life stage is divided into single nobles (unmarried), two-person world (young married but childless), family with cute babies (with young children and the age of the children is less than 10 years old), mature families (with middle-aged children and the age of the children is greater than 10 years old) and retired elderly (age greater than 55 years old); here, the wealth level information is divided into the general public (annual income less than 100,000), quasi-rich (annual income of 100,000-500,000), wealth (annual income of 500,000-1 million) and high net worth (annual income greater than 1 million).

[0061] Table 1

[0062]

[0063]

[0064] The new customer prediction sub-model here can be obtained by ranking the interest levels of insurance products that customers may be interested in under five life stages. Based on the life stage information in the customer portrait information and the sub-model, the insurance product classification corresponding to the target object's similar customer demand tendencies and the scores of each insurance product classification can be obtained.

[0065] Furthermore, the weight coefficient corresponding to the sub-model used in calculating the total score of the insurance product classification can be dynamically adjusted by the insurance business personnel according to the business results in actual applications to improve the accuracy of the insurance proposal; for example, for new customers with focus points, the predicted total score of the insurance product classification can be calculated as: the score of the insurance product classification corresponding to the focus point × 80% + the score of the insurance product classification corresponding to the demand tendency of the palace customer group × 20%; for old users with focus points, the predicted total score of the insurance product classification can be calculated as: the score of the insurance product classification corresponding to the focus point × 80% + the score of the insurance product classification corresponding to the demand tendency of the palace customer group × 20% × 60% + the score of the insurance product classification corresponding to the demand tendency of similar customers × 20 × 40%, among which, two weight coefficients need to be multiplied because there are two focus points corresponding to the insurance product classification, namely life stage and wealth level, but it should be noted that the sum of the weight coefficients of each sub-model should be 100%.

[0066] It can be seen that through the above method, the experience of high-performing insurance sales personnel can be combined with big data models to accurately and reliably target different target objects and whether there are concerns among the target objects, and recommend the most likely insurance plans for them. This will help us understand customers better than the customers themselves, and ultimately improve the compatibility of the insurance proposal sheet with the customer.

[0067] In some embodiments, based on the customer profile information and the primary insurance product prediction model, determining the target primary insurance product corresponding to the target object under the target insurance product category includes:

[0068] Based on customer profile information and the main insurance product prediction model, multiple main insurance products under different target insurance product categories are predicted respectively;

[0069] Determine the target primary insurance product corresponding to the target object among multiple primary insurance products.

[0070] In this embodiment, considering that the target insurance product classification is a major product category, there are a considerable number of main insurances under it. Therefore, when determining the main insurance under the major product category, the customer portrait information is input as an input item into a pre-set main insurance product prediction model, so that multiple main insurance products under the target insurance product classification can be screened, and then the target main insurance product corresponding to the target object can be determined from the multiple main insurance products; in addition, the main insurance product prediction model here can be a pre-established Cube rule model prediction table, which defines one or more main insurances that customers tend to choose under different Cube tags. The Cube tag here refers to the dimensional label corresponding to the customer portrait information such as age, gender, life stage, wealth level, income, etc. Finally, relying on the model to execute the above steps, the target object can be comprehensively considered. The customer portrait information such as the life stage, gender, occupation, income, etc., select the main insurance product with the highest sales as the target main insurance product for the target object, without recommending a main insurance product that exceeds the target object's income level.

[0071] In some embodiments, based on the customer profile information, the target primary insurance product and the preset association rule information, determining the target supplementary insurance product and the target product configuration information associated with the target primary insurance product includes:

[0072] In the preset association rule information, find the top m supplementary insurance products ranked from high to low in terms of association with the customer profile information and the target main insurance product, and use them as the target supplementary insurance products associated with the target main insurance product, where m is a positive integer;

[0073] In the preset association rule information, the target product configuration information corresponding to the customer portrait information and the target main insurance product is determined, and the target product configuration information includes at least one of the following: premium, insured amount, delivery date, and insurance period.

[0074] In this embodiment, after the target main insurance product is confirmed, the above method can be used to further obtain the hot-selling target supplementary insurance product under the target main insurance product, recommend it to the insured for selection, and clarify the target product configuration information, so that the entire insurance proposal is clear.

[0075] In some embodiments, the method further comprises:

[0076] Based on the basic information of the insured, obtain the supplementary insurance products and product configuration information corresponding to each dimensional feature in the basic information from the stored data corresponding to the main insurance product;

[0077] Combine each dimensional feature to obtain multiple dimensional feature combinations, and calculate the correlation between different supplementary insurance products and main insurance products under each dimensional feature combination, and calculate the median of product configuration information under each dimensional feature combination, and use the median as the product configuration information corresponding to the dimensional combination;

[0078] The supplementary insurance products corresponding to each dimensional feature combination and their correlation, as well as product configuration information, are stored as preset association rule information.

[0079] In this embodiment, specifically, the dimensional features here may include gender dimension, age dimension, income dimension, occupation dimension and life stage dimension, etc., and through the above settings, the product configuration information corresponding to each main insurance under different dimensional feature combinations and the correlation with the supplementary insurance product can be obtained, and stored as preset association rule information, so that the target supplementary insurance product corresponding to the target main insurance and the target product configuration information can be determined according to the customer portrait information of the target object and the preset association rule information.

[0080] In some embodiments, the method further comprises:

[0081] Acquire first recommendation data input by the insurance business personnel, and use the first recommendation data to replace at least one of the target primary insurance product, the supplementary insurance product, and the target product configuration information; or,

[0082] Determine the second recommended data according to the business guarantee plan, and use the second recommended data to replace at least one of the target main insurance product, supplementary insurance product and target product configuration information.

[0083] In this embodiment, considering that there are some cases where the insurance business personnel may have more information about the target object and a more in-depth and comprehensive understanding, or they may learn during the communication with the target object that the target object has other ideas about the selection of insurance products, the first recommended data input by the insurance business personnel can be received to replace at least one of the target main insurance product, supplementary insurance product and target product configuration information to better meet the actual insurance needs of the target object, which is conducive to improving the accuracy of the insurance proposal; more specifically, the insurance business personnel can operate on the human-computer interaction interface to input the first recommended data.

[0084] In addition, a business guarantee plan can be pre-set to correct and modify the currently generated insurance proposals that are obviously unreasonable, so as to better meet the insurance needs of the target objects. Specifically, the business guarantee plan here includes but is not limited to setting it according to the current company's main insurance promotion direction, so as to recommend the company's main insurance business to customers, and increase the sales volume of these main insurance businesses on the basis of meeting customer needs.

[0085] In some embodiments, the method further comprises:

[0086] Generate a plan explanation text from the insurance proposal based on the insurance business personnel's experience and / or insurance training information;

[0087] Based on the plan explanation text, digital human technology is used to generate a plan explanation video with the exclusive image of insurance business personnel.

[0088] In this embodiment, it is further considered that in actual applications, there are insurance business personnel who do not know how to explain the insurance proposal in order to facilitate transactions, and there are situations where the explanation of the insurance proposal is not comprehensive and clear, and the words used may not be professional and not attractive enough to customers; in addition, in actual applications, there are situations where insurance business personnel operate insurance products by recording videos or video numbers, but they may not always be able to maintain a full state, and the video shooting time is long and complicated, and there may be situations where the explanation is not good or the shooting is not good and needs to be restarted. Therefore, in this application, the experience words information and / or insurance training information of high-performing insurance business personnel are obtained in advance, so that the insurance proposal is generated based on this information. The program explanation text.

[0089] More specifically, GPT technology (Generative Pre-Trained Transformer) can be used to generate one-stop explanation scripts for products and services for different policyholders, including demand analysis, family protection details and configuration suggestions, insurance product introduction, service reception, service video, and insurance proposal plan interpretation. Through experience script information and / or insurance training information, AI (Artificial Intelligence) based on GPT technology can be used to explain the product and service in a one-stop manner. The training may involve multiple aspects, such as topic entry (including current affairs, children's education, event invitations and customer cases), concept introduction (such as health, medical and elderly care resources, family responsibilities and protection functions), sales skills and the touching points corresponding to the segmented customer matrix to form explanation points and labels, and build the following knowledge base, including customer library (segmentation portraits, behavioral preferences and risk reviews), policy library (including insurance tendencies and insurance type preferences), product library (including product liabilities and service rights). In actual application, the current insurance proposal is processed according to the key point matching to generate the corresponding plan explanation text, which may include an overall introduction to the plan that reflects the characteristics of the insurance proposal, applicable demand scenarios that reflect the needs of the target object, product protection content that reflects the responsibility, value-added service content that reflects the service, and exemplary customer case analysis.

[0090] Furthermore, the voice and image of the insurance business personnel can be cloned using digital human technology to generate a video of the plan explanation in the exclusive image of the insurance business personnel, which explains the text of the plan explanation. The video can then be directly shared to the insured's electronic device through a social account. It is understandable that the plan explanation video can also include a cover, a back cover, an introduction to the insurance business personnel, an introduction to the company's strength, demand stimulation, and a display of the insurance proposal including insurance product liability, features, benefits, and liability statements; in addition, the plan explanation video can also be combined with a PPT presentation to obtain a more diversified and attractive display content, which is not specifically limited here.

[0091] It can be seen that, on the one hand, the above method allows insurance sales personnel to no longer have to worry about the explanation language. Combined with the experience of high-performing sales personnel, more appropriate explanation language can be recommended to them, which is conducive to improving the success rate of signing contracts. On the other hand, the determination of the plan explanation video is conducive to subsequent insurance sales personnel forwarding it to customers and / or posting the insurance proposal explanation video on social media such as video accounts, demonstrating professionalism and promoting insurance sales. In addition, digital people explain insurance products on behalf of insurance sales personnel, which saves time and effort for insurance sales personnel.

[0092] Please refer to Figure 3 , Figure 3 A schematic diagram of the structure of a device for generating an insurance proposal provided by the present invention.

[0093] The device for generating the insurance proposal includes:

[0094] An acquisition module 21 is used to acquire the customer profile information of the policyholder;

[0095] A product classification prediction module 22, for predicting the target insurance product classification corresponding to the target object based on the customer profile information and the product classification prediction model, the target object including the policyholder and / or the policyholder's family members;

[0096] The primary insurance prediction module 23 is used to determine the target primary insurance product corresponding to the target object under the target insurance product classification based on the customer profile information and the primary insurance product prediction model;

[0097] The supplementary insurance and configuration information prediction module 24 is used to determine the target supplementary insurance product and target product configuration information associated with the target main insurance product based on the customer profile information, the target main insurance product and the preset association rule information;

[0098] The proposal generation module 25 is used to generate an insurance proposal including a target primary insurance product, a target supplementary insurance product and target product configuration information.

[0099] For an introduction to the device for generating an insurance proposal provided in this application, please refer to the embodiment of the method for generating an insurance proposal described above, which will not be described in detail here.

[0100] In some embodiments, the product classification prediction model includes at least one of a historical customer multi-classification prediction sub-model, a new customer prediction sub-model, a business experience scoring sub-model, and a customer attention review molecular model; the product classification prediction module 22 includes:

[0101] The category determination module is used to determine the type of the target object and predict the target insurance product classification corresponding to the target object according to the sub-model corresponding to the type of the target object.

[0102] In some embodiments, the category determination module includes a first processing module, a second processing module, a third processing module, a fourth processing module, a score statistics module and a first sorting module; wherein, if the target object is a historical customer and the target object has a focus, the first processing module is triggered; if the target object is a historical customer and the target object has no focus, the second processing module is triggered; if the target object is a new customer and the target object has a focus, the third processing module is triggered; if the target object is a new customer and the target object has no focus, the fourth processing module is triggered;

[0103] The first processing module is used to predict the insurance product classification corresponding to the target object's Gongge customer group demand tendency and the score of each insurance product classification based on the life stage information and wealth level information in the customer portrait information and the business experience scoring sub-model, and to predict the insurance product classification corresponding to the target object's similar customer demand tendency and the score of each insurance product classification based on the customer portrait information and the historical customer multi-classification prediction sub-model, and to predict the insurance product classification corresponding to the target object's focus and the score of each insurance product classification based on the focus information in the customer portrait information and the customer focus comment molecular model;

[0104] The second processing module is used to predict the insurance product classification corresponding to the target object's Gongge customer group demand tendency and the score of each insurance product classification based on the life stage information and wealth level information in the customer portrait information and the business experience scoring sub-model, and to predict the insurance product classification corresponding to the target object's similar customer demand tendency and the score of each insurance product classification based on the customer portrait information and the historical customer multi-classification prediction sub-model;

[0105] The third processing module is used to predict the insurance product classification corresponding to the target object's Gongge customer group demand tendency and the score of each insurance product classification based on the life stage information and wealth level information in the customer portrait information and the business experience scoring sub-model, and predict the insurance product classification corresponding to the target object's similar customer demand tendency and the score of each insurance product classification based on the life stage information and the new customer prediction sub-model in the customer portrait information, and predict the insurance product classification corresponding to the target object's focus and the score of each insurance product classification based on the focus information in the customer portrait information and the customer focus comment molecular model;

[0106] The fourth processing module is used to predict the insurance product classification corresponding to the target object's Gongge customer group demand tendency and the score of each insurance product classification based on the life stage information and wealth level information in the customer portrait information and the business experience scoring sub-model, and predict the insurance product classification corresponding to the target object's similar customer demand tendency and the score of each insurance product classification based on the life stage information in the customer portrait information and the new customer prediction sub-model;

[0107] A score statistics module is used to calculate the total score of the insurance product classification for each insurance product classification; wherein the method of calculating the total score of the insurance product classification includes: multiplying each score of the insurance product classification by the weight coefficient of the corresponding sub-model to obtain multiple products, and calculating the sum of the multiple products to obtain the total score of the insurance product classification;

[0108] The first sorting and screening module is used to classify the top n insurance products with the highest total scores from high to low as the target insurance product classification corresponding to the target object, where n is a positive integer.

[0109] In some embodiments, the primary risk prediction module 23 includes:

[0110] The main insurance prediction submodule is used to predict multiple main insurance products under different target insurance product categories based on customer profile information and the main insurance product prediction model;

[0111] The screening module is used to determine the target primary insurance product corresponding to the target object among multiple primary insurance products.

[0112] In some embodiments, the additional risk and configuration information prediction module 24 includes:

[0113] The second sorting and screening module is used to find the top m supplementary insurance products ranked from high to low in terms of correlation between the customer profile information and the target main insurance product in the preset association rule information, and use them as the target supplementary insurance products associated with the target main insurance product, where m is a positive integer;

[0114] The product configuration information determination module is used to determine the target product configuration information corresponding to the customer portrait information and the target main insurance product in the preset association rule information. The target product configuration information includes at least one of the following: premium, insured amount, delivery date, and insurance period.

[0115] In some embodiments, the insurance proposal generating device further includes:

[0116] The basic information acquisition module is used to obtain the supplementary insurance products and product configuration information corresponding to each dimensional feature in the basic information from the stored data corresponding to the main insurance product based on the basic information of the insured;

[0117] A combination evaluation module is used to combine each dimension feature to obtain multiple dimension feature combinations, and calculate the correlation between different supplementary insurance products and main insurance products under each dimension feature combination, and calculate the median of product configuration information under each dimension feature combination, and use the median as the product configuration information corresponding to the dimension combination;

[0118] The storage module is used to store the supplementary insurance products corresponding to each dimensional feature combination and their correlation, as well as product configuration information, as preset association rule information.

[0119] In some embodiments, the insurance proposal generating device further includes:

[0120] A first replacement module is used to obtain first recommendation data input by an insurance business person, and replace at least one of a target primary insurance product, a supplementary insurance product, and target product configuration information with the first recommendation data; or,

[0121] The second replacement module is used to determine the second recommended data according to the business guarantee plan, and use the second recommended data to replace at least one of the target main insurance product, supplementary insurance product and target product configuration information.

[0122] In some embodiments, the insurance proposal generating device further includes:

[0123] A text generation module is used to generate a text of the insurance proposal based on the experience and speech information of the insurance business personnel and / or the insurance training information;

[0124] The video generation module is used to generate a plan explanation video with an exclusive image of insurance business personnel based on the plan explanation text using digital human technology.

[0125] The present application also provides a readable storage medium on which a program or instruction is stored. When the program or instruction is executed by a processor, the steps of the method for generating an insurance proposal as described above are implemented.

[0126] For an introduction to the readable storage medium provided in this application, please refer to the above-mentioned embodiment of the method for generating an insurance proposal, which will not be repeated here.

[0127] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 4 As shown. The computer device includes a processor, a memory, a network interface and a database connected through a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile and / or volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external client through a network connection. When the computer program is executed by the processor, the functions or steps of a method for generating an insurance proposal are implemented.

[0128] In one embodiment, a computer device is provided. The computer device may be a client, and its internal structure diagram may be as follows: Figure 5 As shown. The computer device includes a processor, a memory, a network interface, a display screen and an input device connected through a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external server through a network connection. When the computer program is executed by the processor, the functions or steps of a method for generating an insurance proposal are implemented.

[0129] It should be noted that, in this article, the terms "comprise", "include" or any other variant thereof are intended to cover non-exclusive inclusion, so that the process, method, article or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise one..." do not exclude the presence of other identical elements in the process, method, article or device including the element. In addition, it should be pointed out that the scope of the methods and devices in the embodiments of the present application is not limited to performing functions in the order shown or discussed, and may also include performing functions in a substantially simultaneous manner or in reverse order according to the functions involved, for example, the described method may be performed in an order different from that described, and various steps may also be added, omitted, or combined. In addition, the features described with reference to certain examples may be combined in other examples.

[0130] The embodiments of the present application are described above in conjunction with the accompanying drawings, but the present application is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present application, ordinary technicians in this field can also make many forms without departing from the purpose of the present application and the scope of protection of the claims, all of which are within the protection of the present application.

Claims

1. A method for generating an insurance proposal, characterized in that: include: Obtain customer profile information of the policyholder; Based on the customer profile information and the product classification prediction model, predict the target insurance product classification corresponding to the target object, the target object including the policyholder and / or the policyholder's family members; Based on the customer portrait information and the primary insurance product prediction model, determine the target primary insurance product corresponding to the target object under the target insurance product classification; Based on the customer profile information, the target primary insurance product and the preset association rule information, determine the target supplementary insurance product and target product configuration information associated with the target primary insurance product; An insurance proposal including the target primary insurance product, the target supplementary insurance product and the target product configuration information is generated.

2. The method according to claim 1, characterized in that The product classification prediction model includes at least one of a historical customer multi-classification prediction sub-model, a new customer prediction sub-model, a business experience scoring sub-model, and a customer attention review molecular model; the target insurance product classification corresponding to the target object is predicted based on the customer portrait information and the product classification prediction model, including: The type of the target object is determined, and the target insurance product classification corresponding to the target object is predicted according to the sub-model corresponding to the type of the target object.

3. The method according to claim 2, characterized in that The predicting of the target insurance product classification corresponding to the target object according to the sub-model corresponding to the type of the target object includes: If the target object is a historical customer and the target object has focus points, then based on the life stage information and wealth level information in the customer portrait information and the business experience scoring sub-model, the insurance product classification corresponding to the target object's grid customer group demand tendency and the score of each insurance product classification are predicted, and based on the customer portrait information and the historical customer multi-classification prediction sub-model, the insurance product classification corresponding to the target object's similar customer demand tendency and the score of each insurance product classification are predicted, and based on the focus point information in the customer portrait information and the customer focus review molecular model, the insurance product classification corresponding to the target object's focus point and the score of each insurance product classification are predicted; If the target object is a historical customer and the target object has no focus point, then based on the life stage information and wealth level information in the customer portrait information and the business experience scoring sub-model, the insurance product classification corresponding to the target object's grid customer group demand tendency and the score of each insurance product classification are predicted, and based on the customer portrait information and the historical customer multi-classification prediction sub-model, the insurance product classification corresponding to the target object's similar customer demand tendency and the score of each insurance product classification are predicted; If the target object is a new customer and the target object has focus points, then based on the life stage information and wealth level information in the customer portrait information and the business experience scoring sub-model, the insurance product classification corresponding to the target object's grid customer group demand tendency and the score of each insurance product classification are predicted, and based on the life stage information in the customer portrait information and the new customer prediction sub-model, the insurance product classification corresponding to the target object's similar customer demand tendency and the score of each insurance product classification are predicted, and based on the focus point information in the customer portrait information and the customer focus review molecular model, the insurance product classification corresponding to the target object's focus point and the score of each insurance product classification are predicted; If the target object is a new customer and the target object has no focus, then based on the life stage information and wealth level information in the customer portrait information and the business experience scoring sub-model, predict the insurance product classification corresponding to the target object's grid customer group demand tendency and the score of each insurance product classification, and based on the life stage information in the customer portrait information and the new customer prediction sub-model, predict the insurance product classification corresponding to the target object's similar customer demand tendency and the score of each insurance product classification; For each insurance product classification, a total score of the insurance product classification is calculated; wherein the method of calculating the total score of the insurance product classification includes: multiplying each score of the insurance product classification by a weight coefficient of a corresponding sub-model to obtain a plurality of products, and calculating the sum of the plurality of products to obtain the total score of the insurance product classification; The top n insurance product categories ranked from high to low in the total scores are taken as the target insurance product categories corresponding to the target object, where n is a positive integer.

4. The method according to claim 1, characterized in that The determining, based on the customer profile information and the primary insurance product prediction model, the target primary insurance product corresponding to the target object under the target insurance product classification includes: Based on the customer profile information and the main insurance product prediction model, respectively predict multiple main insurance products under different target insurance product categories; The target primary insurance product corresponding to the target object is determined from the plurality of primary insurance products.

5. The method according to claim 1, characterized in that The determining, based on the customer profile information, the target primary insurance product and the preset association rule information, the target supplementary insurance product and the target product configuration information associated with the target primary insurance product comprises: In the preset association rule information, search for the top m supplementary insurance products ranked from high to low in terms of association with the customer profile information and the target main insurance product, and use them as the target supplementary insurance products associated with the target main insurance product, where m is a positive integer; In the preset association rule information, the target product configuration information corresponding to the customer portrait information and the target main insurance product is determined, and the target product configuration information includes at least one of the following: premium, insured amount, delivery date, and insurance period.

6. The method according to claim 5, characterized in that The method further comprises: Based on the basic information of the insured, the supplementary insurance products and product configuration information corresponding to each dimensional feature in the basic information are obtained from the stored data corresponding to the main insurance product; Combine each of the dimensional features to obtain multiple dimensional feature combinations, and calculate the correlation between different supplementary insurance products and the main insurance product under each dimensional feature combination, and calculate the median of the product configuration information under each dimensional feature combination, and use the median as the product configuration information corresponding to the dimensional combination; The supplementary insurance products corresponding to each of the dimensional feature combinations and their correlation, as well as product configuration information, are stored as the preset association rule information.

7. The method according to claim 1, characterized in that The method further comprises: Acquire first recommendation data input by an insurance business person, and use the first recommendation data to replace at least one of the target primary insurance product, the supplementary insurance product, and the target product configuration information; or, Determine second recommended data according to the business guarantee plan, and use the second recommended data to replace at least one of the target main insurance product, the supplementary insurance product and the target product configuration information.

8. The method according to claim 1, characterized in that The method further comprises: Based on the experience and speech information of the insurance business personnel and / or the insurance training information, the insurance proposal is generated into a text explaining the plan; Based on the solution explanation text, digital human technology is used to generate a solution explanation video with an exclusive image of insurance business personnel.

9. A device for generating an insurance proposal, characterized in that: include: The acquisition module is used to obtain the customer profile information of the policyholder; A product classification prediction module, for predicting the target insurance product classification corresponding to the target object based on the customer portrait information and the product classification prediction model, wherein the target object includes the policyholder and / or the policyholder's family members; A primary insurance prediction module, used to determine the target primary insurance product corresponding to the target object under the target insurance product classification based on the customer portrait information and the primary insurance product prediction model; A supplementary insurance and configuration information prediction module, used to determine the target supplementary insurance product and target product configuration information associated with the target main insurance product based on the customer portrait information, the target main insurance product and preset association rule information; The proposal generation module is used to generate an insurance proposal including the target main insurance product, the target supplementary insurance product and the target product configuration information.

10. A readable storage medium having a program or instruction stored thereon, characterized in that: When the program or instruction is executed by a processor, the steps of the method for generating an insurance proposal according to any one of claims 1 to 8 are implemented.

Images