CN116843483A

CN116843483A - Vehicle insurance claim settlement method, device, computer equipment and storage medium

Info

Publication number: CN116843483A
Application number: CN202311036736.XA
Authority: CN
Inventors: 彭杉
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: 2023-08-16
Filing date: 2023-08-16
Publication date: 2023-10-03

Abstract

The application discloses a method, a device, computer equipment and a storage medium for vehicle insurance claim settlement, and relates to the technical field of artificial intelligence and the field of financial insurance. According to the method, the related accident data of the current vehicle accident are obtained, wherein the related accident data comprise accident vehicle information and accident description information, the related accident data are imported into a pre-trained vehicle damage prediction model to obtain a vehicle damage prediction result, the insurance policy information of the current accident vehicle is queried based on the accident vehicle information, the related accident data, the vehicle damage prediction result and the insurance policy information of the current accident vehicle are imported into a pre-trained vehicle insurance claim recommendation model, and the vehicle insurance claim result of the current vehicle accident is output. In addition, the application also relates to a blockchain technology, and associated accident data can be stored in the blockchain. The application can reduce the influence of subjective factors in the vehicle insurance claim settlement, help claim settlement staff make more accurate and consistent decisions in the vehicle insurance claim settlement process, and improve the claim settlement efficiency and accuracy.

Description

Vehicle insurance claim settlement method, device, computer equipment and storage medium

Technical Field

The application relates to the technical field of artificial intelligence and the field of financial insurance, in particular to a method, a device, computer equipment and a storage medium for vehicle insurance claim settlement.

Background

The automobile insurance claim settlement refers to a series of processes such as the automobile owner submits claim settlement application, the insurance company performs damage assessment and auditing, and final claim payment, and the automobile insurance claim settlement is generally realized by manually tracking the whole process of automobile insurance claim settlement.

The traditional manual tracking vehicle insurance claim settlement process has the problem of time delay, and the waiting time in the claim settlement process is long due to the dependence on manual processing and communication, so that a customer needs to be happy to wait, and the time of the whole claim settlement process is prolonged. Secondly, the manual processing of the vehicle insurance claim settlement process is easy to cause human errors and omission, and the manual processing may be inaccurate or miss some important information due to factors such as fatigue, emotion and the like, so that the accuracy and efficiency of claim settlement are affected. In addition, the manual tracking of the vehicle insurance claim settlement process is easily influenced by individual differences and subjective factors, and lacks consistency, and different claim settlement processing personnel can process according to own judgment and preference, so that inconsistency of processing results is caused. Finally, the traditional manual tracking vehicle insurance claim settlement process is difficult to perform comprehensive data analysis and optimization, and because the information collection and arrangement rely on manual operation, the data collection and analysis work is complicated, so that an insurance company is difficult to acquire accurate and timely data insight, and the optimization of the process and the improvement of efficiency are limited.

Disclosure of Invention

The embodiment of the application aims to provide a vehicle insurance claim settlement method, a vehicle insurance claim settlement device, computer equipment and a storage medium, which are used for solving the problems of influence of individual differences and subjective factors of received claim processing personnel, low time delay, low accuracy, low claim settlement efficiency, complex data processing process and the like in the existing manual tracking vehicle insurance claim settlement process.

In order to solve the technical problems, the embodiment of the application provides a vehicle insurance claim settlement method, which adopts the following technical scheme:

a method of claim settlement for car insurance comprising:

receiving a car insurance claim settlement instruction, and acquiring associated accident data of a current car accident, wherein the associated accident data comprises accident car information and accident description information;

importing the related accident data into a pre-trained vehicle loss prediction model to obtain a vehicle loss prediction result;

inquiring policy information of a current accident vehicle based on the accident vehicle information;

and importing the associated accident data, the train loss prediction result and the insurance policy information of the current accident train into a pre-trained train insurance claim recommendation model, and outputting the train insurance claim result of the current train accident.

Further, before the relevant accident data is imported into the pre-trained vehicle loss prediction model to obtain the vehicle loss prediction result, the method further comprises the following steps:

Acquiring historical accident data and preprocessing the historical accident data, wherein the preprocessing comprises abnormal value removal processing package, missing data filling processing and data normalization processing;

performing data division on the preprocessed historical accident data according to a preset data division rule to obtain a training set and a testing set;

carrying out running loss prediction on a preset initial prediction model through the training set to obtain an initial prediction result, and iterating the initial prediction model based on the initial prediction result to obtain a trained running loss prediction model;

and testing the vehicle loss prediction model through the test set, and outputting the vehicle loss prediction model which passes the test.

Further, the predicting the running loss of the preset initial prediction model through the training set to obtain an initial prediction result, and iterating the initial prediction model based on the initial prediction result to obtain a trained running loss prediction model, which specifically includes:

extracting features of the historical accident data in the training set to obtain historical accident features;

calculating information gain and mutual information between the historical accident characteristics and a preset target variable, wherein the target variable is a variable which needs to be predicted or inferred in the vehicle insurance claim;

Weighting the historical accident feature based on the information gain and the mutual information;

importing the weighted historical accident characteristics into the initial prediction model to predict the traffic loss, so as to obtain an initial prediction result;

and iterating the initial prediction model based on the initial prediction result and a preset back propagation algorithm until the model is fitted to obtain a trained vehicle loss prediction model.

Further, the calculating the information gain and mutual information between the historical accident feature and a preset target variable specifically includes:

calculating entropy and conditional entropy between the historical accident characteristics and preset target variables;

calculating information gain between the historical accident feature and a preset target variable based on entropy and conditional entropy between the historical accident feature and the preset target variable;

calculating joint probability and marginal probability between the historical accident characteristics and a preset target variable;

and calculating mutual information between the historical accident feature and a preset target variable based on the joint probability and the marginal probability between the historical accident feature and the preset target variable.

Further, the iteration is performed on the initial prediction model based on the initial prediction result and a preset back propagation algorithm until the model is fitted, so as to obtain a trained vehicle loss prediction model, which specifically includes:

Calculating errors of the initial prediction result and a preset standard result based on the loss function of the initial prediction model to obtain a prediction error;

the prediction error is reversely propagated in the initial prediction model, and the magnitude of the prediction error and a preset error threshold value are judged;

and when the prediction error is greater than a preset error threshold, continuously adjusting model parameters of the initial prediction model until the prediction error is less than or equal to the preset error threshold, so as to obtain a trained vehicle loss prediction model.

Further, the car insurance claim recommendation model is a clustering model, and the relevant accident data, the car loss prediction result and the policy information of the current accident car are imported into a pre-trained car insurance claim recommendation model to output the car insurance claim result of the current accident car, and the car insurance claim recommendation model specifically comprises the following steps:

extracting characteristics of the related accident data to obtain current accident characteristics, extracting characteristics of the vehicle loss prediction result to obtain accident vehicle loss characteristics, and extracting characteristics of policy information of the current accident vehicle to obtain vehicle policy characteristics;

constructing a current vehicle accident association feature based on the current accident feature, the accident loss feature and the vehicle policy feature;

Acquiring a matched cluster label from the car insurance claim recommendation model based on the current car accident association characteristics;

clustering the current accident feature, the accident vehicle loss feature and the vehicle policy feature based on the cluster label to obtain a feature clustering result;

and identifying a car insurance claim settlement scheme matched with the current car accident based on the characteristic clustering result, and outputting the car insurance claim settlement scheme.

Further, the constructing the current vehicle accident related feature based on the current accident feature, the accident loss feature and the vehicle policy feature specifically includes:

vectorizing the current accident feature to obtain an accident feature vector, vectorizing the accident vehicle loss feature to obtain a vehicle loss feature vector, vectorizing the vehicle policy feature to obtain a policy feature vector;

mapping the accident feature vector, the train loss feature vector and the policy feature vector to a target feature space to obtain a feature mapping result;

and acquiring the current vehicle accident association features based on the feature mapping result.

In order to solve the technical problems, the embodiment of the application also provides a vehicle insurance claim settlement device, which adopts the following technical scheme:

A vehicle insurance claim device, comprising:

the system comprises a data acquisition module, a vehicle accident analysis module and a vehicle accident analysis module, wherein the data acquisition module is used for receiving a vehicle insurance claim instruction and acquiring associated accident data of a current vehicle accident, and the associated accident data comprises accident vehicle information and accident description information;

the vehicle loss prediction module is used for importing the related accident data into a pre-trained vehicle loss prediction model to obtain a vehicle loss prediction result;

the policy inquiry module is used for inquiring the policy information of the current accident vehicle based on the accident vehicle information;

and the car insurance claim settlement module is used for importing the associated accident data, the car damage prediction result and the insurance policy information of the current accident car into a pre-trained car insurance claim settlement recommendation model and outputting the car insurance claim settlement result of the current car accident.

In order to solve the above technical problems, the embodiment of the present application further provides a computer device, which adopts the following technical schemes:

a computer device comprising a memory having stored therein computer readable instructions which when executed by a processor implement the steps of the vehicle insurance claim method of any of the above claims.

In order to solve the above technical problems, an embodiment of the present application further provides a computer readable storage medium, which adopts the following technical schemes:

A computer readable storage medium having stored thereon computer readable instructions which when executed by a processor implement the steps of the vehicle insurance claim method of any of the above claims.

Compared with the prior art, the embodiment of the application has the following main beneficial effects:

the application discloses a method, a device, computer equipment and a storage medium for vehicle insurance claim settlement, and relates to the technical field of artificial intelligence and the field of financial insurance. The method comprises the steps of obtaining associated accident data of a current vehicle accident by receiving a vehicle insurance claim settlement instruction, wherein the associated accident data comprises accident vehicle information and accident description information, importing the associated accident data into a pre-trained vehicle damage prediction model to obtain a vehicle damage prediction result, inquiring the insurance policy information of the current accident vehicle based on the accident vehicle information, importing the associated accident data, the vehicle damage prediction result and the insurance policy information of the current accident vehicle into a pre-trained vehicle insurance claim settlement recommendation model, and outputting the vehicle insurance claim settlement result of the current vehicle accident. According to the application, the vehicle loss degree is predicted by establishing the vehicle loss prediction model, and intelligent recommendation is performed by matching similar cases or historical claim settlement data based on the vehicle risk claim recommendation model, so that targeted advice and decision support can be provided for the vehicle risk claim settlement process, the influence of subjective factors in the vehicle risk claim settlement is reduced, more accurate and consistent decisions are made by claim settlement processing personnel in the vehicle risk claim settlement process, and the claim settlement efficiency and accuracy are improved.

Drawings

In order to more clearly illustrate the solution of the present application, a brief description will be given below of the drawings required for the description of the embodiments of the present application, it being apparent that the drawings in the following description are some embodiments of the present application, and that other drawings may be obtained from these drawings without the exercise of inventive effort for a person of ordinary skill in the art.

FIG. 1 illustrates an exemplary system architecture diagram in which the present application may be applied;

FIG. 2 illustrates a flow chart of one embodiment of a vehicle insurance claim method in accordance with the present application;

FIG. 3 illustrates a schematic diagram of one embodiment of a vehicle insurance claim device in accordance with the present application;

fig. 4 shows a schematic structural diagram of an embodiment of a computer device according to the application.

Detailed Description

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 in the description of the applications herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application; the terms "comprising" and "having" and any variations thereof in the description of the application and the claims and the description of the drawings above are intended to cover a non-exclusive inclusion. The terms first, second and the like in the description and in the claims or in the above-described figures, are used for distinguishing between different objects and not necessarily for describing a sequential or chronological order.

Reference herein to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment may be included in at least one embodiment of the application. The appearances of such phrases in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. Those of skill in the art will explicitly and implicitly appreciate that the embodiments described herein may be combined with other embodiments.

In order to make the person skilled in the art better understand the solution of the present application, the technical solution of the embodiment of the present application will be clearly and completely described below with reference to the accompanying drawings.

As shown in fig. 1, a system architecture 100 may include terminal devices 101, 102, 103, a network 104, and a server 105. The network 104 is used as a medium to provide communication links between the terminal devices 101, 102, 103 and the server 105. The network 104 may include various connection types, such as wired, wireless communication links, or fiber optic cables, among others.

The user may interact with the server 105 via the network 104 using the terminal devices 101, 102, 103 to receive or send messages or the like. Various communication client applications, such as a web browser application, a shopping class application, a search class application, an instant messaging tool, a mailbox client, social platform software, etc., may be installed on the terminal devices 101, 102, 103.

The terminal devices 101, 102, 103 may be various electronic devices having a display screen and supporting web browsing, including but not limited to smartphones, tablet computers, electronic book readers, MP3 players (Moving Picture Experts Group Audio Layer III, dynamic video expert compression standard audio plane 3), MP4 (Moving Picture Experts Group Audio Layer IV, dynamic video expert compression standard audio plane 4) players, laptop and desktop computers, and the like.

The server 105 may be a server that provides various services, such as a background server that provides support for pages displayed on the terminal devices 101, 102, 103, and may be a stand-alone server, or may be a cloud server that provides cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (Content Delivery Network, CDN), and basic cloud computing services such as big data and artificial intelligence platforms.

It should be noted that, the method for vehicle insurance claim settlement provided by the embodiment of the application is generally executed by a server, and accordingly, the device for vehicle insurance claim settlement is generally disposed in the server.

It should be understood that the number of terminal devices, networks and servers in fig. 1 is merely illustrative. There may be any number of terminal devices, networks, and servers, as desired for implementation.

With continued reference to FIG. 2, a flow chart of one embodiment of a vehicle insurance claim method in accordance with the present application is shown. The embodiment of the application can acquire and process the related data based on the artificial intelligence technology. Among these, 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 extend 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 intelligence software technology mainly comprises a computer vision technology, a robot technology, a biological recognition technology, a voice processing technology, a natural language processing technology, machine learning/deep learning and other directions.

The method aims at solving the problems of low accuracy, low claim settlement efficiency, complex data processing process and the like of the existing manual tracking vehicle insurance claim settlement process, which are caused by the influence of individual differences and subjective factors of received claim processing personnel. The application discloses a car insurance claim settlement method, a device, computer equipment and a storage medium, which relate to the technical field of artificial intelligence and the field of financial insurance.

The vehicle insurance claim settlement method comprises the following steps:

s201, receiving a car insurance claim settlement instruction, and acquiring associated accident data of a current car accident, wherein the associated accident data comprises accident car information and accident description information.

In this embodiment, after receiving the vehicle insurance claim instruction, the server obtains associated accident data of the current vehicle accident, where the associated accident data includes accident vehicle information and accident description information. Vehicle data and accident data of multiple channels and sources of the current accident are acquired through data acquisition, including sources of accident data, vehicle information sources, policy information sources and the like, and the data need to be normalized and standardized for subsequent data analysis and modeling.

S202, the associated accident data are imported into a pre-trained vehicle loss prediction model, and a vehicle loss prediction result is obtained.

In this embodiment, through a large amount of historical data and accident cases, training and optimizing are performed by using a machine learning technology or a deep convolution network, such as a linear regression model, a random forest model, a support vector machine model, a CNN model, a DNN model and the like, so as to obtain a trained vehicle loss prediction model, which can predict the loss condition of an accident vehicle based on vehicle accident data. The vehicle loss prediction model can output a vehicle loss grade and vehicle loss explanation corresponding to the vehicle loss grade, and the vehicle loss grade output by the vehicle loss prediction model can be used for predicting a vehicle insurance claim scheme by combining accident data and insurance policy information of an accident vehicle.

S203, inquiring the policy information of the current accident vehicle based on the accident vehicle information.

In this embodiment, the policy information of the current vehicle accident may be queried through information such as a vehicle brand and a license plate number in the accident vehicle information, and the policy information may include a policy type, a policy amount, an insurance company, etc., and the policy information may be used to determine insurance responsibility, verify compensation range, etc.

S204, the associated accident data, the train loss prediction result and the insurance policy information of the current accident train are imported into a pre-trained train insurance claim recommendation model, and the insurance claim result of the current train accident is output.

In this embodiment, the car insurance claim recommendation model is a pre-trained model, and can recommend based on input data to help decision-making personnel make an optimal decision of car insurance claim, and the output result can include a recommended claim amount, a claim settlement mode, a claim payment proportion and the like. The vehicle insurance claim recommendation model can be obtained by training and optimizing based on a clustering model, a collaborative filtering model and a deep learning model (such as a cyclic neural network and a transducer). And importing the associated accident data, the vehicle loss prediction result and the insurance policy information of the current accident vehicle into a vehicle insurance claim recommendation model to obtain the vehicle insurance claim result of the current vehicle accident, wherein the vehicle insurance claim result can help the claim processor to make more accurate and consistent decisions in the vehicle insurance claim process.

In the embodiment, the vehicle loss degree is predicted by establishing the vehicle loss prediction model, and intelligent recommendation is performed by matching similar cases or historical claim settlement data based on the vehicle risk claim recommendation model, so that targeted suggestion and decision support can be provided for the vehicle risk claim settlement process, the influence of subjective factors in the vehicle risk claim settlement is reduced, more accurate and consistent decisions are made by claim settlement processing personnel in the vehicle risk claim settlement process, and the claim settlement efficiency and accuracy are improved.

Further, before the associated accident data is imported into the pre-trained vehicle loss prediction model to obtain the vehicle loss prediction result, the method further comprises the following steps:

carrying out running loss prediction on a preset initial prediction model through a training set to obtain an initial prediction result, and iterating the initial prediction model based on the initial prediction result to obtain a trained running loss prediction model;

And testing the vehicle loss prediction model through the test set, and outputting the vehicle loss prediction model passing the test.

In this embodiment, before the vehicle loss prediction, a vehicle loss prediction model needs to be trained in advance, a large amount of historical data and accident cases need to be used for training the vehicle loss prediction model, and the historical accident data is obtained and preprocessed, where the preprocessing includes outlier removal processing packet, missing data filling processing and data normalization processing. And then, carrying out data division on the preprocessed historical accident data according to a preset data division rule to obtain a training set and a testing set, wherein the data sets are divided according to a ratio of 9:1, for example. Then selecting proper initial prediction models, wherein the common models comprise linear regression, decision trees, random forests, gradient lifting trees and the like, and the models can be used for regression problems to predict the loss degree by inputting characteristics such as vehicle information, accident descriptions and the like. And carrying out running loss prediction on a preset initial prediction model through a training set to obtain an initial prediction result, iterating the initial prediction model based on the initial prediction result to obtain a trained running loss prediction model, testing the running loss prediction model through a testing set, and outputting the running loss prediction model passing the test.

In the embodiment, the vehicle loss prediction model is trained through a large amount of historical data and accident cases, and the vehicle loss prediction model can be directly based on the vehicle accident data to predict the vehicle loss condition.

Further, carrying out running loss prediction on a preset initial prediction model through a training set to obtain an initial prediction result, and iterating the initial prediction model based on the initial prediction result to obtain a trained running loss prediction model, wherein the method specifically comprises the following steps:

extracting features of historical accident data in the training set to obtain historical accident features;

calculating information gain and mutual information between the historical accident characteristics and preset target variables, wherein the target variables are variables which need to be predicted or inferred in the vehicle insurance claims;

weighting the historical accident characteristics based on the information gain and mutual information;

leading the weighted historical accident characteristics into an initial prediction model for predicting the traffic loss to obtain an initial prediction result;

and iterating the initial prediction model based on the initial prediction result and a preset back propagation algorithm until the model is fitted, so as to obtain a trained vehicle loss prediction model.

In this embodiment, in order to improve the accuracy of the prediction model, considering the characteristics of the vehicle insurance claim data, different weights may be given according to the importance of the features in the process of feature engineering. By weighting the features, the model can be guided to pay more attention to the features with higher weights, so that the prediction accuracy of the model is improved, and the assignment of the feature weights can be determined based on field knowledge, correlation of the features, importance evaluation of the features and the like.

In a specific embodiment of the present application, the feature may be weighted based on calculating the information gain between the historical accident feature and the preset target variable, where the information gain measures the degree of contribution of a feature to reducing uncertainty given a known target variable, and the mutual information is a correlation measure between the feature and the target variable.

The target variable refers to a primary variable in the vehicle insurance claim that needs to be predicted or inferred, and generally relates to the amount of claim or the degree of loss, and specifically, the target variable may be one of the following:

claim amount: predicting the claim amount in the vehicle insurance claim, i.e. predicting the compensation amount required to be obtained by the vehicle owner. This variable is typically used to evaluate the degree of loss and the prediction of the amount of compensation for a car insurance claim.

Degree of loss: predicting the degree of loss in the vehicle insurance claim, i.e. predicting the degree of damage to the vehicle or the degree of impact caused by an accident. This variable is typically used to evaluate the severity of the incident and determine the manner in which the claims are processed.

According to the method, the historical accident data in the training set are subjected to feature extraction to obtain the historical accident features, the information gain and the mutual information between the historical accident features and the preset target variables are calculated, wherein the target variables are variables which need to be predicted or inferred in the vehicle insurance claims, the historical accident features are weighted based on the information gain and the mutual information, the weighted historical accident features are imported into an initial prediction model to conduct driving loss prediction to obtain an initial prediction result, and the initial prediction model is iterated based on the initial prediction result and a preset back propagation algorithm until the model is fitted to obtain a trained vehicle loss prediction model.

In the embodiment, the application gives different weights according to the importance of the features, and the model can be guided to pay more attention to the features with higher weights by weighting the features, so that the prediction accuracy of the model is improved.

Further, calculating information gain and mutual information between the historical accident feature and a preset target variable specifically includes:

and calculating mutual information between the historical accident feature and the preset target variable based on the joint probability and the marginal probability between the historical accident feature and the preset target variable.

In the present embodiment, for discrete features and target variables in the historical accident feature, the Entropy (Entropy) and conditional Entropy (Conditional Entropy) can be used to calculate the information gain, where:

the calculation formula of the entropy is as follows: h (Y) = - Σp (Y) ×log2 (P (Y)), where P (Y) represents the probability of the target variable Y;

The calculation formula of the conditional entropy is as follows: h (y|x) =Σp (X) ×h (y|x), where P (X) represents the probability of feature X and H (y|x) represents the entropy of the target variable Y given feature X;

the information gain is calculated by the following formula: IG (X) =h (Y) -H (y|x), representing the degree of uncertainty reduction of feature X to the target variable Y, by calculating the information gain for each feature, the relationship between them and the target variable can be compared.

For continuous features and target variables in the historical accident feature, mutual information can be used for calculating the correlation between the continuous features and the target variables, and the calculation formula of the mutual information is as follows: MI (X, Y) =Σp (X, Y) =log 2 (p (X, Y)/(p (X) ×p (Y))), where p (X, Y) represents the joint probability of the feature X and the target variable Y, and p (X) and p (Y) represent the marginal probabilities of the feature X and the target variable Y, respectively, and mutual information measures the correlation between the feature X and the target variable Y and the amount of shared information.

In the embodiment, the information gain is calculated through the entropy and the conditional entropy between the historical accident feature and the preset target variable, the mutual information is calculated through the joint probability and the marginal probability between the historical accident feature and the preset target variable, the historical accident feature is weighted through the information gain and the mutual information, and the model is guided to pay more attention to the feature with higher weight, so that the prediction precision of the model is improved.

For example, there is a vehicle insurance claim data set that includes the following features: vehicle brand, vehicle model, accident type, weather conditions, model targets for predicting claim amounts. The following results were obtained by calculation:

vehicle brand: the information gain is 0.4, and the mutual information is 0.6;

vehicle model: the information gain is 0.3, and the mutual information is 0.4;

type of accident: the information gain is 0.2, and the mutual information is 0.3;

weather conditions: the information gain is 0.1, and the mutual information is 0.2;

based on these information gain or mutual information values, each feature may be given a respective weight, and assuming that the maximum information gain or mutual information value is set to 1, the corresponding weight may be assigned as follows:

vehicle brand: the weight is 1;

vehicle model: the weight is 0.75;

type of accident: the weight is 0.5;

weather conditions: the weight is 0.25;

in this way, different weights are assigned to different features based on their correlation with the target variable, which weights can be used to weight the processed features during model training and prediction to improve prediction accuracy.

Further, iterating the initial prediction model based on the initial prediction result and a preset back propagation algorithm until the model is fitted to obtain a trained vehicle loss prediction model, which specifically comprises:

Calculating errors of an initial prediction result and a preset standard result based on a loss function of the initial prediction model to obtain a prediction error;

and when the prediction error is greater than a preset error threshold, continuously adjusting model parameters of the initial prediction model until the prediction error is less than or equal to the preset error threshold, and obtaining the trained vehicle loss prediction model.

In this embodiment, the initial prediction model may use a back propagation algorithm to perform model iteration, calculate the error between the initial prediction result and the preset standard result through the loss function of the initial prediction model to obtain a prediction error, back propagate the prediction error in the initial prediction model, determine the magnitude of the prediction error and the preset error threshold in the initial prediction model, and continuously adjust the model parameters of the initial prediction model when the prediction error is greater than the preset error threshold until the prediction error is less than or equal to the preset error threshold, so as to obtain the trained vehicle loss prediction model.

Further, the car insurance claim recommendation model is a clustering model, the associated accident data, the car loss prediction result and the insurance policy information of the current accident car are imported into the car insurance claim recommendation model trained in advance, and the car insurance claim result of the current car accident is output, and the car insurance claim recommendation model specifically comprises the following steps:

Extracting characteristics of associated accident data to obtain current accident characteristics, extracting characteristics of a vehicle loss prediction result to obtain accident vehicle loss characteristics, and extracting characteristics of policy information of a current accident vehicle to obtain vehicle policy characteristics;

acquiring a matched cluster label from a car insurance claim recommendation model based on the current car accident association characteristics;

In this embodiment, a clustering algorithm may be used to match a car insurance claim settlement scheme of a current car accident, firstly, a suitable clustering algorithm, such as K-means, hierarchical clustering, DBSCAN, and the like, is selected, corresponding parameters are set to construct a car insurance claim settlement recommendation model, a recommendation rule or decision rule is established for each cluster label by defining the cluster label, a processing mode, claim amount, and the like for recommending car insurance claim settlement are established, and the recommendation rule may be formulated based on historical data, expert knowledge, business experience, and the like, or may be learned and optimized according to specific conditions by using a machine learning algorithm.

When the vehicle insurance claim settlement scheme is matched with the current vehicle accident, the characteristics of associated accident data are extracted to obtain the current accident characteristics, the characteristics of vehicle damage prediction results are extracted to obtain the accident vehicle damage characteristics, the characteristics of policy information of the current accident vehicle are extracted to obtain the vehicle policy characteristics, the current vehicle accident associated characteristics are constructed based on the current accident characteristics, the accident vehicle damage characteristics and the vehicle policy characteristics, matched cluster labels are obtained from the vehicle insurance claim settlement recommendation model based on the current vehicle accident associated characteristics, corresponding recommendation rules or decision rules are searched according to the labels of the clusters to which the vehicle accident belongs, and the recommendation results of the vehicle insurance claim settlement, such as recommendation processing mode, claim amount range, claim settlement time prediction and the like, are given according to the rules.

In the embodiment, the clustering algorithm is used for matching the car insurance claim settlement scheme of the current car accident, so that targeted suggestion and decision support are provided for the car insurance claim settlement process, the influence of subjective factors in the car insurance claim settlement is reduced, more accurate and consistent decisions are made by claim settlement processing personnel in the car insurance claim settlement process, and the claim settlement efficiency and accuracy are improved.

Further, the method for constructing the current vehicle accident related feature based on the current accident feature, the accident loss feature and the vehicle insurance policy feature specifically comprises the following steps:

mapping the accident feature vector, the vehicle loss feature vector and the policy feature vector to a target feature space to obtain a feature mapping result;

In this embodiment, a plurality of feature vectors may be mapped into the same feature space by means of feature vector mapping, so as to obtain the associated feature. The accident feature vector is obtained by vectorizing the current accident feature, the accident loss feature is vectorized to obtain the loss feature vector, the vehicle policy feature is vectorized to obtain the policy feature vector, the accident feature vector, the loss feature vector and the policy feature vector are mapped to a target feature space to obtain a feature mapping result, and the current vehicle accident related feature is obtained based on the feature mapping result.

In the embodiment, the application realizes feature association through feature vectorization and feature vector mapping, and obtains the current vehicle accident association feature by mapping all feature vectors to the same feature space.

In the above embodiment, the application discloses a vehicle insurance claim settlement method, and relates to the technical field of artificial intelligence and the field of financial insurance. The method comprises the steps of obtaining associated accident data of a current vehicle accident by receiving a vehicle insurance claim settlement instruction, wherein the associated accident data comprises accident vehicle information and accident description information, importing the associated accident data into a pre-trained vehicle damage prediction model to obtain a vehicle damage prediction result, inquiring the insurance policy information of the current accident vehicle based on the accident vehicle information, importing the associated accident data, the vehicle damage prediction result and the insurance policy information of the current accident vehicle into a pre-trained vehicle insurance claim settlement recommendation model, and outputting the vehicle insurance claim settlement result of the current vehicle accident. According to the application, the vehicle loss degree is predicted by establishing the vehicle loss prediction model, and intelligent recommendation is performed by matching similar cases or historical claim settlement data based on the vehicle risk claim recommendation model, so that targeted advice and decision support can be provided for the vehicle risk claim settlement process, the influence of subjective factors in the vehicle risk claim settlement is reduced, more accurate and consistent decisions are made by claim settlement processing personnel in the vehicle risk claim settlement process, and the claim settlement efficiency and accuracy are improved.

In this embodiment, the electronic device (for example, the server shown in fig. 1) on which the vehicle insurance claim method operates may receive the instruction or acquire the data through a wired connection manner or a wireless connection manner. It should be noted that the wireless connection may include, but is not limited to, 3G/4G connections, wiFi connections, bluetooth connections, wiMAX connections, zigbee connections, UWB (ultra wideband) connections, and other now known or later developed wireless connection means.

It is emphasized that, to further ensure the privacy and security of the associated incident data, the associated incident data may also be stored in a blockchain node.

The blockchain is a novel application mode of computer technologies such as distributed data storage, point-to-point transmission, consensus mechanism, encryption algorithm and the like. The Blockchain (Blockchain), which is essentially a decentralised database, is a string of data blocks that are generated by cryptographic means in association, each data block containing a batch of information of network transactions for verifying the validity of the information (anti-counterfeiting) and generating the next block. The blockchain may include a blockchain underlying platform, a platform product services layer, an application services layer, and the like.

Those skilled in the art will appreciate that implementing all or part of the processes of the methods of the embodiments described above may be accomplished by way of computer readable instructions, stored on a computer readable storage medium, which when executed may comprise processes of embodiments of the methods described above. The storage medium may be a nonvolatile storage medium such as a magnetic disk, an optical disk, a Read-Only Memory (ROM), or a random access Memory (Random Access Memory, RAM).

It should be understood that, although the steps in the flowcharts of the figures are shown in order as indicated by the arrows, these steps are not necessarily performed in order as indicated by the arrows. The steps are not strictly limited in order and may be performed in other orders, unless explicitly stated herein. Moreover, at least some of the steps in the flowcharts of the figures may include a plurality of sub-steps or stages that are not necessarily performed at the same time, but may be performed at different times, the order of their execution not necessarily being sequential, but may be performed in turn or alternately with other steps or at least a portion of the other steps or stages.

With further reference to fig. 3, as an implementation of the method shown in fig. 2, the present application provides an embodiment of a vehicle insurance claim device, where the embodiment of the device corresponds to the embodiment of the method shown in fig. 2, and the device may be specifically applied to various electronic devices.

As shown in fig. 3, the vehicle insurance claim device 300 according to the present embodiment includes:

the data acquisition module 301 is configured to receive a vehicle insurance claim instruction, and acquire associated accident data of a current vehicle accident, where the associated accident data includes accident vehicle information and accident description information;

The vehicle loss prediction module 302 is configured to import the associated accident data into a pre-trained vehicle loss prediction model, so as to obtain a vehicle loss prediction result;

a policy inquiry module 303, configured to inquire policy information of a current accident vehicle based on accident vehicle information;

the car insurance claim settlement module 304 is used for importing the associated accident data, the car damage prediction result and the insurance policy information of the current accident car into a pre-trained car insurance claim settlement recommendation model, and outputting the car insurance claim settlement result of the current car accident.

Further, the vehicle insurance claim device 300 further includes:

the preprocessing module is used for acquiring historical accident data and preprocessing the historical accident data, wherein the preprocessing comprises abnormal value removal processing package, missing data filling processing and data normalization processing;

the data set dividing module is used for carrying out data division on the preprocessed historical accident data according to a preset data dividing rule to obtain a training set and a testing set;

the model training module is used for predicting the running loss of a preset initial prediction model through a training set to obtain an initial prediction result, and iterating the initial prediction model based on the initial prediction result to obtain a trained running loss prediction model;

And the model test module is used for testing the vehicle loss prediction model through the test set and outputting the vehicle loss prediction model which passes the test.

Further, the model training module specifically includes:

the first feature extraction unit is used for carrying out feature extraction on the historical accident data in the training set to obtain the historical accident feature;

the information gain and mutual information calculation unit is used for calculating information gain and mutual information between the historical accident characteristics and preset target variables, wherein the target variables are variables which need to be predicted or inferred in the vehicle insurance claims;

the characteristic weighting unit is used for weighting the historical accident characteristics based on the information gain and mutual information;

the vehicle loss prediction unit is used for guiding the weighted historical accident characteristics into the initial prediction model to predict the vehicle loss so as to obtain an initial prediction result;

and the reverse iteration unit is used for iterating the initial prediction model based on the initial prediction result and a preset reverse propagation algorithm until the model is fitted, so as to obtain a trained vehicle loss prediction model.

Further, the information gain and mutual information calculation unit specifically includes:

the entropy calculating subunit is used for calculating entropy and conditional entropy between the historical accident characteristics and a preset target variable;

The information gain calculating subunit is used for calculating the information gain between the historical accident feature and the preset target variable based on the entropy and the conditional entropy between the historical accident feature and the preset target variable;

the probability calculation subunit is used for calculating joint probability and marginal probability between the historical accident characteristics and a preset target variable;

and the mutual information calculation subunit is used for calculating the mutual information between the historical accident feature and the preset target variable based on the joint probability and the marginal probability between the historical accident feature and the preset target variable.

Further, the reverse iteration unit specifically includes:

the error calculation subunit is used for calculating errors of the initial prediction result and a preset standard result based on the loss function of the initial prediction model to obtain a prediction error;

the back propagation subunit is used for back propagating the prediction error in the initial prediction model and judging the magnitude of the prediction error and a preset error threshold value;

and the iteration single updating subunit is used for continuously adjusting the model parameters of the initial prediction model when the prediction error is greater than a preset error threshold value until the prediction error is less than or equal to the preset error threshold value, so as to obtain the trained vehicle loss prediction model.

Further, the car insurance claim recommendation model is a cluster model, and the car insurance claim module 304 specifically includes:

the second feature extraction unit is used for extracting features of associated accident data to obtain current accident features, extracting features of a vehicle loss prediction result to obtain accident vehicle loss features, and extracting features of policy information of a current accident vehicle to obtain vehicle policy features;

the feature combination unit is used for constructing current vehicle accident association features based on the current accident features, the accident vehicle loss features and the vehicle insurance policy features;

the cluster matching unit is used for acquiring matched cluster labels from the car insurance claim recommendation model based on the current car accident association characteristics;

the feature clustering unit is used for clustering the current accident feature, the accident vehicle loss feature and the vehicle policy feature based on the cluster label to obtain feature clustering results;

and the vehicle insurance claim identification unit is used for identifying a vehicle insurance claim settlement scheme matched with the current vehicle accident based on the characteristic clustering result and outputting the vehicle insurance claim settlement scheme.

Further, the feature combining unit specifically includes:

the feature vectorization subunit is used for vectorizing the current accident feature to obtain an accident feature vector, vectorizing the accident vehicle loss feature to obtain a vehicle loss feature vector, vectorizing the vehicle policy feature to obtain a policy feature vector;

The feature mapping subunit is used for mapping the accident feature vector, the train loss feature vector and the policy feature vector to a target feature space to obtain a feature mapping result;

and the joint feature acquisition subunit is used for acquiring the current vehicle accident associated feature based on the feature mapping result.

In the above embodiment, the application discloses a vehicle insurance claim settlement device, and relates to the technical field of artificial intelligence and the field of financial insurance. The method comprises the steps of obtaining associated accident data of a current vehicle accident by receiving a vehicle insurance claim settlement instruction, wherein the associated accident data comprises accident vehicle information and accident description information, importing the associated accident data into a pre-trained vehicle damage prediction model to obtain a vehicle damage prediction result, inquiring the insurance policy information of the current accident vehicle based on the accident vehicle information, importing the associated accident data, the vehicle damage prediction result and the insurance policy information of the current accident vehicle into a pre-trained vehicle insurance claim settlement recommendation model, and outputting the vehicle insurance claim settlement result of the current vehicle accident. According to the application, the vehicle loss degree is predicted by establishing the vehicle loss prediction model, and intelligent recommendation is performed by matching similar cases or historical claim settlement data based on the vehicle risk claim recommendation model, so that targeted advice and decision support can be provided for the vehicle risk claim settlement process, the influence of subjective factors in the vehicle risk claim settlement is reduced, more accurate and consistent decisions are made by claim settlement processing personnel in the vehicle risk claim settlement process, and the claim settlement efficiency and accuracy are improved.

In order to solve the technical problems, the embodiment of the application also provides computer equipment. Referring specifically to fig. 4, fig. 4 is a basic structural block diagram of a computer device according to the present embodiment.

The computer device 4 comprises a memory 41, a processor 42, a network interface 43 communicatively connected to each other via a system bus. It should be noted that only computer device 4 having components 41-43 is shown in the figures, but it should be understood that not all of the illustrated components are required to be implemented and that more or fewer components may be implemented instead. It will be appreciated by those skilled in the art that the computer device herein is a device capable of automatically performing numerical calculations and/or information processing in accordance with predetermined or stored instructions, the hardware of which includes, but is not limited to, microprocessors, application specific integrated circuits (Application Specific Integrated Circuit, ASICs), programmable gate arrays (fields-Programmable Gate Array, FPGAs), digital processors (Digital Signal Processor, DSPs), embedded devices, etc.

The computer equipment can be a desktop computer, a notebook computer, a palm computer, a cloud server and other computing equipment. The computer equipment can perform man-machine interaction with a user through a keyboard, a mouse, a remote controller, a touch pad or voice control equipment and the like.

The memory 41 includes at least one type of readable storage medium including flash memory, hard disk, multimedia card, card memory (e.g., SD or DX memory, etc.), random Access Memory (RAM), static Random Access Memory (SRAM), read Only Memory (ROM), electrically Erasable Programmable Read Only Memory (EEPROM), programmable Read Only Memory (PROM), magnetic memory, magnetic disk, optical disk, etc. In some embodiments, the storage 41 may be an internal storage unit of the computer device 4, such as a hard disk or a memory of the computer device 4. In other embodiments, the memory 41 may also be an external storage device of the computer device 4, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash Card (Flash Card) or the like, which are provided on the computer device 4. Of course, the memory 41 may also comprise both an internal memory unit of the computer device 4 and an external memory device. In this embodiment, the memory 41 is typically used to store an operating system and various application software installed on the computer device 4, such as computer readable instructions of a vehicle insurance claim method. Further, the memory 41 may be used to temporarily store various types of data that have been output or are to be output.

The processor 42 may be a central processing unit (Central Processing Unit, CPU), controller, microcontroller, microprocessor, or other data processing chip in some embodiments. The processor 42 is typically used to control the overall operation of the computer device 4. In this embodiment, the processor 42 is configured to execute computer readable instructions stored in the memory 41 or process data, such as computer readable instructions for executing the vehicle insurance claim method.

The network interface 43 may comprise a wireless network interface or a wired network interface, which network interface 43 is typically used for establishing a communication connection between the computer device 4 and other electronic devices.

In the above embodiments, the application discloses a computer device, and relates to the technical field of artificial intelligence and the field of financial insurance. The method comprises the steps of obtaining associated accident data of a current vehicle accident by receiving a vehicle insurance claim settlement instruction, wherein the associated accident data comprises accident vehicle information and accident description information, importing the associated accident data into a pre-trained vehicle damage prediction model to obtain a vehicle damage prediction result, inquiring the insurance policy information of the current accident vehicle based on the accident vehicle information, importing the associated accident data, the vehicle damage prediction result and the insurance policy information of the current accident vehicle into a pre-trained vehicle insurance claim settlement recommendation model, and outputting the vehicle insurance claim settlement result of the current vehicle accident. According to the application, the vehicle loss degree is predicted by establishing the vehicle loss prediction model, and intelligent recommendation is performed by matching similar cases or historical claim settlement data based on the vehicle risk claim recommendation model, so that targeted advice and decision support can be provided for the vehicle risk claim settlement process, the influence of subjective factors in the vehicle risk claim settlement is reduced, more accurate and consistent decisions are made by claim settlement processing personnel in the vehicle risk claim settlement process, and the claim settlement efficiency and accuracy are improved.

The present application also provides another embodiment, namely, a computer-readable storage medium storing computer-readable instructions executable by at least one processor to cause the at least one processor to perform the steps of the vehicle insurance claim method as described above.

In the above embodiments, the present application discloses a computer readable storage medium, which relates to the technical field of artificial intelligence and the field of financial insurance. The method comprises the steps of obtaining associated accident data of a current vehicle accident by receiving a vehicle insurance claim settlement instruction, wherein the associated accident data comprises accident vehicle information and accident description information, importing the associated accident data into a pre-trained vehicle damage prediction model to obtain a vehicle damage prediction result, inquiring the insurance policy information of the current accident vehicle based on the accident vehicle information, importing the associated accident data, the vehicle damage prediction result and the insurance policy information of the current accident vehicle into a pre-trained vehicle insurance claim settlement recommendation model, and outputting the vehicle insurance claim settlement result of the current vehicle accident. According to the application, the vehicle loss degree is predicted by establishing the vehicle loss prediction model, and intelligent recommendation is performed by matching similar cases or historical claim settlement data based on the vehicle risk claim recommendation model, so that targeted advice and decision support can be provided for the vehicle risk claim settlement process, the influence of subjective factors in the vehicle risk claim settlement is reduced, more accurate and consistent decisions are made by claim settlement processing personnel in the vehicle risk claim settlement process, and the claim settlement efficiency and accuracy are improved.

From the above description of the embodiments, it will be clear to those skilled in the art that the above-described embodiment method may be implemented by means of software plus a necessary general hardware platform, but of course may also be implemented by means of hardware, but in many cases the former is a preferred embodiment. Based on such understanding, the technical solution of the present application may be embodied essentially or in a part contributing to the prior art in the form of a software product stored in a storage medium (e.g. ROM/RAM, magnetic disk, optical disk) comprising instructions for causing a terminal device (which may be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to perform the method according to the embodiments of the present application.

The application is operational with numerous general purpose or special purpose computer system environments or configurations. For example: personal computers, server computers, hand-held or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments that include 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.

It is apparent that the above-described embodiments are only some embodiments of the present application, but not all embodiments, and the preferred embodiments of the present application are shown in the drawings, which do not limit the scope of the patent claims. This application may be embodied in many different forms, but rather, embodiments are provided in order to provide a thorough and complete understanding of the present disclosure. Although the application has been described in detail with reference to the foregoing embodiments, it will be apparent to those skilled in the art that modifications may be made to the embodiments described in the foregoing description, or equivalents may be substituted for elements thereof. All equivalent structures made by the content of the specification and the drawings of the application are directly or indirectly applied to other related technical fields, and are also within the scope of the application.

Claims

1. A method of claim settlement for car insurance comprising:

2. The vehicle insurance claim 1, wherein before said introducing said associated accident data into a pre-trained vehicle loss prediction model to obtain a vehicle loss prediction result, further comprising:

3. The vehicle insurance claim 2, wherein the predicting the vehicle loss through the training set to a preset initial prediction model to obtain an initial prediction result, and iterating the initial prediction model based on the initial prediction result to obtain a trained vehicle loss prediction model, specifically comprising:

4. A method of claim 3, wherein said calculating information gain and mutual information between said historical accident signature and a predetermined target variable comprises:

5. The vehicle insurance claim 3, wherein the iterative process is performed on the initial prediction model based on the initial prediction result and a preset back propagation algorithm until the model is fitted to obtain a trained vehicle loss prediction model, and specifically includes:

6. The vehicle insurance claim method according to any one of claims 1 to 5, wherein the vehicle insurance claim recommendation model is a cluster model, and the step of importing the associated accident data, the vehicle loss prediction result and the policy information of the current accident vehicle into the pre-trained vehicle insurance claim recommendation model and outputting the vehicle insurance claim result of the current vehicle accident comprises:

7. The vehicle insurance claim 6, wherein said constructing a current vehicle accident-related feature based on said current accident feature, said accident-loss feature, and said vehicle policy feature, comprises:

8. A vehicle insurance claim device, comprising:

9. A computer device comprising a memory and a processor, the memory having stored therein computer readable instructions which when executed by the processor implement the steps of the vehicle insurance claim method of any of claims 1 to 7.

10. A computer readable storage medium having stored thereon computer readable instructions which when executed by a processor implement the steps of the vehicle insurance claim method of any of claims 1 to 7.