CN119722344A

CN119722344A - Insurance pricing method, device, electronic equipment and medium based on data fusion

Info

Publication number: CN119722344A
Application number: CN202411793117.XA
Authority: CN
Inventors: 王利云
Original assignee: Ping An Property and Casualty Insurance Company of China Ltd
Current assignee: Ping An Property and Casualty Insurance Company of China Ltd
Priority date: 2024-12-05
Filing date: 2024-12-05
Publication date: 2025-03-28
Anticipated expiration: 2044-12-05
Also published as: CN119722344B

Abstract

The embodiments of the present application provide an insurance pricing method, device, electronic device and medium based on data fusion, which belongs to the field of financial technology. The method includes: in response to the user's pricing instruction, obtaining the basic product information corresponding to the pricing instruction through the pricing system; receiving the identity information sent by the user through the pricing system, and determining the user-related data and the credit platform bound to the pricing system according to the identity information; in response to the user's authorization instruction, obtaining the credit score corresponding to the user through the credit platform; determining the comprehensive discount parameters according to the credit score and the user-related data; conducting risk assessment on the basic product information, credit score and user-related data to obtain the risk assessment results; conducting pricing analysis on the risk assessment results and the comprehensive discount parameters through the pricing model to obtain the target insurance pricing. The embodiments of the present application fully reflect the customer's risk level through multi-dimensional data, and realize dynamic adjustment of the customer's insurance pricing.

Description

Insurance pricing method and device based on data fusion, electronic equipment and medium

Technical Field

The application relates to the technical field of financial science and technology, in particular to a insurance pricing method, a device, electronic equipment and a medium based on data fusion.

Background

The rapid development of unmanned aerial vehicle technology has promoted the development of low-altitude economy, and unmanned aerial vehicle has played important roles in a plurality of industries such as commodity circulation, agriculture, infrastructure inspection. However, since the unmanned aerial vehicle flight involves a plurality of risk factors, such as operation habits of the aircraft, weather conditions, terrain complexity, etc., the existing insurance pricing system often adopts a long-term fixed pricing mode, and the individual demands of the clients cannot be flexibly reflected. This stiffness in pricing model not only limits the personalized development of insurance products, but also ignores the actual impact of customer behavior on risk levels. The insurance company can not make full use of the comprehensive data of the client to formulate more reasonable insurance pricing, so that the real risk level of the client can not be fully reflected, the client can not enjoy dynamic premium preference in different use stages, and the insurance purchasing experience and viscosity of the client are reduced.

Disclosure of Invention

The embodiment of the application mainly aims to provide a data fusion-based insurance pricing method, a data fusion-based insurance pricing device, electronic equipment and a data fusion-based insurance pricing medium, which can fully reflect the risk level of a customer through multidimensional data and realize dynamic adjustment of insurance pricing of the customer.

To achieve the above object, a first aspect of an embodiment of the present application provides a method for pricing insurance based on data fusion, the method including:

Responding to a pricing instruction of a user, and acquiring product basic information corresponding to the pricing instruction through a pricing system;

receiving identity information sent by the user through the pricing system, and determining user associated data and a credit platform bound with the pricing system according to the identity information, wherein the user associated data is used for representing purchase records and claim settlement information of the user;

responding to the authorization instruction of the user, and acquiring a credit score corresponding to the user through the credit platform;

determining a composite discount parameter from the credit score and the user-associated data;

performing risk assessment on the product basic information, the credit score and the user associated data to obtain a risk assessment result;

And carrying out pricing analysis on the risk assessment result and the comprehensive discount parameters through a preset pricing model to obtain target insurance pricing.

In some embodiments, the determining the composite discount parameter from the credit score and the user-associated data comprises:

Determining the highest credit score and the lowest credit score of the user in a preset time period according to the credit score;

Normalizing the credit score according to the highest credit score and the lowest credit score to obtain a target credit score;

Performing exponential decay operation on the target credit score based on a preset adjustment parameter to obtain a credit score discount, wherein the adjustment parameter is used for controlling the decay rate of the discount;

determining the number of claims, dangerous seed weight and total number of insurance according to the user associated data;

determining a correlation discount according to the number of claims, the dangerous seed weight and the total policy number;

And determining a comprehensive discount parameter according to the credit scoring discount and the relevance discount.

In some embodiments, said determining the composite discount parameter from the credit score discount and the relevance discount comprises:

Determining a first priority corresponding to the credit scoring discount and determining a second priority corresponding to the relevance discount;

Setting a first weight parameter corresponding to the credit scoring discount according to the first priority, and setting a second weight parameter corresponding to the relevance discount according to the second priority;

Multiplying the credit score discount with the first weight parameter to obtain a credit score, and multiplying the correlation discount with the second weight parameter to obtain a correlation score;

and determining comprehensive discount parameters according to the credit score and the relevance score.

In some embodiments, the performing risk assessment on the product basic information, the credit score and the user associated data to obtain a risk assessment result includes:

Determining product identification and product history data according to the product basic information, and carrying out association analysis on the credit score and the user association data according to the product identification to obtain an association relationship;

taking the product identifier as a node, and connecting a plurality of nodes according to the association relationship to construct a network map;

Feature fusion is carried out on the network map through a preset graph neural network, so that feature representation is obtained;

carrying out feature processing on the product history data to obtain time sequence features and non-time sequence features;

Carrying out data prediction on the time sequence characteristics through a preset long-short-period memory network to obtain a hidden state;

And carrying out feature enhancement on the hidden state based on a preset attention mechanism to obtain comprehensive information, and calculating a risk score of the time sequence feature according to the comprehensive information and the non-time sequence feature to obtain a risk assessment result.

In some embodiments, the feature fusion of the network map through a preset graph neural network to obtain a feature representation includes:

For each node in the network map, assigning a feature vector to the node;

Selecting a test node from the network map, and determining an initial feature vector of the test node and a neighbor node set around the test node;

Performing linear transformation on neighbor nodes in the neighbor node set through the graph neural network to obtain a first parameter, and performing linear transformation on the test node to obtain a second parameter;

Determining a neighbor feature vector set corresponding to the neighbor node set, determining a first feature representation according to the initial feature vector, and determining a second feature representation according to the neighbor feature vector set;

Inputting the first parameter, the second parameter, the first feature representation and the second feature representation into the graph neural network, so that the graph neural network performs hierarchical updating on the test node based on the first parameter, the second parameter, the first feature representation and the second feature representation to obtain a feature representation.

In some embodiments, the feature enhancement is performed on the hidden state based on a preset attention mechanism to obtain comprehensive information, and a risk score of the time series feature is calculated according to the comprehensive information and the non-time series feature to obtain a risk assessment result, where the risk assessment result includes:

calculating the attention score of the hidden state based on a preset attention mechanism;

determining the attention weight of the current time step according to the attention score;

the context of the time series feature is weighted and summed based on the attention weight and the hidden state to obtain comprehensive information;

splicing the comprehensive information and the non-time sequence features to obtain splicing features;

and carrying out risk scoring on the spliced features to obtain a risk assessment result.

In some embodiments, the pricing analysis on the risk assessment result and the comprehensive discount parameter through a preset pricing model, to obtain a target insurance pricing, includes:

determining a basic premium applied by the user according to the user association data;

And carrying out pricing analysis according to the basic premium, the risk assessment result and the comprehensive discount parameter to obtain target insurance pricing.

In order to achieve the above object, a second aspect of the embodiments of the present application provides an insurance pricing device based on data fusion, where the device includes:

the data acquisition module is used for responding to the pricing instruction of the user and acquiring product basic information corresponding to the pricing instruction through the pricing system;

The data determining module is used for receiving the identity information sent by the user through the pricing system, and determining user association data and a credit platform bound with the pricing system according to the identity information, wherein the user association data is used for representing the purchase record and the claim settlement information of the user;

the credit score acquisition module is used for responding to the authorization instruction of the user and acquiring the credit score corresponding to the user through the credit platform;

A discount calculation module for determining a comprehensive discount parameter according to the credit score and the user-associated data;

the risk assessment module is used for carrying out risk assessment on the product basic information, the credit score and the user associated data to obtain a risk assessment result;

And the pricing analysis module is used for carrying out pricing analysis on the risk assessment result and the comprehensive discount parameter through a preset pricing model to obtain target insurance pricing.

To achieve the above object, a third aspect of the embodiments of the present application provides an electronic device, which includes a memory and a processor, where the memory stores a computer program, and the processor executes the computer program to implement the insurance pricing method based on data fusion according to the first aspect.

To achieve the above object, a fourth aspect of the embodiments of the present application proposes a computer-readable storage medium storing a computer program which, when executed by a processor, implements the data fusion-based insurance pricing method according to the first aspect.

According to the insurance pricing method, the device, the electronic equipment and the storage medium based on data fusion, the pricing instruction of the user is responded, the product basic information corresponding to the pricing instruction is obtained through the pricing system, so that the basic information of an insurance product can be obtained, the basic information can be conveniently used as a reference for risk assessment later, identity information sent by the user is received through the pricing system, user association data and a credit platform bound with the pricing system are determined according to the identity information, purchase records of other products purchased by the user and current claim settlement records can be obtained, and therefore whether the user has a claim settlement or not and specific information of the claim settlement can be judged. Responding to an authorization instruction of a user, acquiring a credit score corresponding to the user through a credit platform, reflecting the credit condition of the user through the credit score, improving the accuracy of risk management, reducing risk loss, determining comprehensive discount parameters according to the credit score and user association data, reasonably setting discount parameters according to the credit condition, historical claim settlement condition and insurance purchase condition of the user, carrying out risk assessment on product basic information, the credit score and user association data, fully reflecting the risk level of the user through analysis of multidimensional data, obtaining a risk assessment result, realizing comprehensive assessment on insurance risk, carrying out pricing analysis on the risk assessment result and the comprehensive discount parameters through a preset pricing model, flexibly adjusting premium in combination with the insurance condition and actual risk of the user, and obtaining target insurance pricing, thereby improving more personalized premium benefits for the user. According to the embodiment of the application, through multidimensional data collection, fusion and analysis, the insurance rate of the user is dynamically adjusted, and through docking the third-party credit platform and integrating data such as purchase records of other insurance products, more personalized premium offers can be provided for the user.

Drawings

FIG. 1 is a flow chart of a data fusion based insurance pricing method provided by an embodiment of the application;

Fig. 2 is a flowchart of step S104 in fig. 1;

fig. 3 is a flowchart of step S206 in fig. 2;

fig. 4 is a flowchart of step S105 in fig. 1;

fig. 5 is a flowchart of step S403 in fig. 4;

fig. 6 is a flowchart of step S406 in fig. 4;

Fig. 7 is a flowchart of step S106 in fig. 1;

FIG. 8 is a schematic diagram of a data fusion-based insurance pricing device according to an embodiment of the application;

fig. 9 is a schematic diagram of a hardware structure of an electronic device according to an embodiment of the present application.

Detailed Description

The present application will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present application more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the application.

It should be noted that although functional block division is performed in a device diagram and a logic sequence is shown in a flowchart, in some cases, the steps shown or described may be performed in a different order than the block division in the device, or in the flowchart. The terms first, second and the like in the description and in the claims and in the above-described figures, are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order.

Unless defined otherwise, 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 the application only and is not intended to be limiting of the application.

First, several nouns involved in the present application are parsed:

Natural language processing (Natural Language Processing, NLP) NLP is a branch of artificial intelligence, which is a interdisciplinary of computer science and linguistics, and is often referred to as computational linguistics, where NLP is processed, understood, and applied in human language (e.g., chinese, english, etc.). Natural language processing includes parsing, semantic analysis, chapter understanding, and the like. Natural language processing is commonly used in the technical fields of machine translation, handwriting and print character recognition, voice recognition and text-to-speech conversion, information intent recognition, information extraction and filtering, text classification and clustering, public opinion analysis and opinion mining, and the like, and relates to data mining, machine learning, knowledge acquisition, knowledge engineering, artificial intelligence research, linguistic research related to language calculation, and the like.

According to the insurance pricing method and device based on data fusion, the electronic equipment and the storage medium, provided by the embodiment of the application, the risk level of the client can be fully reflected through multidimensional data, and the dynamic adjustment of the insurance pricing of the client can be realized.

The following embodiments are specifically described, first, to describe an insurance pricing method based on data fusion in an embodiment of the present application.

The embodiment of the application can acquire and process the related data based on the artificial intelligence technology. Wherein artificial intelligence (ARTIFICIAL INTELLIGENCE, AI) is the theory, method, technique, and application system that uses a digital computer or a digital computer-controlled machine to simulate, extend, and expand human intelligence, sense the environment, acquire knowledge, and use knowledge to obtain optimal results.

Artificial intelligence infrastructure technologies generally include technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing technologies, operation/interaction systems, mechatronics, and the like. The artificial intelligent software technology mainly comprises a computer vision technology, a robot technology, a biological recognition technology, a module management technology of an online guest receiving system, a natural language processing technology, machine learning/deep learning and other directions.

The embodiment of the application provides a insurance pricing method based on data fusion, and relates to the technical field of financial science and technology. The insurance pricing method based on data fusion provided by the embodiment of the application can be applied to a terminal, a server and software running in the terminal or the server. In some embodiments, the terminal may be a smart phone, a tablet computer, a notebook computer, a desktop computer, etc., the server may be configured as an independent physical server, may be configured as a server cluster or a distributed system formed by a plurality of physical servers, and may be configured as a cloud server for providing cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDNs, and basic cloud computing services such as big data and artificial intelligence platforms, and the software may be an application for implementing an insurance pricing method based on data fusion, but is not limited to the above form.

The application is operational with numerous general purpose or special purpose computer system environments or configurations. Such as a personal computer, a server computer, a hand-held or portable device, a tablet device, a multiprocessor system, a microprocessor-based system, a set top box, a programmable consumer electronics, a network PC, a minicomputer, a mainframe computer, a distributed computing environment that includes any of the above systems or devices, and the like. The application may be described in the general context of computer-executable instructions, such as program modules, being executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform particular tasks or implement particular abstract data types. The application may also be practiced in distributed computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules may be located in both local and remote computer storage media including memory storage devices.

In the embodiments of the present application, when related processing is performed according to user information, user behavior data, user history data, user location information, and other data related to user identity or characteristics, permission or consent of the user is obtained first, and the collection, use, processing, and the like of the data comply with related laws and regulations and standards of related countries and regions. In addition, when the embodiment of the application needs to acquire the sensitive personal information of the user, the independent permission or independent consent of the user is acquired through popup or jump to a confirmation page and the like, and after the independent permission or independent consent of the user is definitely acquired, the necessary relevant data of the user for enabling the embodiment of the application to normally operate is acquired.

In order to solve the above problems, the present embodiment provides a method, an apparatus, an electronic device, and a storage medium for insurance pricing based on data fusion, in response to a pricing instruction of a user, obtain product basic information corresponding to the pricing instruction through a pricing system, so as to obtain basic information of an insurance product, facilitate subsequent reference to risk assessment, receive identity information sent by the user through the pricing system, determine user association data and a credit platform bound to the pricing system according to the identity information, and obtain a purchase record of purchasing other products and a current claim settlement record of the user, so as to determine whether the user has a claim and specific information of the claim settlement. Responding to an authorization instruction of a user, acquiring a credit score corresponding to the user through a credit platform, reflecting the credit condition of the user through the credit score, improving the accuracy of risk management, reducing risk loss, determining comprehensive discount parameters according to the credit score and user association data, reasonably setting discount parameters according to the credit condition, historical claim settlement condition and insurance purchase condition of the user, carrying out risk assessment on product basic information, the credit score and user association data, fully reflecting the risk level of the user through analysis of multidimensional data, obtaining a risk assessment result, realizing comprehensive assessment on insurance risk, carrying out pricing analysis on the risk assessment result and the comprehensive discount parameters through a preset pricing model, flexibly adjusting premium in combination with the insurance condition and actual risk of the user, and obtaining target insurance pricing, thereby improving more personalized premium benefits for the user. According to the embodiment of the application, through multidimensional data collection, fusion and analysis, the insurance rate of the user is dynamically adjusted, and through docking the third-party credit platform and integrating data such as purchase records of other insurance products, more personalized premium offers can be provided for the user.

The following is a detailed description with reference to the accompanying drawings.

FIG. 1 is an alternative flow chart of a data fusion-based insurance pricing method provided by an embodiment of the application, where the method of FIG. 1 may include, but is not limited to, steps S101 through S106.

In step S101, in response to a pricing instruction of a user, product basic information corresponding to the pricing instruction is acquired through a pricing system.

In step S101 of some embodiments, a user may send a pricing instruction by triggering a related control on the pricing system, and in response to the pricing instruction of the user, obtain product base information corresponding to the pricing instruction through the pricing system, where the product base information includes base information of a product, such as a brand, model, market pricing, and the like, to facilitate subsequent reference as a base reference for risk assessment.

It should be noted that, the pricing system in the embodiment of the present application is provided with an intelligent question-answering module, where the intelligent question-answering module includes a plurality of preset questions, where the preset questions may be single-choice questions, multiple-choice questions or gap-filling questions. When the preset questions are single-choice questions or multi-choice questions, the user can select corresponding options to answer, and when the preset questions are blank-filling questions, the user can answer in answer areas corresponding to each preset question. For example, the intelligent question-answering module is provided with problems such as flight frequency, flight habit, knowledge understanding degree of the unmanned aerial vehicle and the like, and gives corresponding options, at this time, a user can answer the corresponding options according to different problems, and after the user answers, a corresponding pricing instruction is generated to perform the insurance pricing process.

Step S102, receiving identity information sent by a user through a pricing system, and determining user association data and a credit platform bound with the pricing system according to the identity information.

It should be noted that the user-associated data is used to characterize the purchase records and claims information of the user.

In step S102 of some embodiments, after obtaining the product basic information, the embodiment of the present application further receives identity information sent by the user through the pricing system, and determines user association data according to the identity information, so that other insurance products purchased or being held by the user at the insurance company, such as automobile insurance, pet insurance, health insurance, etc., can be determined through the user association data, and meanwhile, historical claim settlement conditions of the user before the user can be determined through the user association data, so that dynamic pricing can be performed by combining the purchase history and claim settlement history of the user later, and flexibility of insurance pricing is improved. Meanwhile, the embodiment of the application also determines a credit platform bound with the pricing system according to the identity information, such as a payment treasury platform, a WeChat platform and the like, so that the credit of the user can be acquired conveniently.

Step S103, responding to the authorization instruction of the user, and obtaining the credit score corresponding to the user through a credit platform.

In step S103 of some embodiments, in response to the authorization instruction of the user, the pricing system sends a query request instruction to the credit platform, and the credit platform queries the credit score corresponding to the identity information of the user and returns a corresponding result.

It can be understood that the credit score in the embodiment of the present application may be determined according to the type of the credit platform, for example, when the credit platform is a payment treasured, the credit score at this time is a sesame credit score of the payment treasured, when the credit platform is a WeChat, the credit score at this time is a WeChat credit score, and the like, which is not particularly limited.

Notably, the pricing system in the embodiment of the present application updates the credit score in real time, so as to ensure the accuracy of the assessment, wherein the higher the credit score, the greater the premium benefit obtained by the user.

Step S104, determining the comprehensive discount parameters according to the credit scores and the user association data.

In step S104 of some embodiments, the comprehensive discount parameters are determined according to the credit score and the user associated data, so that a higher credit score can be obtained to obtain a more obvious discount, and a credit score discount can be obtained, meanwhile, a sharp change caused by a logarithmic function in a low time sharing can be avoided, smooth calculation of premium offers of users can be realized, the discount difference between different credit scores is ensured to be not significant, and discounts of customers with low credit score and customers with high credit score can be balanced.

Step S105, carrying out risk assessment on the product basic information, the credit score and the user associated data to obtain a risk assessment result.

In step S105 of some embodiments, risk assessment is performed on the product basic information, the credit score and the user associated data to obtain a risk assessment result, so that comprehensive risk assessment can be realized by combining multidimensional information, accuracy of risk assessment is improved, and comprehensive analysis on historical data is realized.

And S106, pricing analysis is carried out on the risk assessment result and the comprehensive discount parameters through a preset pricing model, and target insurance pricing is obtained.

In step S106 of some embodiments, pricing analysis is performed on the risk assessment result and the comprehensive discount parameter through a preset pricing model, so that comprehensive risk assessment can be realized by combining multidimensional information, target insurance pricing is obtained, accuracy of risk assessment is improved, and comprehensive analysis on historical data is realized.

In some embodiments, the present application further records data by using a blockchain technique, so as to ensure transparency and non-tamper-resistance of each premium calculation and adjustment process. And asynchronous uplink operation can be carried out on the data, specifically, the data enters the queue in the form of text, file or picture in consideration of the influence of the block chain performance, and the uplink is completed within 5 minutes in the queue, so that the efficiency is ensured.

Specifically, the data for each premium calculation and adjustment process is signed and uploaded to the blockchain. The form of the data uplink comprises text, files or pictures and the like, and each link of the premium adjustment is ensured to be non-tamper-proof and traceable. Considering the limitation of the operation performance on the chain, the system waits for the data to be wound in the queue through an asynchronous winding mechanism, so that the data writing is ensured to be completed efficiently. And the system can combine the data of the third party platform and the internal history data through the data stored on the chain, and adopts Bayesian optimization to carry out real-time pricing adjustment, thereby ensuring individuation and fairness of client premium.

In some embodiments, the embodiment of the application can flexibly adjust the premium according to the real-time operation behavior of the user, the user associated data and the credit score, so as to ensure that the premium accords with the actual risk level of the user. Users of different task types, operating environments, and credit scores will get different premium adjustment policies. For example, a drone user performs multiple high risk tasks, but the credit score is high, and the system gives an additional 5% premium discount based on his credit record.

Referring to fig. 2, in some embodiments, step S104 may further include, but is not limited to, steps S201 to S206.

Step S201, determining the highest credit score and the lowest credit score of the user in the preset time according to the credit score.

Step S202, normalizing the credit score according to the highest credit score and the lowest credit score to obtain a target credit score.

And step S203, carrying out exponential decay operation on the target credit score based on the preset regulation parameters to obtain a credit score discount.

It should be noted that the adjustment parameter is used to control the discounted decay rate.

In steps S201 to S203 of some embodiments, in determining the comprehensive discount parameters according to the credit score and the user associated data, the embodiments of the present application determine the highest credit score and the lowest credit score of the user within the preset duration according to the credit score, by determining the lowest and the highest credit scores, a standardized comparison standard can be provided for the credit status of different users or entities, so as to facilitate the subsequent accurate adjustment of premium pricing, and then normalize the credit score according to the highest credit score and the lowest credit score, so that the value ranges of different features are consistent, thereby accelerating the convergence process of the optimization algorithm such as gradient descent, so as to obtain the target credit score, and then perform an exponential decay operation on the target credit score based on the preset adjustment parameters, so that the higher credit score can obtain a more obvious discount, so as to obtain the credit score discount, and meanwhile avoid the abrupt change caused by the logarithmic function in low time, realize the smooth calculation of premium offers of the user, ensure that the discount difference between different credit scores is not significant, and balance the low-score and high-score customer discount.

Specifically, in the embodiment of the present application, the exponential decay operation of the target credit score is represented as follows:

;

Wherein, Is the normalized credit score, i.e., the target credit score; is the highest credit score; Is the lowest credit score; Is the original credit score.

;

Wherein, Is the normalized credit score; Is a tuning parameter, used to control the discounted decay rate, Is a credit score discount.

It will be appreciated that customers with higher credit scores in embodiments of the application enjoy greater discounts while ensuring that low credit score customers are not discounted too low. For example, a client with a credit score of 700 would have a reduced discount spread compared to a client with a credit score of 800.

And step S204, determining the number of claims, the risk weight and the total insurance number according to the user associated data.

Step S205, determining the correlation discount according to the number of the claims, the risk weight and the total policy number.

Step S206, determining the comprehensive discount parameters according to the credit score discount and the correlation discount.

In steps S204 to S206 of some embodiments, the embodiments of the present application further determine the number of claims, the risk weight and the total policy according to the user associated data, so as to implement comprehensive collection of the historical claims data of the user and the purchased insurance data, and then determine the relevant discount according to the number of claims, the risk weight and the total policy, so that the premium corresponding to the user who has purchased more products or has less claims is higher, and implement flexible adjustment of premium pricing, and then determine the comprehensive discount parameter according to the credit score discount and the relevant discount, and implement dynamic adjustment of the discount parameter through comprehensive combination of multiple data, so as to determine the discount parameter by combining the purchased product condition, the historical claims condition and the credit score condition of the user, and improve the accuracy and flexibility of insurance pricing.

Specifically, the correlation discount in the embodiment of the present application is calculated as follows:

;

Wherein, Is a correlation discount; Is a dangerous seed weight; The number of claims is settled; Is the total guard number.

It can be understood that in the embodiment of the application, the dangerous seed of the user client without the claim record obtains the greatest weight preference, and the dangerous seed with higher claim settlement times reduces the corresponding discount proportion.

It should be noted that, the preset duration in the embodiment of the present application may be set according to the needs of the user, for example, one year, two years, or half month, etc., and the embodiment of the present application is not limited specifically.

Referring to fig. 3, in some embodiments, step S206 may further include, but is not limited to, steps S301 to S304.

Step S301, a first priority corresponding to the credit score discount is determined, and a second priority corresponding to the relevance discount is determined.

Step S302, a first weight parameter corresponding to the credit score discount is set according to the first priority, and a second weight parameter corresponding to the relevance discount is set according to the second priority.

Step S303, the credit score discount is multiplied by the first weight parameter to obtain a credit score, and the correlation discount is multiplied by the second weight parameter to obtain a correlation score.

Step S304, determining the comprehensive discount parameters according to the credit score and the relevance score.

In steps S301 to S304 of some embodiments, in determining the comprehensive discount parameters according to the credit score discount and the correlation discount, the embodiments of the present application determine the first priority corresponding to the credit score discount and determine the second priority corresponding to the correlation discount, so as to determine the priority relationship between the credit score discount and the correlation discount, then set the first weight parameter corresponding to the credit score discount according to the first priority and set the second weight parameter corresponding to the correlation discount according to the second priority, so that different weight parameters can be set according to different priorities, the discount with higher priority can be set with a larger weight parameter, the discount with lower priority can be set with a smaller weight parameter, then the credit score discount is multiplied with the first weight parameter to obtain the credit score, and the correlation score is multiplied with the second weight parameter to obtain the correlation score, and then the comprehensive discount parameters are determined according to the credit score and the correlation score, so that the comprehensive parameters can be reasonably determined, and the service for the user can be provided with individuation.

Specifically, the operation of determining the comprehensive discount parameters according to the embodiment of the present application is as follows:

;

Wherein, Scoring discounts for credits; Is a relatedness discount; Is a first weight parameter; is a second weight parameter.

It may be understood that, in the embodiment of the present application, the sum of the weights of the first weight parameter and the second weight parameter is 1, and the first priority and the second priority may be set according to the needs of the user, for example, the first priority is set to be higher than the second priority, where the first weight parameter is greater than the second weight parameter, and the embodiment of the present application is not limited specifically.

Referring to fig. 4, in some embodiments, step S105 may further include, but is not limited to including, step S401 to step S406.

Step S401, determining product identification and product history data according to the product basic information, and carrying out association analysis on credit scores and user association data according to the product identification to obtain association relation.

Step S402, product identifiers are used as nodes, and a plurality of nodes are connected according to the association relation to construct a network map.

And S403, carrying out feature fusion on the network map through a preset map neural network to obtain feature representation.

And step S404, carrying out feature processing on the product history data to obtain time sequence features and non-time sequence features.

Step S405, data prediction is performed on the time series characteristics through a preset long-short-period memory network to obtain a hidden state.

Step S406, feature enhancement is performed on the hidden state based on a preset attention mechanism to obtain comprehensive information, and a risk score of the time sequence feature is calculated according to the comprehensive information and the non-time sequence feature to obtain a risk assessment result.

In steps S401 to S406 of some embodiments, in the process of performing risk assessment on product basic information, credit score and user association data, the embodiment of the application determines product identifiers and product history data according to the product basic information, performs association analysis on the credit score and the user association data according to the product identifiers, analyzes association degrees between the credit score and the user association data and the product identifiers to obtain association relations, uses the product identifiers as nodes, and connects a plurality of nodes according to the association relations to construct a network map, wherein each node represents different products, edges between the nodes represent common characteristics without using the association between the nodes, and edges existing between the two nodes are used as data sources, so that the correspondence between the products can be clearly described, and the relation between the product data can be intuitively displayed through the map. And then, carrying out feature fusion on the network map through a preset graph neural network, and improving the model convergence speed while ensuring the balance of the feature fusion to obtain the feature representation. Then carrying out feature processing on the product history data, converting the product history data into time series features, using normalization and standardization to ensure that feature values are in a reasonable range to obtain time series features and non-time series features, then carrying out data prediction on the time series features through a preset long-short-term memory network to enable the long-term memory network to capture the risk modes of the product and pay attention to key risk points in the risk modes, and finally, carrying out feature enhancement on the hidden state based on a preset attention mechanism, thereby obtaining comprehensive information of important moments in a time sequence, calculating risk scores of the time sequence features according to the comprehensive information and non-time sequence features, and obtaining a risk assessment result, thereby realizing comprehensive risk assessment by combining multidimensional information, improving the accuracy of risk assessment and realizing comprehensive analysis on historical data.

Specifically, the specific process of the embodiment of the application for predicting the data of the time series characteristics through the preset long-period and short-period memory network is as follows:

=;

;

Wherein, The weight matrix of the forgetting gate is used for controlling the forgetting proportion of the state at the previous moment.For the weight matrix of the input gate, the degree to which new input data enters the cell state is determined.Is a fusion matrix of input data and previous states for updating the cell state.And state information representing the forget gate and controlling the previous time step.To input the gate, the inflow of new information is controlled.And the control signal of the output gate is used for determining the weight output at the current moment.Long-term dependencies in the time series are recorded for the cell status.And outputting the hidden state at the current moment.Is a sigmoid function for compressing values between 0 and 1.And (3) inputting the current time step.Is a bias term for forgetting gates.Is a bias term of the input gate.Is a bias term for the state of the cell.Is the input of the current time step.

Referring to fig. 5, in some embodiments, step S403 may further include, but is not limited to, steps S501 to S505.

Step S501, for each node in the network map, a feature vector is assigned to the node.

Step S502, selecting a test node from the network map, and determining an initial feature vector of the test node and a neighbor node set around the test node.

Step S503, performing linear transformation on the neighbor nodes in the neighbor node set through the graph neural network to obtain a first parameter, and performing linear transformation on the test node to obtain a second parameter.

Step S504, a neighbor feature vector set corresponding to the neighbor node set is determined, a first feature representation is determined according to the initial feature vector, and a second feature representation is determined according to the neighbor feature vector set.

Step S505, inputting the first parameter, the second parameter, the first feature representation and the second feature representation into the graph neural network, so that the graph neural network performs hierarchical update on the test node based on the first parameter, the second parameter, the first feature representation and the second feature representation to obtain the feature representation.

In steps S501 to S505 of some embodiments, in the process of feature fusion of a network map through a preset graph neural network, for each node in the network map, a feature vector is allocated to a node, and the feature vector is used as an initial feature vector, where the feature vector may be an attribute of the node, for example, a brand, a model, etc. of an unmanned aerial vehicle, then a test node is randomly selected in the network map, and an initial feature vector of the test node and a set of neighboring nodes around the test node are determined, so as to implement message transfer between the test node and the neighboring nodes, then linear transformation is performed on the neighboring nodes in the set of neighboring nodes through the graph neural network to obtain a first parameter, and linear transformation is performed on the test node, and finally, inputting the first parameter, the second parameter, the first feature representation and the second feature representation into the graph neural network so that the graph neural network performs hierarchical updating on the test node based on the first parameter, the second parameter, the first feature representation and the second feature representation, and the model convergence speed can be improved while the balance of feature fusion is ensured, and the feature representation of the information of the aggregated neighbor nodes and the self information is obtained, thereby extracting the features in the multidimensional data and improving the accuracy of risk assessment.

Specifically, the level update of the test node in the embodiment of the application is represented as follows:

;

Wherein, Represent the firstTest node in layerIs characterized by; representing test nodes Is a neighbor node set; The standardized coefficients are expressed and used for preventing the node characteristic values from exploding or disappearing, ensuring the balance of characteristic fusion and improving the convergence rate of the model; representing a first feature representation; representing a second feature representation; representing a weight matrix; representing a nonlinear activation function; Representation and test node Adjacent neighbor nodes.

Referring to fig. 6, in some embodiments, step S406 may further include, but is not limited to, steps S601 to S605.

In step S601, an attention score of the hidden state is calculated based on a preset attention mechanism.

Step S602, determining the attention weight of the current time step according to the attention score.

Step S603, weighting and summing the contexts of the time sequence features based on the attention weight and the hidden state to obtain comprehensive information.

And step S604, splicing the comprehensive information and the non-time sequence features to obtain splicing features.

And step S605, risk scoring is carried out on the spliced features, and a risk assessment result is obtained.

In steps S601 to S605 of some embodiments, in the process of performing feature enhancement on a hidden state based on a preset attention mechanism to obtain comprehensive information, calculating a risk score of a time sequence feature according to the comprehensive information and a non-time sequence feature to obtain a risk evaluation result, the embodiment of the application calculates an attention score of the hidden state based on the preset attention mechanism, so that the importance of the current time can be measured by the attention score, and then determines an attention weight of the current time step according to the attention score, so that the relative importance of the current time step can be measured, and then performs weighted summation on the context of the time sequence feature based on the attention weight and the hidden state, so as to obtain comprehensive information of important time in a time sequence, and then performs splicing on the comprehensive information and the non-time sequence feature to obtain a spliced feature, and then performs risk scoring on the spliced feature to obtain a risk evaluation result, so that comprehensive risk evaluation can be realized by combining multi-dimensional information, and the accuracy of risk evaluation is improved, and comprehensive analysis on historical data is realized.

Specifically, the calculation process of weighted summation of the context of time series features based on the attention weight and hidden state is as follows:

;

Wherein, Is a weight matrix for calculating the attention score.Is a bias term for the attention mechanism.Is a time stepFor measuring the importance of the current moment.Time stepRepresents the relative importance of that moment.The context vector after weighted summation represents the comprehensive information of important moments in the time sequence.

Specifically, the calculation process of risk scoring for the splicing features in the embodiment of the application is as follows:

;

Wherein, Representing a time sequence context vectorNon-time series characteristicsIs a splice feature of (2).Representing a weight matrix for connecting the composite featuresTo risk scoring。Representing a rank bias vector for use in conjunction withThe linear outputs of the models are adjusted together.Representing an activation function, such as a sigmoid function.

Referring to fig. 7, in some embodiments, step S106 may further include, but is not limited to, steps S701 to S702.

Step S701, determining the basic premium applied by the user according to the user association data.

And step S702, pricing analysis is carried out according to the basic premium, the risk assessment result and the comprehensive discount parameter, and the target insurance pricing is obtained.

In steps S701 to S702 of some embodiments, in the process of performing pricing analysis on the risk assessment result and the comprehensive discount parameter through the preset pricing model, the embodiment of the application determines the basic premium applied by the user according to the user association data, and performs pricing analysis according to the basic premium, the risk assessment result and the comprehensive discount parameter to obtain the target insurance pricing, so that the insurance rate of the user can be dynamically adjusted by combining the multidimensional data, and more personalized pricing is realized.

Specifically, the embodiment of the application adopts a Bayesian optimization technology to dynamically adjust the insurance rate of the client, and the concrete process is as follows:

;

Wherein, Time of dayI.e., target insurance pricing.Is the basic premium.Is a risk adjustment coefficient.Is a comprehensive discount parameter.

;

Wherein, Is a mean function of the predicted distribution.Is a covariance function used to measure the similarity between data points.

Referring to fig. 8, the embodiment of the application further provides an insurance pricing device based on data fusion, where the device includes:

a data acquisition module 801, configured to respond to a pricing instruction of a user, and acquire product basic information corresponding to the pricing instruction through a pricing system;

the data determining module 802 is configured to receive identity information sent by a user through the pricing system, and determine user association data and a credit platform bound with the pricing system according to the identity information, where the user association data is used to characterize a purchase record and claim settlement information of the user;

the credit score obtaining module 803 is configured to obtain a credit score corresponding to the user through the credit platform in response to an authorization instruction of the user;

A discount calculation module 804 for determining a comprehensive discount parameter according to the credit score and the user-associated data;

The risk assessment module 805 is configured to perform risk assessment on the product basic information, the credit score, and the user-related data, to obtain a risk assessment result;

and the pricing analysis module 806 is configured to perform pricing analysis on the risk assessment result and the comprehensive discount parameter through a preset pricing model, so as to obtain the target insurance pricing.

The specific implementation of the insurance pricing device based on data fusion is basically the same as the specific embodiment of the insurance pricing method based on data fusion, and will not be described herein.

The embodiment of the application also provides electronic equipment, which comprises a memory, a processor, a program stored on the memory and capable of running on the processor and a data bus for realizing connection communication between the processor and the memory, wherein the program is executed by the processor to realize the insurance pricing method based on data fusion. The electronic equipment can be any intelligent terminal including a tablet personal computer, a vehicle-mounted computer and the like.

Referring to fig. 9, fig. 9 illustrates a hardware structure of an electronic device according to another embodiment, the electronic device includes:

The processor 901 may be implemented by a general purpose CPU (Central Processing Unit ), a microprocessor, an application specific integrated circuit (Application SpecificIntegrated Circuit, ASIC), or one or more integrated circuits, etc. for executing related programs to implement the technical solution provided by the embodiments of the present application;

The Memory 902 may be implemented in the form of a Read Only Memory (ROM), a static storage device, a dynamic storage device, or a random access Memory (Random Access Memory, RAM). Memory 902 may store an operating system and other application programs, and when the technical solutions provided in the embodiments of the present disclosure are implemented by software or firmware, relevant program codes are stored in memory 902, and the processor 901 invokes the insurance pricing method based on data fusion to perform the embodiments of the present disclosure;

an input/output interface 903 for inputting and outputting information;

The communication interface 904 is configured to implement communication interaction between the device and other devices, and may implement communication in a wired manner (e.g. USB, network cable, etc.), or may implement communication in a wireless manner (e.g. mobile network, WIFI, bluetooth, etc.);

a bus 905 that transfers information between the various components of the device (e.g., the processor 901, the memory 902, the input/output interface 903, and the communication interface 904);

Wherein the processor 901, the memory 902, the input/output interface 903 and the communication interface 904 are communicatively coupled to each other within the device via a bus 905.

The embodiment of the application also provides a computer readable storage medium, wherein the computer readable storage medium stores a computer program which realizes the insurance pricing method based on data fusion when being executed by a processor.

The memory, as a non-transitory computer readable storage medium, may be used to store non-transitory software programs as well as non-transitory computer executable programs. In addition, the memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one magnetic disk storage device, flash memory device, or other non-transitory solid state storage device. In some embodiments, the memory optionally includes memory remotely located relative to the processor, the remote memory being connectable to the processor through a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.

According to the insurance pricing method, the device, the electronic equipment and the storage medium based on data fusion, provided by the embodiment of the application, the pricing instruction of the user is responded, the product basic information corresponding to the pricing instruction is obtained through the pricing system, so that the basic information of an insurance product can be obtained, the basic information can be conveniently used as a reference for risk assessment later, the identity information sent by the user is received through the pricing system, the user associated data and the credit platform bound with the pricing system are determined according to the identity information, the purchasing record of other products purchased by the user and the current claim settlement record can be obtained, and therefore whether the user has the claim settlement and the specific information of the claim settlement can be judged. Responding to an authorization instruction of a user, acquiring a credit score corresponding to the user through a credit platform, reflecting the credit condition of the user through the credit score, improving the accuracy of risk management, reducing risk loss, determining comprehensive discount parameters according to the credit score and user association data, reasonably setting discount parameters according to the credit condition, historical claim settlement condition and insurance purchase condition of the user, carrying out risk assessment on product basic information, the credit score and user association data, fully reflecting the risk level of the user through analysis of multidimensional data, obtaining a risk assessment result, realizing comprehensive assessment on insurance risk, carrying out pricing analysis on the risk assessment result and the comprehensive discount parameters through a preset pricing model, flexibly adjusting premium in combination with the insurance condition and actual risk of the user, and obtaining target insurance pricing, thereby improving more personalized premium benefits for the user. According to the embodiment of the application, through multidimensional data collection, fusion and analysis, the insurance rate of the user is dynamically adjusted, and through docking the third-party credit platform and integrating data such as purchase records of other insurance products, more personalized premium offers can be provided for the user.

The embodiments described in the embodiments of the present application are for more clearly describing the technical solutions of the embodiments of the present application, and do not constitute a limitation on the technical solutions provided by the embodiments of the present application, and those skilled in the art can know that, with the evolution of technology and the appearance of new application scenarios, the technical solutions provided by the embodiments of the present application are equally applicable to similar technical problems.

It will be appreciated by those skilled in the art that the solutions shown in fig. 1-9 are not limiting on the embodiments of the application and may include more or fewer steps than shown, or certain steps may be combined, or different steps.

The above described apparatus embodiments are merely illustrative, wherein the units illustrated as separate components may or may not be physically separate, i.e. may be located in one place, or may be distributed over a plurality of network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

Those of ordinary skill in the art will appreciate that all or some of the steps of the methods, systems, functional modules/units in the devices disclosed above may be implemented as software, firmware, hardware, and suitable combinations thereof.

The terms "first," "second," "third," "fourth," and the like in the description of the application and in the above figures, if any, are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments of the application described herein may be implemented in sequences other than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.

It should be understood that in the present application, "at least one (item)" means one or more, and "a plurality" means two or more. "and/or" is used to describe an association relationship of an associated object, and indicates that three relationships may exist, for example, "a and/or B" may indicate that only a exists, only B exists, and three cases of a and B exist simultaneously, where a and B may be singular or plural. The character "/" generally indicates that the context-dependent object is an "or" relationship. "at least one of" or the like means any combination of these items, including any combination of single item(s) or plural items(s). For example, at least one of a, b or c may represent a, b, c, "a and b", "a and c", "b and c", or "a and b and c", wherein a, b, c may be single or plural.

In the several embodiments provided by the present application, it should be understood that the disclosed apparatus and method may be implemented in other manners. For example, the above-described apparatus embodiments are merely illustrative, and for example, the above-described division of units is merely a logical function division, and there may be another division manner in actual implementation, for example, a plurality of units or components may be combined or may be integrated into another system, or some features may be omitted, or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be an indirect coupling or communication connection via some interfaces, devices or units, which may be in electrical, mechanical or other form.

The units described above as separate components may or may not be physically separate, and components shown as units may or may not be physical units, may be located in one place, or may be distributed over a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

In addition, each functional unit in the embodiments of the present application may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units may be implemented in hardware or in software functional units.

The integrated units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present application may be embodied in essence or a part contributing to the prior art or all or part of the technical solution in the form of a software product stored in a storage medium, including multiple instructions to cause a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method of the various embodiments of the present application. The storage medium includes various media capable of storing programs, such as a U disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (Random Access Memory RAM), a magnetic disk, or an optical disk.

The preferred embodiments of the present application have been described above with reference to the accompanying drawings, and are not thereby limiting the scope of the claims of the embodiments of the present application. Any modifications, equivalent substitutions and improvements made by those skilled in the art without departing from the scope and spirit of the embodiments of the present application shall fall within the scope of the claims of the embodiments of the present application.

Claims

1. A method of insurance pricing based on data fusion, the method comprising:

2. The data fusion-based insurance pricing method according to claim 1, wherein said determining a composite discount parameter according to said credit score and said user-associated data comprises:

3. The data fusion-based insurance pricing method of claim 2, wherein said determining a composite discount parameter from said credit scoring discount and said relevance discount comprises:

4. The data fusion-based insurance pricing method according to claim 1, wherein performing risk assessment on the product base information, the credit score and the user-associated data to obtain risk assessment results comprises:

5. The data fusion-based insurance pricing method according to claim 4, wherein the feature fusion is performed on the network map through a preset graph neural network to obtain a feature representation, and the method comprises:

For each node in the network map, assigning a feature vector to the node;

6. The data fusion-based insurance pricing method according to claim 4, wherein the feature enhancement is performed on the hidden state based on a preset attention mechanism to obtain comprehensive information, and the risk score of the time series feature is calculated according to the comprehensive information and the non-time series feature to obtain a risk assessment result, including:

7. The data fusion-based insurance pricing method according to claim 1, wherein the pricing analysis is performed on the risk assessment result and the comprehensive discount parameter through a preset pricing model to obtain a target insurance pricing, and the method comprises:

8. An insurance pricing device based on data fusion, the device comprising:

9. An electronic device comprising a memory storing a computer program and a processor that when executing the computer program implements the data fusion-based insurance pricing method according to any of claims 1 to 7.

10. A computer readable storage medium storing a computer program, wherein the computer program when executed by a processor implements the data fusion based insurance pricing method of any of claims 1 to 7.