Insurance method, insurance device, electronic device, and storage medium

Abstract

This application provides an insurance application method, device, electronic device, and storage medium, belonging to the field of artificial intelligence technology and applied in fintech and healthcare scenarios. The method includes: extracting information from the object's identity information, health information, and insurance product information to obtain an identity field for the object's identity information, a health field for the object's health information, and a product field for the insurance product information; filling the identity field, health field, and product field into a preset insurance application template to obtain insurance application data; obtaining the insured object's historical insurance application counts; performing a risk assessment on the insured object based on the insurance application data and historical insurance application counts to obtain risk assessment data; constructing the insured object's insurance application behavior characteristics based on the historical insurance application counts; generating an insurance application process based on the insurance application behavior characteristics and risk assessment data; and conducting the insurance application according to the insurance application process. This application improves the flexibility of insurance application.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence technology and is applied to the fields of fintech and healthcare, particularly to an insurance application method, insurance application device, electronic device, and storage medium. Background Technology

[0002] In related technologies, users must follow a standardized application process when applying for insurance. However, the application process uses a uniform template, requiring all users to perform the same steps, resulting in low flexibility. Therefore, improving application flexibility has become an urgent problem to be solved. Summary of the Invention

[0003] The main objective of this application is to provide an insurance application method, device, electronic device, and storage medium, which aims to improve the flexibility of insurance application.

[0004] To achieve the above objectives, a first aspect of this application proposes an insurance application method, the method comprising: Obtain the basic information of the insured and the information of the insured product; wherein, the basic information of the insured includes the insured's identity information and the insured's health information; Information is extracted from the object's identity information, the object's health information, and the insurance product information to obtain the identity field of the object's identity information, the health field of the object's health information, and the product field of the insurance product information. The identity field, the health field, and the product field are then filled into a preset insurance template to obtain insurance data. Obtain the historical number of insurance policies for the insured individual; Based on the insurance data and the historical number of insurance applications, a risk assessment is performed on the insured object to obtain risk assessment data; Based on the historical number of insurance applications, the insurance application behavior characteristics of the insured object are constructed; An insurance application process is generated based on the characteristics of the insurance application behavior and the risk assessment data, and the insurance application is carried out in accordance with the insurance application process.

[0005] In some embodiments, the step of performing a risk assessment on the insured based on the insurance data and the historical number of insurance applications to obtain risk assessment data includes: The risk assessment model is determined based on the number of historical insurance policies. The risk assessment data is obtained by conducting a risk assessment on the insured object based on the insurance data and the risk assessment model.

[0006] In some embodiments, the risk assessment mode includes a first-time insurance assessment mode or a non-first-time insurance assessment mode. The step of conducting a risk assessment on the insured based on the insurance data and the risk assessment mode to obtain the risk assessment data includes: The insurance data is assessed using a pre-set risk assessment model to obtain initial assessment data. If the risk assessment mode is the first-time insurance assessment mode, then the initial assessment data will be used as the risk assessment data; If the risk assessment mode is the non-first-time insurance assessment mode, then the historical assessment data of the insured object is obtained, and the risk assessment data is determined based on the initial assessment data and the historical assessment data.

[0007] In some embodiments, the preset risk assessment model is trained according to the following steps: Send the preset initial evaluation model to at least two terminals; Obtain a reference evaluation model obtained by the terminal through local training of the preset initial evaluation model; wherein, the reference evaluation model is trained by the terminal based on local sample data, and the local sample data includes sample insurance data and risk labels of the sample insurance data; The model parameters of the reference assessment model are aggregated to obtain the preset risk assessment model.

[0008] In some embodiments, constructing the insurance behavior characteristics of the insured based on the number of historical insurance applications includes: If the number of historical insurance purchases is 0, then the insured objects are classified according to the insurance purchase data to obtain the object group category of the insured objects; Obtain the first historical behavioral characteristics of the object group category and the current behavioral characteristics of the insured object; The insurance application behavior characteristics are constructed based on the first historical behavior characteristics and the current behavior characteristics.

[0009] In some embodiments, constructing the insurance behavior characteristics of the insured based on the number of historical insurance applications includes: If the number of historical insurance purchases is greater than 0, then the second historical behavioral characteristics and current behavioral characteristics of the insured object are obtained; Calculate the feature deviation between the current behavior feature and the second historical behavior feature; The insurance behavior features are constructed based on the second historical behavior features and the feature deviation.

[0010] In some embodiments, generating the insurance application process based on the insurance application behavior characteristics and the risk assessment data includes: The object status of the insured is determined based on the characteristics of the insurance purchase behavior and the risk assessment data; Calculate the action score of each candidate process generated action in the preset process generation action space based on the object state; The candidate process with the highest action score is selected to generate the insurance application process.

[0011] To achieve the above objectives, a second aspect of this application provides an insurance application device, the device comprising: The first acquisition module is used to acquire the basic information of the insured object and the information of the insured product; wherein, the basic information of the object includes the object's identity information and the object's health information; The information extraction module is used to extract information from the object's identity information, the object's health information, and the insurance product information to obtain the identity field of the object's identity information, the health field of the object's health information, and the product field of the insurance product information. The identity field, the health field, and the product field are then filled into a preset insurance template to obtain insurance data. The second acquisition module is used to acquire the historical number of insurance policies of the insured object; The risk assessment module is used to assess the risk of the insured based on the insurance data and the number of historical insurance applications, and to obtain risk assessment data. The feature construction module is used to construct the insurance behavior features of the insured object based on the number of historical insurance purchases; The insurance application module is used to generate an insurance application process based on the insurance application behavior characteristics and the risk assessment data, and to apply for insurance according to the insurance application process.

[0012] To achieve the above objectives, a third aspect of this application provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the method described in the first aspect.

[0013] To achieve the above objectives, a fourth aspect of the present application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method described in the first aspect.

[0014] The insurance application method, device, electronic device, and computer-readable storage medium proposed in this application obtain basic information about the insured and the insured product information. Using information extraction technology, structured identity, health, and product fields are extracted from this information and automatically populated into an insurance application template to obtain standardized insurance data. Simultaneously, the historical number of insurance applications by the insured is obtained to construct insurance behavior characteristics, and risk assessment is performed in conjunction with the insurance data to obtain risk assessment data. Based on the insurance behavior characteristics and risk assessment data, a personalized insurance application process is generated for the insured, and insurance is applied for according to the process, improving the flexibility and efficiency of the insurance application process. Attached Figure Description

[0015] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is a schematic diagram of the application environment for the insurance application method provided in the embodiments of this application; Figure 2 This is a flowchart of the insurance application method provided in the embodiments of this application; Figure 3 yes Figure 2 The flowchart of step S240 in the text; Figure 4 yes Figure 3 The flowchart of step S320 in the middle; Figure 5 This is a flowchart of the training process of the preset risk assessment model provided in the embodiments of this application; Figure 6 yes Figure 2 The flowchart of step S250 in the text; Figure 7 yes Figure 2 Another flowchart of step S250 in the process; Figure 8 yes Figure 2 The flowchart of step S260 in the text; Figure 9 This is a schematic diagram of the insurance application device provided in the embodiments of this application; Figure 10 This is a schematic diagram of the hardware structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0018] It should be noted that although functional modules are divided in the device schematic diagram and a logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than the module division in the device or the order in the flowchart. The terms "first," "second," etc., in the specification, claims, and the aforementioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.

[0019] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0020] In related technologies, users must follow a standardized application process when applying for insurance. However, the application process uses a uniform template, requiring all users to perform the same steps, resulting in low flexibility. Therefore, improving application flexibility has become an urgent problem to be solved.

[0021] Based on this, embodiments of this application provide an insurance application method, an insurance application device, an electronic device, and a computer-readable storage medium, aiming to improve insurance application flexibility.

[0022] The insurance application method, insurance application device, electronic device, and computer-readable storage medium provided in this application are specifically described through the following embodiments. First, the insurance application method in this application embodiment is described.

[0023] The insurance application method provided in this application relates to the field of artificial intelligence technology. The insurance application method provided in this application can be applied to a terminal, a server, or software running on either a terminal or a server. In some embodiments, the terminal can be a smartphone, tablet, laptop, desktop computer, etc.; the server can be configured as an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms; the software can be an application implementing the insurance application method, but is not limited to the above forms.

[0024] This application can be used in a wide variety of general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices. This application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform specific tasks or implement specific abstract data types. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0025] It should be noted that in all specific embodiments of this application, when processing data related to user identity or characteristics, such as user information, user behavior data, user historical data, and user location information, user permission or consent is obtained first. Furthermore, the collection, use, and processing of this data comply with relevant laws, regulations, and standards. In addition, when embodiments of this application require access to sensitive personal information of users, separate permission or consent from the user is obtained through pop-ups or redirection to confirmation pages. Only after obtaining the user's separate permission or consent is the necessary user-related data required for the proper functioning of these embodiments acquired.

[0026] The insurance application method provided in this application can be applied to, for example, Figure 1In this application environment, the client communicates with the server via a network. The server can obtain the basic information of the insured and the information of the insured product through the client. The basic information includes the insured's identity information and health information. Information is extracted from the identity information, health information, and insured product information to obtain the identity field, the health field, and the product field. These fields are then filled into a preset insurance template to obtain insurance data. The server also obtains the insured's historical insurance application count; performs a risk assessment on the insured based on the insurance data and historical application count to obtain risk assessment data; constructs the insured's insurance behavior characteristics based on the historical application count; and generates an insurance application process based on the insurance behavior characteristics and risk assessment data, and executes the insurance application according to the process. In this application, a personalized insurance application process is generated for the insured based on the insurance behavior characteristics and risk assessment data, and the insurance application operation is executed according to the personalized process, thereby improving the flexibility of insurance application. The client can be, but is not limited to, various personal computers, laptops, smartphones, tablets, and portable wearable devices. The server can be implemented using a standalone server or a server cluster consisting of multiple servers. The following detailed description uses specific embodiments to illustrate this application.

[0027] Figure 2 This is an optional flowchart of the insurance application method provided in this application embodiment. Figure 2 The method may include, but is not limited to, steps S210 to S260.

[0028] Step S210: Obtain the basic information of the insured and the information of the insured product; wherein, the basic information of the insured includes the insured's identity information and the insured's health information; Step S220: Extract information from the object's identity information, object's health information, and insurance product information to obtain the identity field of the object's identity information, the health field of the object's health information, and the product field of the insurance product information. Then, fill the identity field, health field, and product field into the preset insurance template to obtain the insurance data. Step S230: Obtain the number of times the insured has purchased insurance in the past; Step S240: Conduct a risk assessment on the insured based on the insurance data and the number of historical insurance applications to obtain risk assessment data; Step S250: Construct the insurance behavior characteristics of the insured based on the number of historical insurance applications; Step S260: Generate an insurance application process based on the characteristics of the insurance application behavior and risk assessment data, and apply for insurance according to the insurance application process.

[0029] Steps S210 to S260 as shown in the embodiments of this application generate a personalized insurance application process for the insured based on the characteristics of the insurance application behavior and risk assessment data, and perform the insurance application operation according to the personalized insurance application process, rather than following a fixed process step, which greatly improves the flexibility of insurance application.

[0030] In step S210 of some embodiments, the policyholder initiates an insurance application request through the front-end page of the business system. The back-end service responds to the application request, extracting the object identifier and product identifier of the insured object from the application request. The insured object is the insured person in the insurance application business, and the application request is used to request the execution of the insurance application operation on the insured object. The object identifier is used to uniquely identify the insured object, and the product identifier is used to determine the specific insurance product for this application. To achieve data interoperability, avoid duplicate information entry, and improve information authenticity and reusability, the business system connects to other platforms through standardized interfaces, and obtains the basic information of the insured object across platforms based on the object identifier, reducing manual entry by users, thereby improving application efficiency and reducing entry errors. The basic information of the object is the basic information that distinguishes different insured objects, including object identity information and object health information. Among them, the object identity information is used to identify the identity of the insured object, such as name, age, occupation, and ID number, which can be obtained from the identity management platform; the object health information is the health data of the insured object, such as physical examination reports and past medical history, which can be obtained from the medical platform. The transmission of basic object information uses Advanced Encryption Standard (AES) to ensure data privacy and security. Based on the product identifier, the insured product information is retrieved from the business system. This information includes details of the specific insurance product being insured, such as insurance coverage, insurance period, insured amount, and premium calculation rules.

[0031] In step S220 of some embodiments, Named Entity Recognition (NER) is performed on the object identity information, object health information, and insurance product information to extract structured key fields, resulting in an identity field corresponding to the object identity information, a health field corresponding to the object health information, and a product field corresponding to the insurance product information. The identity field is used to identify the identity of the insured object, such as name, ID number, and contact information. The health field is used to identify the health status of the insured object, such as health indicators (blood pressure, blood sugar, heart rate, etc.) and medical history information. The product field is used to identify the insurance product, such as the insurance type and coverage type.

[0032] Different types of insurance have different risk points, and therefore focus on different identity, health, and product fields. For example, medical insurance focuses more on the insured's health status, while accident insurance focuses more on the insured's job category. Therefore, different application templates are needed for different types of insurance. An application template is a data structure or form framework used in insurance business to standardize the collection and storage of application information. The insurance type is determined based on the product information, and the application template is loaded according to the insurance type. The identity, health, and product fields are mapped to their corresponding template field names to obtain the application data.

[0033] In step S230 of some embodiments, the historical number of insurance policies of the insured object is obtained. The historical number of insurance policies refers to the number of times the insured object has been insured as the insured in the past period. The historical number of insurance policies is an integer greater than or equal to 0.

[0034] Please see Figure 3 In some embodiments, step S240 may include, but is not limited to, steps S310 to S320: Step S310: Determine the risk assessment model based on the number of historical insurance policies; Step S320: Conduct a risk assessment on the insured based on the insurance data and risk assessment model to obtain risk assessment data.

[0035] In step S310 of some embodiments, the risk characteristics of different insured individuals vary significantly, and a uniform risk assessment model cannot accurately identify individual risks. The number of historical insurance applications can reflect the insured individual's insurance behavior and potential risks. For example, a high number of applications in a short period carries a higher risk. To improve the accuracy of risk assessment, this application embodiment selects different risk assessment models based on the number of historical insurance applications. Specifically, if the number of historical insurance applications is 0, it indicates that the insured individual is acting as an insured for the first time within the statistical scope, and the risk assessment model is determined to be the first-time insurance application assessment model. If the number of historical insurance applications is greater than 0, it indicates that the insured individual has acted as an insured multiple times within the statistical scope, and the risk assessment model is determined to be the non-first-time insurance application assessment model.

[0036] In step S320 of some embodiments, a risk assessment is performed on the insured based on the insurance data, according to either the first-time insurance assessment mode or the non-first-time insurance assessment mode, to obtain the insured's risk assessment data. The risk assessment data is used to quantify the insured's risk level and can include risk categories, risk scores, etc. Risk categories include low risk, medium risk, and high risk. The higher the risk score, the higher the insured's risk level.

[0037] Steps S310 to S320 above differentiate risk assessment models based on the number of historical insurance applications and conduct differentiated risk assessments for different insured individuals according to the risk assessment models, which can improve the accuracy of risk assessment.

[0038] Please see Figure 4 In some embodiments, step S320 may include, but is not limited to, steps S410 to S430: Step S410: Perform a risk assessment on the insurance data using a preset risk assessment model to obtain initial assessment data; Step S420: If the risk assessment mode is the first-time insurance assessment mode, then the initial assessment data will be used as the risk assessment data. Step S430: If the risk assessment mode is the non-first-time insurance assessment mode, then obtain the historical assessment data of the insured, and determine the risk assessment data based on the initial assessment data and the historical assessment data.

[0039] In step S410 of some embodiments, the insurance data is input into a preset risk assessment model for risk assessment to obtain initial assessment data. The preset risk assessment model is a regression model or a classification model. The regression model can be an XGBoost regression model, a neural network, etc., and the classification model can be a random forest, an XGBoost classification model, etc. If the preset risk assessment model is a classification model, the initial assessment data is a risk category, which includes low risk, medium risk, or high risk. If the preset risk assessment model is a regression model, the initial assessment data is a risk score, such as 80 points.

[0040] In step S420 of some embodiments, if the risk assessment mode is the first-time insurance assessment mode, it means that the risk assessment of the insured object is being carried out for the first time, and the initial assessment data output by the preset risk assessment model is used as the risk assessment data.

[0041] In step S430 of some embodiments, if the risk assessment mode is a non-first-time insurance assessment mode, it means that a risk assessment has been conducted on the insured object before, and historical assessment data of the insured object can be obtained. The historical assessment data is the assessment data obtained by conducting risk assessments on the insured object in the past historical insurance periods. The initial assessment data and the historical assessment data are subjected to time-series correlation analysis to identify the risk evolution trend, correct the single assessment bias, and output the comprehensive risk assessment result to obtain the risk assessment data.

[0042] It is important to note that the initial assessment data and historical assessment data must be of the same data type. For example, both historical and initial assessment data may be risk categories or risk scores. If the data types are different, they need to be converted to the same data type.

[0043] If both the initial assessment data and the historical assessment data are of the risk category type, then the risk category that appears most frequently is selected as the risk assessment data. For example, if the initial assessment data shows 3 instances of low risk and 1 instance of high risk, and the initial assessment data is low risk, then the frequency of low risk is 4, and the frequency of high risk is 1. Since the frequency of low risk is greater than that of high risk, low risk is determined as the risk assessment data.

[0044] If both the initial assessment data and the historical assessment data are data types of risk scores, then the average of the initial assessment data and the historical assessment data is calculated, and this average is used as the risk assessment data. For example, if the initial assessment data is 70, and the historical assessment data are 40, 30, and 60, then the risk assessment data is (70+40+30+60) / 4=50.

[0045] Steps S410 to S430 above, by integrating current assessment data and historical assessment data, can achieve a more accurate risk assessment.

[0046] Insurance data, due to privacy concerns, cannot be centrally shared and is scattered across various insurance companies. Furthermore, the data sample distribution within each insurance company is unbalanced, with fewer high-risk samples. Insurance companies can only train risk assessment models using their local insurance data, resulting in insufficient model training, difficulty in covering all risk scenarios, and poor generalization performance. This application's embodiment utilizes federated learning technology to train the risk assessment model, improving model training performance while ensuring the privacy of insurance data.

[0047] Please see Figure 5 In some embodiments, the training process of the preset risk assessment model may include, but is not limited to, steps S510 to S530: Step S510: Send the preset initial evaluation model to at least two terminals; Step S520: Obtain a reference evaluation model obtained by the terminal through local training of the preset initial evaluation model; wherein, the reference evaluation model is trained by the terminal based on local sample data, and the local sample data includes sample insurance data and risk labels of sample insurance data; Step S530: Aggregate model parameters of the reference assessment model to obtain a preset risk assessment model.

[0048] In step S510 of some embodiments, the model parameters are initialized to obtain a preset initial evaluation model, and the preset initial evaluation model is sent to at least two terminals, which are insurance institution nodes participating in federated learning.

[0049] In step S520 of some embodiments, each terminal trains a preset initial evaluation model based on its local sample data to obtain a reference evaluation model, and then uploads the reference evaluation model. The local sample data includes the terminal's local sample insurance data and risk labels for the sample insurance data. The risk labels indicate a risk category or risk score.

[0050] During the upload of the reference evaluation model, differential privacy or homomorphic encryption techniques can be used to protect the model parameters and prevent the leakage of insurance data information. Furthermore, a trusted execution environment or secure multi-party computation can be introduced to further ensure data security.

[0051] In step S530 of some embodiments, an aggregation algorithm is used to perform a weighted average of the model parameters of the reference evaluation models trained by each terminal to obtain an aggregated reference evaluation model. The aggregated reference evaluation model is sent to the terminal, and steps S520 to S530 are repeated until the aggregated reference evaluation model converges to obtain a preset risk assessment model.

[0052] Through steps S510 to S530, the insurance companies can jointly optimize the model parameters without leaving the data domain, thereby improving the model's ability to identify rare risks and its overall generalization performance.

[0053] Please see Figure 6 In some embodiments, step S250 may include, but is not limited to, steps S610 to S630: Step S610: If the number of historical insurance applications is 0, then classify the insured objects according to the insurance data to obtain the object group category of the insured objects; Step S620: Obtain the first historical behavioral characteristics of the target group category and the current behavioral characteristics of the insured object; Step S630: Construct insurance behavior characteristics based on the first historical behavior characteristics and the current behavior characteristics.

[0054] In step S610 of some embodiments, a group category and its group behavior characteristics are determined. The group category is used to identify the type of insured individuals in terms of insurance attributes, such as high-risk practitioners or those concerned about chronic diseases. Group behavior characteristics are the common patterns exhibited by the group category in its insurance behavior, such as chronic disease-concerned individuals purchasing health insurance before their health deteriorates, or high-risk practitioners purchasing accident insurance in a short period of time. Group behavior characteristics can include insurance frequency, insurance type preference, insurance purchase patterns, policy cancellation behavior, and claims behavior.

[0055] The process involves large-scale collection of insurance application records from multiple pre-defined insured individuals. These records may contain multiple insurance applications from the same individuals over a given period. Behavioral features are extracted from these records using feature extraction networks (such as the Transformer model or BERT model), resulting in insurance application feature vectors for each pre-defined individual. Clustering algorithms, such as K-means clustering, are then used to analyze these feature vectors, identifying cluster categories and their feature centers. The feature centers are the average of the features from multiple insurance application feature vectors within each cluster category. Finally, the cluster categories are mapped to group categories, and the feature centers are used as the group's behavioral characteristics.

[0056] If the number of historical insurance applications is 0, behavioral features are extracted from the insurance application data using a feature extraction network to obtain the current behavioral features. The similarity between the current behavioral features and the behavioral features of each group is calculated, and the behavioral feature of the group with the highest similarity is taken as the first historical behavioral feature. The group category to which the first historical behavioral feature belongs is taken as the target group category of the insured.

[0057] In step S620 of some embodiments, the current behavioral characteristics can reflect the static attributes of the current insured object, such as illness or insurance products, but cannot reflect dynamic behaviors such as whether they frequently apply for insurance or whether they conceal their health conditions. Since the insured object has no historical insurance records, this application embodiment obtains first historical behavioral characteristics to reflect the general behavioral patterns of this type of object. By combining the first historical behavioral characteristics with the current behavioral characteristics of the insured object, the insurance behavior of the insured object can be accurately characterized.

[0058] In step S630 of some embodiments, feature fusion is performed on the first historical behavior features and the current behavior features to obtain the insurance application behavior features of the insured object. Feature fusion can be performed using methods such as feature concatenation, feature weighting, and attention fusion.

[0059] Through the above steps S610 to S630, the insurance behavior characteristics of the first-time insured can be accurately constructed.

[0060] Please see Figure 7 In some embodiments, step S250 may include, but is not limited to, steps S710 to S730: Step S710: If the number of historical insurance purchases is greater than 0, then obtain the second historical behavioral characteristics and current behavioral characteristics of the insured object; Step S720: Calculate the feature deviation between the current behavioral feature and the second historical behavioral feature; Step S730: Construct insurance behavior characteristics based on the second historical behavior characteristics and the characteristic deviation.

[0061] In step S710 of some embodiments, if the number of historical insurance applications is greater than 0, the historical insurance application records of the insured object in the historical period are obtained. The historical insurance application records include multiple insurance application data in that historical period. A feature extraction network is used to extract behavioral features from the historical insurance application records to obtain the dynamic behavior of the insured object in the historical period, resulting in a second historical behavioral feature. Then, a feature extraction network is used to extract behavioral features from the insurance application data to obtain the current behavioral details of the insured object, resulting in the current behavioral feature.

[0062] In step S720 of some embodiments, the feature deviation between the current behavior feature and the second historical behavior feature is calculated, and the difference between the current insurance behavior and the historical insurance behavior is quantified based on the feature deviation to reflect the behavior changes of the insured object, thereby capturing abnormal insurance behavior.

[0063] Specifically, for each feature dimension, the absolute difference between the current behavioral feature and the second historical behavioral feature is calculated. The dimensional weights of the feature dimensions are obtained; the higher the dimensional weight, the more important the feature dimension. The dimensional weights and absolute difference values ​​are then weighted to calculate the feature deviation. The formula for calculating the feature deviation is as follows: , in, Indicates the degree of deviation of the feature; Indicates the first Each feature dimension; Indicates the number of feature dimensions; Indicates the first Dimension weights for each feature dimension; Indicates the current behavioral characteristic in the th order. Feature values ​​of each feature dimension; Indicates the second historical behavioral characteristic in the first Feature values ​​of each feature dimension.

[0064] In step S730 of some embodiments, the feature deviation can reflect changes in the insured's behavior, such as a change in the type of insurance due to deteriorating health or an increase in the sum insured due to increased income. By combining the second historical behavioral features and the feature deviation, the individual profile of the insured can be dynamically updated, achieving accurate construction of the insured's behavioral features. Specifically, feature fusion is performed on the second historical behavioral features and the feature deviation to obtain the insured's behavioral features. Feature fusion can employ methods such as feature concatenation, feature weighting, and attention fusion.

[0065] Through the above steps S710 to S730, the insurance behavior characteristics of non-first-time insured individuals can be accurately constructed.

[0066] In related technologies, the insurance application process uses a uniform template. Regardless of health status, application channel, or product type, all applicants must complete the same steps, such as health declaration, information confirmation, and payment. This approach fails to simplify the process for low-risk users or provide precise guidance for high-risk users, resulting in rigid and inflexible steps. To improve application flexibility, this application's embodiment automatically generates an application process tailored to the applicant based on their application behavior characteristics and risk assessment data. This allows for the execution of the application process according to the chosen process, thereby enhancing application flexibility.

[0067] Please see Figure 8 In some embodiments, step S260 may include, but is not limited to, steps S810 to S830: Step S810: Determine the object status of the insured based on the characteristics of the insurance application behavior and risk assessment data; Step S820: Calculate the action score of each candidate process generated action in the preset process generation action space based on the object state; Step S830: Select the candidate process with the highest action score to generate the action and generate the insurance application process.

[0068] In step S810 of some embodiments, a reinforcement learning strategy is used to generate the insurance application process. In reinforcement learning, states and actions constitute the basic units for the interaction between the agent and the environment. The insurance application behavior characteristics and risk assessment data are used as states to obtain the object state of the insured object.

[0069] In step S820 of some embodiments, the object state is input into a deep Q-network for action reasoning. The deep Q-network outputs the Q-value of each candidate process generation action in the preset process generation action space, and the Q-value is used as the action score of the candidate process generation action. The preset process generation action space refers to a finite set of actions predefined based on insurance business rules for generating the insurance application process. The set contains multiple candidate process generation actions. Candidate process generation actions are actions that generate the insurance application process, such as information collection, underwriting depth adjustment (e.g., simplified underwriting, standard underwriting, manual underwriting, etc.), additional value-added services (e.g., exclusive consultation, product recommendation), setting the policy effective time, risk control verification, etc.

[0070] In step S830 of some embodiments, the candidate process generating action with the highest action score is selected to obtain the optimal action in the current state. The candidate process generating action is then executed to orchestrate the process and generate the insurance application process. For example, executing the simple underwriting action results in the insurance application process as follows: identity verification → basic health declaration → automatic pricing → immediate underwriting; executing the standard underwriting action results in the insurance application process as follows: identity verification → complete health declaration → manual review → pricing → underwriting.

[0071] Through the above steps S810 to S830, the insurance application process can be adaptively adjusted according to the insured, thus achieving a flexible insurance application process design.

[0072] Please see Figure 9 This application also provides an insurance application device that can implement the above-mentioned insurance application method. The insurance application device includes: The first acquisition module 910 is used to acquire the basic information of the insured and the information of the insured product; wherein, the basic information of the insured includes the identity information and the health information of the insured. The information extraction module 920 is used to extract information from the object's identity information, object's health information, and insurance product information, to obtain the identity field of the object's identity information, the health field of the object's health information, and the product field of the insurance product information, and to fill the identity field, health field, and product field into the preset insurance template to obtain the insurance data; The second acquisition module 930 is used to acquire the historical number of insurance policies of the insured; Risk assessment module 940 is used to conduct risk assessments on insured individuals based on insurance data and historical insurance frequency, and to obtain risk assessment data. Feature construction module 950 is used to construct the insurance behavior features of the insured based on the number of historical insurance applications; The insurance application module 960 is used to generate an insurance application process based on insurance application behavior characteristics and risk assessment data, and to conduct insurance applications according to the application process.

[0073] This application also provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the above-described insurance application method. This electronic device can be any smart terminal, including tablet computers, in-vehicle computers, etc.

[0074] Please see Figure 10 , Figure 10 The hardware structure of an electronic device according to another embodiment is illustrated. The electronic device includes: The processor 1010 can be implemented using a general-purpose central processing unit (CPU), microprocessor, application specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this application. The memory 1020 can be implemented as a read-only memory (ROM), static storage device, dynamic storage device, or random access memory (RAM). The memory 1020 can store the operating system and other applications. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 1020 and is called and executed by the processor 1010 to implement the insurance application method of the embodiments of this application. The input / output interface 1030 is used to implement information input and output; The communication interface 1040 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, network cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.). Bus 1050 transmits information between various components of the device (e.g., processor 1010, memory 1020, input / output interface 1030, and communication interface 1040); The processor 1010, memory 1020, input / output interface 1030 and communication interface 1040 are connected to each other within the device via bus 1050.

[0075] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described insurance application method.

[0076] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0077] The insurance application method, device, electronic equipment, and computer storage medium provided in this application generate personalized insurance application processes for the insured based on insurance behavior characteristics and risk assessment data, and perform insurance application operations according to the personalized insurance application process, rather than following fixed process steps, which greatly improves the flexibility of insurance application.

[0078] The embodiments described in this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided by the embodiments of this application. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this application are also applicable to similar technical problems.

[0079] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of this application, and may include more or fewer steps than shown, or combine certain steps, or different steps.

[0080] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0081] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or suitable combinations thereof.

[0082] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0083] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0084] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of the units described above is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0085] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0086] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0087] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing programs, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0088] It should be noted that any AI models, software tools, or components not belonging to this company appearing in the embodiments of this application are merely illustrative examples and do not represent actual use. All user personal information involved in the embodiments of this application has been authorized (with the knowledge and consent) by the relevant parties or has been fully authorized by all parties, and the executing entity may obtain it through various legal and compliant means. The collection, storage, use, processing, transmission, provision, and disclosure of the information, data, and signals involved all comply with relevant laws and regulations and do not violate public order and good morals.

[0089] The preferred embodiments of the present application have been described above with reference to the accompanying drawings, but this does not limit the scope of the claims of the present application. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and substance of the embodiments of the present application shall be within the scope of the claims of the present application.

Claims

1. An underwriting method, characterized by, The method includes: Obtain the basic information of the insured and the information of the insured product; wherein, the basic information of the insured includes the insured's identity information and the insured's health information; Information is extracted from the object's identity information, the object's health information, and the insurance product information to obtain the identity field of the object's identity information, the health field of the object's health information, and the product field of the insurance product information. The identity field, the health field, and the product field are then filled into a preset insurance template to obtain insurance data. Obtain the historical number of insurance policies for the insured individual; Based on the insurance data and the historical number of insurance applications, a risk assessment is performed on the insured object to obtain risk assessment data; Based on the historical number of insurance applications, the insurance application behavior characteristics of the insured object are constructed; An insurance application process is generated based on the characteristics of the insurance application behavior and the risk assessment data, and the insurance application is carried out in accordance with the insurance application process.

2. The method of claim 1, wherein, The risk assessment of the insured based on the insurance data and the historical number of insurance applications, to obtain risk assessment data, includes: The risk assessment model is determined based on the number of historical insurance policies. The risk assessment data is obtained by conducting a risk assessment on the insured object based on the insurance data and the risk assessment model.

3. The method according to claim 2, characterized in that, The risk assessment model includes a first-time insurance assessment model or a non-first-time insurance assessment model. The step of conducting a risk assessment on the insured based on the insurance data and the risk assessment model to obtain the risk assessment data includes: The insurance data is assessed using a pre-set risk assessment model to obtain initial assessment data. If the risk assessment mode is the first-time insurance assessment mode, then the initial assessment data will be used as the risk assessment data; If the risk assessment mode is the non-first-time insurance assessment mode, then the historical assessment data of the insured object is obtained, and the risk assessment data is determined based on the initial assessment data and the historical assessment data.

4. The method according to claim 3, characterized in that, The preset risk assessment model is trained according to the following steps: Send the preset initial evaluation model to at least two terminals; Obtain a reference evaluation model obtained by the terminal through local training of the preset initial evaluation model; wherein, the reference evaluation model is trained by the terminal based on local sample data, and the local sample data includes sample insurance data and risk labels of the sample insurance data; The model parameters of the reference assessment model are aggregated to obtain the preset risk assessment model.

5. The method according to any one of claims 1 to 4, characterized in that, The process of constructing the insurance behavior characteristics of the insured based on the historical number of insurance applications includes: If the number of historical insurance purchases is 0, then the insured objects are classified according to the insurance purchase data to obtain the object group category of the insured objects; Obtain the first historical behavioral characteristics of the object group category and the current behavioral characteristics of the insured object; The insurance application behavior characteristics are constructed based on the first historical behavior characteristics and the current behavior characteristics.

6. The method according to any one of claims 1 to 4, characterized in that, The process of constructing the insurance behavior characteristics of the insured based on the historical number of insurance applications includes: If the number of historical insurance purchases is greater than 0, then the second historical behavioral characteristics and current behavioral characteristics of the insured object are obtained; Calculate the feature deviation between the current behavior feature and the second historical behavior feature; The insurance behavior features are constructed based on the second historical behavior features and the feature deviation.

7. The method according to any one of claims 1 to 4, characterized in that, The process of generating an insurance application based on the characteristics of the insurance application behavior and the risk assessment data includes: The object status of the insured is determined based on the characteristics of the insurance purchase behavior and the risk assessment data; Calculate the action score of each candidate process generated action in the preset process generation action space based on the object state; The candidate process with the highest action score is selected to generate the insurance application process.

8. An insurance application device, characterized in that, The device includes: The first acquisition module is used to acquire the basic information of the insured object and the information of the insured product; wherein, the basic information of the object includes the object's identity information and the object's health information; The information extraction module is used to extract information from the object's identity information, the object's health information, and the insurance product information to obtain the identity field of the object's identity information, the health field of the object's health information, and the product field of the insurance product information. The identity field, the health field, and the product field are then filled into a preset insurance template to obtain insurance data. The second acquisition module is used to acquire the historical number of insurance policies of the insured object; The risk assessment module is used to assess the risk of the insured based on the insurance data and the number of historical insurance applications, and to obtain risk assessment data. The feature construction module is used to construct the insurance behavior features of the insured object based on the number of historical insurance purchases; The insurance application module is used to generate an insurance application process based on the insurance application behavior characteristics and the risk assessment data, and to apply for insurance according to the insurance application process.

9. An electronic device, characterized in that, The electronic device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the method according to any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the method described in any one of claims 1 to 7.

Images