CN117764742A - Method and device for determining insurance health index - Google Patents
Method and device for determining insurance health index Download PDFInfo
- Publication number
- CN117764742A CN117764742A CN202311553364.8A CN202311553364A CN117764742A CN 117764742 A CN117764742 A CN 117764742A CN 202311553364 A CN202311553364 A CN 202311553364A CN 117764742 A CN117764742 A CN 117764742A
- Authority
- CN
- China
- Prior art keywords
- insurance
- target
- user
- relationship
- determining
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 230000036541 health Effects 0.000 title claims abstract description 81
- 238000000034 method Methods 0.000 title claims abstract description 45
- 238000012549 training Methods 0.000 claims description 31
- 239000008280 blood Substances 0.000 claims description 18
- 210000004369 blood Anatomy 0.000 claims description 18
- 238000003062 neural network model Methods 0.000 claims description 18
- 238000004891 communication Methods 0.000 claims description 17
- 230000002159 abnormal effect Effects 0.000 claims description 16
- 238000013058 risk prediction model Methods 0.000 claims description 13
- 238000010276 construction Methods 0.000 claims description 6
- 238000001914 filtration Methods 0.000 claims description 5
- 238000012545 processing Methods 0.000 abstract description 9
- 201000010099 disease Diseases 0.000 description 8
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 8
- 230000000694 effects Effects 0.000 description 7
- 239000000284 extract Substances 0.000 description 5
- 230000008901 benefit Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 238000010801 machine learning Methods 0.000 description 3
- 238000013473 artificial intelligence Methods 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 238000013500 data storage Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000000513 principal component analysis Methods 0.000 description 2
- 230000004083 survival effect Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000003058 natural language processing Methods 0.000 description 1
- 238000011022 operating instruction Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 239000002689 soil Substances 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Landscapes
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
Abstract
The invention discloses a method and a device for determining an insurance health index, relates to the technical field of data processing, and mainly aims to solve the problem that the accuracy of determining the existing health index is poor. Comprising the following steps: acquiring insurance basic data of different users, and extracting user characteristic information from the insurance basic data, wherein the user characteristic information is used for representing contents belonging to different insurance attributes; constructing a social relationship knowledge graph of the user characteristic information according to a structured network, wherein the structured network is configured based on at least one social relationship; and after the target insurance risk prediction of the target user is completed based on the target insurance base data, carrying out weight configuration on the target insurance risk prediction based on the social relationship knowledge graph, and determining the insurance health index of the target user based on the configured weight value and the target insurance risk prediction result.
Description
Technical Field
The present invention relates to the field of data processing technologies, and in particular, to a method and an apparatus for determining an insurance health index.
Background
With the application of big data in the insurance field, the calculation of insurance underwriting, premium, etc. is usually predicted based on user's insurance base data. In order to ensure the underwriting effect of the insurance product, the health condition of the user is generally determined, for example, the health index of the user is predicted based on the insurance base data, so as to realize the processing effect of the insurance business based on the health index.
At present, the determination of the health index of the existing user usually directly adopts an artificial intelligence algorithm to predict the basic data of the user, but the direct prediction based on the artificial intelligence algorithm can neglect the guiding effect or influence of other users on the health index of the user, so that the accuracy of the determination of the health index is reduced, and the treatment effect of the health index on insurance business is influenced.
Disclosure of Invention
In view of the above, the present invention provides a method and a device for determining an insurance health index, which are mainly aimed at solving the problem of poor accuracy of determining the existing health index.
According to one aspect of the present invention, there is provided a method of determining an insurance health index, comprising:
acquiring insurance basic data of different users, and extracting user characteristic information from the insurance basic data, wherein the user characteristic information is used for representing contents belonging to different insurance attributes;
Constructing a social relationship knowledge graph of the user characteristic information according to a structured network, wherein the structured network is configured based on at least one social relationship;
and after the target insurance risk prediction of the target user is completed based on the target insurance base data, carrying out weight configuration on the target insurance risk prediction based on the social relationship knowledge graph, and determining the insurance health index of the target user based on the configured weight value and the target insurance risk prediction result.
Further, before the social relationship knowledge graph of the user characteristic information is constructed according to the structured network, the method further comprises:
identifying text information used for representing social relations in the insurance base data, wherein the social relations comprise blood relation, friend relation, working relation and certificate relation;
determining a relationship branch and a relationship quantity based on the text information, and configuring the structured network based on the relationship branch and the relationship quantity.
Further, the constructing the social relationship knowledge graph of the user characteristic information according to the structured network includes:
determining that the user characteristic information belongs to a target relationship branch in the blood relationship, the friend relationship, the work relationship and the certificate relationship;
And drawing the user identification of the user in the structured network according to the target relationship branch to obtain a social relationship knowledge graph.
Further, the weight configuration for the target insurance risk prediction based on the social relationship knowledge graph includes:
acquiring social relationship influence risk rules, wherein the social relationship influence risk rules are used for representing judgment methods for risks caused by different insurance businesses and insurance products;
and determining a user relationship branch corresponding to the target user in the social relationship knowledge graph according to the social relationship influence risk rule, and configuring a weight value of the target insurance risk prediction based on the user relationship branch or the branch layer number of the user relationship branch.
Further, before the weight configuration is performed on the target insurance risk prediction based on the social relationship knowledge graph, the method further includes:
acquiring a historical insurance risk training sample, and constructing a neural network model;
when training the neural network model based on the historical insurance risk training sample, determining a model root mean square error and a model average absolute error, and completing training of the neural network model when the model root mean square error and the model average absolute error match a preset error threshold value to obtain the risk prediction model;
And predicting the target insurance basic data of the target user based on the risk prediction model with model training completed, so as to obtain the target insurance prediction result.
Further, before the extracting the user characteristic information from the insurance base data, the method further includes:
extracting data attributes in the insurance base data, and determining abnormal base data in the insurance base data according to the data attributes;
and filling and/or filtering the abnormal basic data according to the data attribute to obtain insurance basic data of the user characteristic information to be extracted.
Further, after determining the insurance health index of the target user based on the configured weight value and the target insurance risk prediction result, the method further includes:
if the insurance health index is greater than the risk threshold of the target insurance service selected by the target user, determining a product transaction object between the target user and the target insurance service so as to execute the target insurance service based on the product transaction object;
and if the insurance health index is smaller than or equal to the risk threshold of the target insurance service selected by the target user, generating insurance service early warning information of the target user.
According to another aspect of the present invention, there is provided a determining apparatus of an insurance health index, including:
the system comprises an acquisition module, a storage module and a storage module, wherein the acquisition module is used for acquiring insurance basic data of different users and extracting user characteristic information from the insurance basic data, wherein the user characteristic information is used for representing contents belonging to different insurance attributes;
the construction module is used for constructing a social relationship knowledge graph of the user characteristic information according to a structural network, and the structural network is configured based on at least one social relationship;
and the determining module is used for carrying out weight configuration on the target insurance risk prediction based on the social relationship knowledge graph after the target insurance risk prediction of the target user is completed based on the target insurance base data, and determining the insurance health index of the target user based on the configured weight value and the target insurance risk prediction result.
Further, the apparatus further comprises:
the identification module is used for identifying text information used for representing social relations in the insurance base data, wherein the social relations comprise blood relation, friend relation, working relation and certificate relation;
and the configuration module is used for determining a relation branch and a relation quantity based on the text information and configuring the structured network based on the relation branch and the relation quantity.
Further, the construction module is specifically configured to determine that the user characteristic information belongs to a target relationship branch in the blood relationship, the friend relationship, the work relationship and the certificate relationship; and drawing the user identification of the user in the structured network according to the target relationship branch to obtain a social relationship knowledge graph.
Further, the determining module is specifically configured to obtain a social relationship influence risk rule, where the social relationship influence risk rule is used to characterize a judging method for risk caused to different insurance services and insurance products; and determining a user relationship branch corresponding to the target user in the social relationship knowledge graph according to the social relationship influence risk rule, and configuring a weight value of the target insurance risk prediction based on the user relationship branch or the branch layer number of the user relationship branch.
Further, the apparatus further comprises: the prediction module is used for predicting the number of the blocks,
the acquisition module is also used for acquiring a historical insurance risk training sample and constructing a neural network model;
the determining module is further configured to determine a model root mean square error and a model average absolute error when the neural network model is trained based on the historical insurance risk training sample, and complete training of the neural network model when the model root mean square error and the model average absolute error match a preset error threshold value, so as to obtain the risk prediction model;
And the prediction module is used for predicting the target insurance basic data of the target user based on the risk prediction model which is trained by the model, and obtaining the target insurance prediction result.
Further, the apparatus further comprises:
the extraction module is used for extracting data attributes in the insurance base data and determining abnormal base data in the insurance base data according to the data attributes; and filling and/or filtering the abnormal basic data according to the data attribute to obtain insurance basic data of the user characteristic information to be extracted.
Further, the apparatus further comprises: the generation module is configured to generate a first set of data,
the determining module is further configured to determine a product transaction object between the target user and the target insurance service if the insurance health index is greater than a risk threshold of the target insurance service selected by the target user, so as to execute the target insurance service based on the product transaction object;
the generation module is configured to generate insurance business early warning information of the target user if the insurance health index is less than or equal to a risk threshold of the target insurance business selected by the target user.
According to still another aspect of the present invention, there is provided a storage medium having stored therein at least one executable instruction for causing a processor to perform operations corresponding to the method of determining an insurance health index as described above.
According to still another aspect of the present invention, there is provided a terminal including: the device comprises a processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface complete communication with each other through the communication bus;
the memory is used for storing at least one executable instruction, and the executable instruction enables the processor to execute the operation corresponding to the method for determining the insurance health index.
By means of the technical scheme, the technical scheme provided by the embodiment of the invention has at least the following advantages:
compared with the prior art, the embodiment of the invention obtains the insurance basic data of different users, and extracts the user characteristic information from the insurance basic data, wherein the user characteristic information is used for representing the contents belonging to different insurance attributes; constructing a social relationship knowledge graph of the user characteristic information according to a structured network, wherein the structured network is configured based on at least one social relationship; after target insurance risk prediction of a target user is completed based on target insurance basic data, weight configuration is carried out on the target insurance risk prediction based on the social relationship knowledge graph, and an insurance health index of the target user is determined based on the configured weight value and a target insurance risk prediction result, so that the accuracy of health index determination is greatly improved based on the relationship among users as an expected basis of the user health index, and the processing effect of the health index-based insurance business is improved.
The foregoing description is only an overview of the present invention, and is intended to be implemented in accordance with the teachings of the present invention in order that the same may be more clearly understood and to make the same and other objects, features and advantages of the present invention more readily apparent.
Drawings
Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The drawings are only for purposes of illustrating the preferred embodiments and are not to be construed as limiting the invention. Also, like reference numerals are used to designate like parts throughout the figures. In the drawings:
FIG. 1 is a flowchart of a method for determining an insurance health index according to an embodiment of the present invention;
FIG. 2 shows a social relationship knowledge graph provided by the embodiment of the invention;
FIG. 3 shows another social relationship knowledge graph provided by an embodiment of the invention;
FIG. 4 is a block diagram showing the constitution of a determining device for insurance health index according to an embodiment of the present invention;
fig. 5 shows a schematic structural diagram of a terminal according to an embodiment of the present invention.
Detailed Description
Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.
The embodiment of the invention provides a method for determining an insurance health index, as shown in fig. 1, the method comprises the following steps:
101. and acquiring insurance basic data of different users, and extracting user characteristic information from the insurance basic data.
In the embodiment of the invention, the current execution end can be a cloud server or a terminal server for executing insurance health index determination so as to acquire insurance base data of different users. The insurance basic data is used for representing data generated or recorded by a user purchasing an insurance product, at this time, the insurance product includes, but is not limited to, personal insurance, property insurance, agricultural insurance, car insurance and the like, the corresponding insurance basic data includes, but is not limited to, historical full data of various insurance policies, after the current execution end obtains the insurance basic data, user characteristic information is extracted from the insurance basic data, at this time, the user characteristic information is used for representing content belonging to different insurance attributes, including personal attribute information (such as data including gender, age, occupation, academic, marital status, annual income level, location area, disease ICD code, whether disability, disability level, personal insurance risk, etc.), family attribute information (such as data of family number, family annual income, family occupation distribution, family highest school, family residence, family health risk, etc.), regional attribute information (such as data of environment, soil, water quality, cultural background, disease morbidity, five-year morbidity, serious disease risk, etc.), and the embodiment of the invention is not particularly limited.
102. And constructing a social relationship knowledge graph of the user characteristic information according to the structured network.
In the embodiment of the invention, the structured network is configured based on at least one social relationship, the current execution end pre-configures the structured network, the social relationship among different users is represented by each relationship branch in the network, and when the social relationship knowledge graph is constructed, the target relationship branch of the knowledge graph constructed by each user characteristic information is determined, so that the user characteristic information is drawn in the structured network to obtain the social relationship knowledge graph. The social relationship knowledge graph records social relationships among different users including, but not limited to, blood relationship, friend relationship, work relationship and certificate relationship, as shown in fig. 2, so that the social relationship after a certain user and other users can be queried based on the social relationship knowledge graph.
103. And after the target insurance risk prediction of the target user is completed based on the target insurance base data, carrying out weight configuration on the target insurance risk prediction based on the social relationship knowledge graph, and determining the insurance health index of the target user based on the configured weight value and the target insurance risk prediction result.
In the embodiment of the invention, in the scene of determining the insurance health data of the target user, the current execution end carries out target insurance risk prediction through the target insurance basic data of the target user, and at the moment, the risk prediction model can be obtained by training a machine learning algorithm in advance for carrying out the insurance risk prediction. In addition, after the target insurance risk prediction is completed, the weight configuration is performed for the performed target insurance risk prediction based on the social relationship knowledge relationship graph, at this time, the weight configuration may be configured based on whether the social relationship has a relationship and whether the target user has a relationship of underwriting or insurance product transaction, for example, when the target user has other users with a relationship of blood, if the other users have an underwriting history, a higher weight value is configured for the target insurance risk prediction of the target user so as to embody that the target user has a higher insurance risk, or the target insurance risk prediction is the prediction of the target disease insurance, the configuration of the weight value may also be performed based on the five-year survival rate of the disease in the region where the user is located, and when the five-year survival rate is higher, the weight is smaller. And after the target insurance risk prediction is completed, calculating an insurance health index of the target user based on the configured weight value and the target risk prediction result, wherein the insurance health index is used for representing the health score of the target user aiming at the non-insurance risk. In different application scenarios, if the insurance risk is a personal insurance risk, the insurance health index is a health score of a disease, and if the insurance risk is a property insurance risk, the insurance health index is a health score of property loss, and the embodiment of the invention is not particularly limited.
In a specific implementation scenario, when determining the insurance health index of the target user based on the configured weight value and the target insurance risk prediction results, each target insurance risk prediction result is a disease risk prediction result for predicting different diseases, the insurance health index S is calculated by combining the configured weight value and the configured weight value, the calculation formula is,s is an insurance health index, X is a target insurance risk prediction result, a is a weight value corresponding to the target insurance risk prediction result, i is an integer variable of 1-n, and n is the number of the target insurance risk prediction result.
In another embodiment of the present invention, for further defining and describing, before the step of constructing the social relationship knowledge graph of the user feature information according to the structured network, the method further includes:
identifying text information used for representing social relations in the insurance base data;
determining a relationship branch and a relationship quantity based on the text information, and configuring the structured network based on the relationship branch and the relationship quantity.
In order to accurately construct a social relationship knowledge graph, weight configuration is carried out by taking the relationship in the graph as insurance risk prediction, the determination accuracy of insurance health indexes is improved, and a current execution end firstly identifies text information in insurance basic data. The text information with social relationship can be identified by a natural language processing technology, at this time, since the insurance base data is extracted from texts in various table forms, when the text identification is performed, the text information can be extracted based on the social relationship identifier marked in the table, for example, the text information can be identified by identifying the word identifier marked with "relatives" in the insurance base data, and the embodiment of the invention is not limited specifically. Specifically, the social relationship includes a blood relationship, a friend relationship, a work relationship and a certificate relationship, the blood relationship includes relationships of father, mother, son, grandson, third generation relatives and the like, the friend relationship includes relationships of classmates or general friends and the like, the work relationship includes relationships of colleagues, work going and the like, the certificate relationship includes relationships of spouse, guardian relationship, doctor-patient and the like which are established based on legal relationships, and the embodiment of the invention is not particularly limited. In addition, the certificate relationship may include, in addition to the already-divorced partner relationship, an already-divorced partner relationship, etc. so as to include a relationship branch of the already-divorced partner relationship when the structured network is constructed, and embodiments of the present invention are not limited in detail.
It should be noted that, after the text information is identified by the current executing end, in order to configure the diversified structured network, a relationship branch and a relationship number are determined based on the text information, where at this time, one relationship branch is used to represent one relationship between two users, and the relationship number user represents the relationship number of the social relationship between multiple users, which may be based on statistics of the social relationship determined by the text information, and the embodiment of the present invention is not limited specifically. Further, the structured network is configured according to the relationship branches and the relationship number, for example, the relationship branches of the user a, the user b, the user c and the user d are blood-source branches and friend branches respectively, and the relationship number is 3, and the structured network is constructed as shown in fig. 3.
In another embodiment of the present invention, for further defining and describing, the step of constructing a social relationship knowledge graph of the user feature information according to a structured network includes:
determining that the user characteristic information belongs to a target relationship branch in the blood relationship, the friend relationship, the work relationship and the certificate relationship;
and drawing the user identification of the user in the structured network according to the target relationship branch to obtain a social relationship knowledge graph.
In order to improve the construction effectiveness of the social relationship knowledge graph, when the current execution end constructs the social relationship knowledge graph, specifically, when the current execution end constructs the social relationship knowledge graph, the current execution end firstly determines that the user characteristic information belongs to target relationship branches in blood relationship, friend relationship, working relationship and certificate relationship. At this time, the target relationship branch is a branch determined according to a relationship identified from the user feature information, so as to obtain at least one target relationship branch.
After determining the target relationship branch, the current execution end draws the user identifier of the user and the structured network according to the determined target relationship branch, and at this time, the structured network is a blank network which is configured and also has different relationships, and the current execution end only draws the user identifier (such as a user name, an identity card number, etc.) for representing the user identity in the blank network according to the corresponding relationship branch.
In another embodiment of the present invention, for further defining and describing, the step of performing weight configuration for the target insurance risk prediction based on the social relationship knowledge graph includes:
Acquiring social relationship influence risk rules;
and determining a user relationship branch corresponding to the target user in the social relationship knowledge graph according to the social relationship influence risk rule, and configuring a weight value of the target insurance risk prediction based on the user relationship branch or the branch layer number of the user relationship branch.
In order to accurately perform weight configuration for target risk prediction, so that weight configuration for different insurance risk predictions is met, a current execution end firstly acquires a social relationship influence risk rule to determine a weight value based on the rule. The social relationship influence risk rule is used for representing a judging method of risk caused by different insurance services and insurance products, for example, a method of risk judgment caused by certain insurance products based on insurance basic data is that a machine learning algorithm predicts the insurance risk of a user, or a method of risk judgment for certain insurance service operation based on insurance basic data predicts the insurance service risk of the user based on regularized logic rules, and the embodiment of the invention is not particularly limited. Further, the current execution end determines a user relationship branch corresponding to the target user in the social relationship knowledge graph according to the social relationship influence risk rule, at this time, the current execution end may perform screening determination based on a preset mapping relationship between rules corresponding to different insurance products and insurance services and different user relationship branches, for example, if the social relationship influence risk rule is a machine learning algorithm for performing risk prediction on the personal insurance products, the user relationship branch is determined to include a blood relationship based on the mapping relationship, and the embodiment of the invention is not particularly limited. After determining the user relationship branch, performing weight value configuration based on the user relationship branch or the number of branch layers of the user relationship branch as the target insurance risk prediction, wherein the number of branch layers is the number of the determined user relationship branches, in an implementation scene based on the number of branch layers configuration weight values, if the number of branch layers is large, the number of the weight value configuration is small, the number of the weight value configuration is between 0 and 1, the total weight value is 1, for example, the number of the branch layers is 4, the configured weight value is 0.4, and the embodiment of the invention is not particularly limited. In the implementation scenario of configuring the weight value based on the user relationship branch, the weight value can be configured according to the relationship with smaller and smaller risk influence corresponding to the blood relationship, the friend relationship, the working relationship and the certificate relationship, for example, the risk influence of the user relationship branch of the blood relationship is maximum, the configured weight value is maximum, for example, 0.4, and in addition, if all the weight values are blood relationship, the weight value can be sequentially reduced according to the father-son relationship, the mother-son relationship, the grandson relationship, the third generation relationship and the like.
In another embodiment of the present invention, for further defining and describing, before the step of performing weight configuration for the target insurance risk prediction based on the social relationship knowledge graph, the method further includes:
acquiring a historical insurance risk training sample, and constructing a neural network model;
when training the neural network model based on the historical insurance risk training sample, determining a model root mean square error and a model average absolute error, and completing training of the neural network model when the model root mean square error and the model average absolute error match a preset error threshold value to obtain the risk prediction model;
and predicting the target insurance basic data of the target user based on the risk prediction model with model training completed, so as to obtain the target insurance prediction result.
In a specific implementation scenario, the current execution end acquires a historical insurance risk training sample in advance, and constructs a neural network model, such as a BP neural network, so as to perform model training on the neural network model based on the historical insurance risk training sample. And when the model root mean square error and the model average absolute error are matched with a preset error threshold, training of the neural network model is completed to obtain a risk prediction model, and target insurance base data of a target user is predicted through the risk prediction model trained by the completed model to obtain a target insurance prediction result. The method for calculating the root mean square error RMSE and the mean absolute error MAE of the model is not particularly limited in the embodiment of the invention.
In the model training process, the historical insurance risk training samples are divided into a training set and a testing set, the input parameters are subjected to dimension reduction processing through Principal Component Analysis (PCA), and the contribution degree of the input parameters is determined. In a specific scenario, main components with contribution degrees greater than 60% can be further sequenced and extracted to serve as input data, user attribute information of blood-cause relationships is determined to be the most dominant influence input parameter by combining main component analysis, the contribution degree is 81%, and the contribution degree is 70% of user attribute information of working relationships. Finally, model training is performed based on the extracted input parameters, and the Root Mean Square Error (RMSE) and the model average absolute error (MAE) index are used for evaluation so as to continuously optimize the prediction result and accurately complete health index prediction model training.
In another embodiment of the present invention, for further defining and describing, before the step of extracting the user characteristic information from the insurance base data, the method further includes:
extracting data attributes in the insurance base data, and determining abnormal base data in the insurance base data according to the data attributes;
And filling and/or filtering the abnormal basic data according to the data attribute to obtain insurance basic data of the user characteristic information to be extracted.
In order to ensure the validity of the insurance base data and thus improve the accuracy of insurance health index determination, the current execution end firstly extracts the data attribute in the insurance base data before extracting the user characteristic information so as to determine the abnormal base data in the insurance base data according to the data attribute. The data attribute is used to represent a data storage form, or attributes such as a data storage form, including but not limited to date data, character string data, picture data, etc., and the embodiment of the present invention is not limited specifically. Specifically, when determining the abnormal basic data, the determination may be made based on a specific data form of the data attribute, for example, the insurance basic data is date data, the date data is stored according to year, month and day, and if the insurance basic data is month, day and minute, the insurance basic data is determined to be the abnormal basic data, and the embodiment of the present invention is not limited specifically. Furthermore, when the abnormal basic data is redundant data, the insurance basic data is deleted, and when the abnormal basic data is missing data, the abnormal basic data is filled according to the data attribute, so that the effective insurance basic data is obtained.
In a specific implementation scenario, the insurance base data includes name, gender, age, etc., if the identification card number intercepts 7-12 bits of birth date data and 2 bits of data are added, deleting the added two bits of data, or when the same identification card number corresponds to a plurality of names, preserving latest name information according to time. In addition, when the missing value is processed, the unique ID is generated for the minor and personnel without the identification card number to ensure the uniqueness, and the 17 th bit of the identification card number can be judged to be even or odd to fill the gender for the field with the empty gender, and the cleaned data is stored in a structured way.
In another embodiment of the present invention, for further defining and describing, after determining the insurance health index of the target user based on the configured weight value and the target insurance risk prediction result, the method further includes:
if the insurance health index is greater than the risk threshold of the target insurance service selected by the target user, determining a product transaction object between the target user and the target insurance service so as to execute the target insurance service based on the product transaction object;
And if the insurance health index is smaller than or equal to the risk threshold of the target insurance service selected by the target user, generating insurance service early warning information of the target user.
In order to achieve the effective purpose of carrying out insurance business processing based on the insurance health index, the current execution end compares the determined insurance health index with the risk threshold of the target insurance business selected by the target user. The insurance service may be transaction of insurance products, insurance underwriting, etc., at this time, different insurance services correspond to different risk thresholds, after determining the target insurance service, a risk threshold preconfigured with the target insurance service is obtained, if the insurance health index is greater than the risk threshold, it is indicated that the target user has a safer implementation purpose for the insurance service, so that a product transaction object between the target user and the target insurance service is determined. The product transaction object is a financial object for performing product or business transaction, including but not limited to electronic payment enterprises, electronic banking enterprises and the like, so that a target insurance business is executed based on the product transaction object. If the insurance health index is smaller than or equal to the risk threshold of the target insurance service selected by the target user, the target user is indicated to have dangerous realization purposes for the insurance service, such as spoofing insurance, and the like, so that insurance service early warning information of the target user is generated.
Compared with the prior art, the embodiment of the invention obtains the insurance basic data of different users and extracts the user characteristic information from the insurance basic data, wherein the user characteristic information is used for representing the contents belonging to different insurance attributes; constructing a social relationship knowledge graph of the user characteristic information according to a structured network, wherein the structured network is configured based on at least one social relationship; after target insurance risk prediction of a target user is completed based on target insurance basic data, weight configuration is carried out on the target insurance risk prediction based on the social relationship knowledge graph, and an insurance health index of the target user is determined based on the configured weight value and a target insurance risk prediction result, so that the accuracy of health index determination is greatly improved based on the relationship among users as an expected basis of the user health index, and the processing effect of the health index-based insurance business is improved.
Further, as an implementation of the method shown in fig. 1, an embodiment of the present invention provides a device for determining an insurance health index, as shown in fig. 4, where the device includes:
The acquiring module 21 is configured to acquire insurance base data of different users, and extract user feature information from the insurance base data, where the user feature information is used to characterize content belonging to different insurance attributes;
a construction module 22, configured to construct a social relationship knowledge graph of the user feature information according to a structured network, where the structured network is configured based on at least one social relationship;
the determining module 23 is configured to perform weight configuration for the target insurance risk prediction based on the social relationship knowledge graph after the target insurance risk prediction of the target user is completed based on the target insurance base data, and determine an insurance health index of the target user based on the configured weight value and the target insurance risk prediction result.
Further, the apparatus further comprises:
the identification module is used for identifying text information used for representing social relations in the insurance base data, wherein the social relations comprise blood relation, friend relation, working relation and certificate relation;
and the configuration module is used for determining a relation branch and a relation quantity based on the text information and configuring the structured network based on the relation branch and the relation quantity.
Further, the construction module is specifically configured to determine that the user characteristic information belongs to a target relationship branch in the blood relationship, the friend relationship, the work relationship and the certificate relationship; and drawing the user identification of the user in the structured network according to the target relationship branch to obtain a social relationship knowledge graph.
Further, the determining module is specifically configured to obtain a social relationship influence risk rule, where the social relationship influence risk rule is used to characterize a judging method for risk caused to different insurance services and insurance products; and determining a user relationship branch corresponding to the target user in the social relationship knowledge graph according to the social relationship influence risk rule, and configuring a weight value of the target insurance risk prediction based on the user relationship branch or the branch layer number of the user relationship branch.
Further, the apparatus further comprises: the prediction module is used for predicting the number of the blocks,
the acquisition module is also used for acquiring a historical insurance risk training sample and constructing a neural network model;
the determining module is further configured to determine a model root mean square error and a model average absolute error when the neural network model is trained based on the historical insurance risk training sample, and complete training of the neural network model when the model root mean square error and the model average absolute error match a preset error threshold value, so as to obtain the risk prediction model;
And the prediction module is used for predicting the target insurance basic data of the target user based on the risk prediction model which is trained by the model, and obtaining the target insurance prediction result.
Further, the apparatus further comprises:
the extraction module is used for extracting data attributes in the insurance base data and determining abnormal base data in the insurance base data according to the data attributes; and filling and/or filtering the abnormal basic data according to the data attribute to obtain insurance basic data of the user characteristic information to be extracted.
Further, the apparatus further comprises: the generation module is configured to generate a first set of data,
the determining module is further configured to determine a product transaction object between the target user and the target insurance service if the insurance health index is greater than a risk threshold of the target insurance service selected by the target user, so as to execute the target insurance service based on the product transaction object;
the generation module is configured to generate insurance business early warning information of the target user if the insurance health index is less than or equal to a risk threshold of the target insurance business selected by the target user.
Compared with the prior art, the embodiment of the invention obtains the insurance basic data of different users and extracts the user characteristic information from the insurance basic data, wherein the user characteristic information is used for representing the contents belonging to different insurance attributes; constructing a social relationship knowledge graph of the user characteristic information according to a structured network, wherein the structured network is configured based on at least one social relationship; after target insurance risk prediction of a target user is completed based on target insurance basic data, weight configuration is carried out on the target insurance risk prediction based on the social relationship knowledge graph, and an insurance health index of the target user is determined based on the configured weight value and a target insurance risk prediction result, so that the accuracy of health index determination is greatly improved based on the relationship among users as an expected basis of the user health index, and the processing effect of the health index-based insurance business is improved.
According to one embodiment of the present invention, there is provided a storage medium storing at least one executable instruction for performing the method of determining an insurance health index in any of the above-described method embodiments.
Fig. 5 is a schematic structural diagram of a terminal according to an embodiment of the present invention, and the specific embodiment of the present invention is not limited to the specific implementation of the terminal.
As shown in fig. 5, the terminal may include: a processor (processor) 302, a communication interface (Communications Interface) 304, a memory (memory) 306, and a communication bus 308.
Wherein: processor 302, communication interface 304, and memory 306 perform communication with each other via communication bus 308.
A communication interface 304 for communicating with network elements of other devices, such as clients or other servers.
The processor 302 is configured to execute the program 310, and may specifically perform relevant steps in the above-described embodiment of the method for determining an insurance health index.
In particular, program 310 may include program code including computer-operating instructions.
The processor 302 may be a central processing unit CPU, or a specific integrated circuit ASIC (Application Specific Integrated Circuit), or one or more integrated circuits configured to implement embodiments of the present invention. The one or more processors included in the terminal may be the same type of processor, such as one or more CPUs; but may also be different types of processors such as one or more CPUs and one or more ASICs.
Memory 306 for storing programs 310. Memory 306 may comprise high-speed RAM memory or may also include non-volatile memory (non-volatile memory), such as at least one disk memory.
Program 310 may be specifically operable to cause processor 302 to:
acquiring insurance basic data of different users, and extracting user characteristic information from the insurance basic data, wherein the user characteristic information is used for representing contents belonging to different insurance attributes;
constructing a social relationship knowledge graph of the user characteristic information according to a structured network, wherein the structured network is configured based on at least one social relationship;
and after the target insurance risk prediction of the target user is completed based on the target insurance base data, carrying out weight configuration on the target insurance risk prediction based on the social relationship knowledge graph, and determining the insurance health index of the target user based on the configured weight value and the target insurance risk prediction result.
It will be appreciated by those skilled in the art that the modules or steps of the invention described above may be implemented in a general purpose computing device, they may be concentrated on a single computing device, or distributed across a network of computing devices, they may alternatively be implemented in program code executable by computing devices, so that they may be stored in a memory device for execution by computing devices, and in some cases, the steps shown or described may be performed in a different order than that shown or described, or they may be separately fabricated into individual integrated circuit modules, or multiple modules or steps within them may be fabricated into a single integrated circuit module for implementation. Thus, the present invention is not limited to any specific combination of hardware and software.
The above description is only of the preferred embodiments of the present invention and is not intended to limit the present invention, but various modifications and variations can be made to the present invention by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (10)
1. A method of determining an insurance health index, comprising:
acquiring insurance basic data of different users, and extracting user characteristic information from the insurance basic data, wherein the user characteristic information is used for representing contents belonging to different insurance attributes;
constructing a social relationship knowledge graph of the user characteristic information according to a structured network, wherein the structured network is configured based on at least one social relationship;
and after the target insurance risk prediction of the target user is completed based on the target insurance base data, carrying out weight configuration on the target insurance risk prediction based on the social relationship knowledge graph, and determining the insurance health index of the target user based on the configured weight value and the target insurance risk prediction result.
2. The method according to claim 1, wherein before the social relationship knowledge graph of the user feature information is constructed according to the structured network, the method further comprises:
Identifying text information used for representing social relations in the insurance base data, wherein the social relations comprise blood relation, friend relation, working relation and certificate relation;
determining a relationship branch and a relationship quantity based on the text information, and configuring the structured network based on the relationship branch and the relationship quantity.
3. The method according to claim 2, wherein the constructing the social relationship knowledge graph of the user feature information according to the structured network comprises:
determining that the user characteristic information belongs to a target relationship branch in the blood relationship, the friend relationship, the work relationship and the certificate relationship;
and drawing the user identification of the user in the structured network according to the target relationship branch to obtain a social relationship knowledge graph.
4. The method of claim 1, wherein the weighting the target insurance risk prediction based on the social relationship knowledge graph comprises:
acquiring social relationship influence risk rules, wherein the social relationship influence risk rules are used for representing judgment methods for risks caused by different insurance businesses and insurance products;
And determining a user relationship branch corresponding to the target user in the social relationship knowledge graph according to the social relationship influence risk rule, and configuring a weight value of the target insurance risk prediction based on the user relationship branch or the branch layer number of the user relationship branch.
5. The method of claim 1, wherein prior to the weighting the target insurance risk prediction based on the social relationship knowledge graph, the method further comprises:
acquiring a historical insurance risk training sample, and constructing a neural network model;
when training the neural network model based on the historical insurance risk training sample, determining a model root mean square error and a model average absolute error, and completing training of the neural network model when the model root mean square error and the model average absolute error match a preset error threshold value to obtain the risk prediction model;
and predicting the target insurance basic data of the target user based on the risk prediction model with model training completed, so as to obtain the target insurance prediction result.
6. The method of claim 1, wherein prior to extracting user characteristic information from the insurance base data, the method further comprises:
Extracting data attributes in the insurance base data, and determining abnormal base data in the insurance base data according to the data attributes;
and filling and/or filtering the abnormal basic data according to the data attribute to obtain insurance basic data of the user characteristic information to be extracted.
7. The method of any of claims 1-6, wherein after the determining the target user's insurance health index based on the configured weight values and target insurance risk prediction results, the method further comprises:
if the insurance health index is greater than the risk threshold of the target insurance service selected by the target user, determining a product transaction object between the target user and the target insurance service so as to execute the target insurance service based on the product transaction object;
and if the insurance health index is smaller than or equal to the risk threshold of the target insurance service selected by the target user, generating insurance service early warning information of the target user.
8. A device for determining an insurance health index, comprising:
the system comprises an acquisition module, a storage module and a storage module, wherein the acquisition module is used for acquiring insurance basic data of different users and extracting user characteristic information from the insurance basic data, wherein the user characteristic information is used for representing contents belonging to different insurance attributes;
The construction module is used for constructing a social relationship knowledge graph of the user characteristic information according to a structural network, and the structural network is configured based on at least one social relationship;
and the determining module is used for carrying out weight configuration on the target insurance risk prediction based on the social relationship knowledge graph after the target insurance risk prediction of the target user is completed based on the target insurance base data, and determining the insurance health index of the target user based on the configured weight value and the target insurance risk prediction result.
9. A storage medium having stored therein at least one executable instruction for causing a processor to perform operations corresponding to the method of determining an insurance health index according to any one of claims 1 to 7.
10. A terminal, comprising: the device comprises a processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface complete communication with each other through the communication bus;
the memory is configured to store at least one executable instruction that causes the processor to perform operations corresponding to the method for determining an insurance health index according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311553364.8A CN117764742A (en) | 2023-11-20 | 2023-11-20 | Method and device for determining insurance health index |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311553364.8A CN117764742A (en) | 2023-11-20 | 2023-11-20 | Method and device for determining insurance health index |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117764742A true CN117764742A (en) | 2024-03-26 |
Family
ID=90320905
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311553364.8A Pending CN117764742A (en) | 2023-11-20 | 2023-11-20 | Method and device for determining insurance health index |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117764742A (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160048766A1 (en) * | 2014-08-13 | 2016-02-18 | Vitae Analytics, Inc. | Method and system for generating and aggregating models based on disparate data from insurance, financial services, and public industries |
| CN108734479A (en) * | 2018-04-12 | 2018-11-02 | 阿里巴巴集团控股有限公司 | Data processing method, device, equipment and the server of Insurance Fraud identification |
| US20220351209A1 (en) * | 2021-04-29 | 2022-11-03 | Swiss Reinsurance Company Ltd. | Automated fraud monitoring and trigger-system for detecting unusual patterns associated with fraudulent activity, and corresponding method thereof |
| CN115660814A (en) * | 2022-11-08 | 2023-01-31 | 中国工商银行股份有限公司 | Risk prediction method and device, computer readable storage medium and electronic equipment |
| CN116934507A (en) * | 2023-09-19 | 2023-10-24 | 国任财产保险股份有限公司 | Intelligent claim settlement method and system based on big data driving |
-
2023
- 2023-11-20 CN CN202311553364.8A patent/CN117764742A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160048766A1 (en) * | 2014-08-13 | 2016-02-18 | Vitae Analytics, Inc. | Method and system for generating and aggregating models based on disparate data from insurance, financial services, and public industries |
| CN108734479A (en) * | 2018-04-12 | 2018-11-02 | 阿里巴巴集团控股有限公司 | Data processing method, device, equipment and the server of Insurance Fraud identification |
| US20220351209A1 (en) * | 2021-04-29 | 2022-11-03 | Swiss Reinsurance Company Ltd. | Automated fraud monitoring and trigger-system for detecting unusual patterns associated with fraudulent activity, and corresponding method thereof |
| CN115660814A (en) * | 2022-11-08 | 2023-01-31 | 中国工商银行股份有限公司 | Risk prediction method and device, computer readable storage medium and electronic equipment |
| CN116934507A (en) * | 2023-09-19 | 2023-10-24 | 国任财产保险股份有限公司 | Intelligent claim settlement method and system based on big data driving |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11546223B2 (en) | Systems and methods for conducting more reliable assessments with connectivity statistics | |
| US11475143B2 (en) | Sensitive data classification | |
| US10785241B2 (en) | URL attack detection method and apparatus, and electronic device | |
| CN107025596B (en) | Risk assessment method and system | |
| EP3713191B1 (en) | Identifying legitimate websites to remove false positives from domain discovery analysis | |
| US10637826B1 (en) | Policy compliance verification using semantic distance and nearest neighbor search of labeled content | |
| CN110659318B (en) | Big data-based policy pushing method, system and computer equipment | |
| CN108629413A (en) | Neural network model training, trading activity Risk Identification Method and device | |
| MX2012003721A (en) | Systems and methods for social graph data analytics to determine connectivity within a community. | |
| WO2011134086A1 (en) | Systems and methods for conducting reliable assessments with connectivity information | |
| CN106874253A (en) | Recognize the method and device of sensitive information | |
| CN113656808A (en) | Data security evaluation method, device, equipment and storage medium | |
| CN107229605B (en) | Text similarity calculation method and device | |
| CN114338064A (en) | Method, device, equipment and storage medium for identifying network traffic type | |
| CN113010785B (en) | User recommendation method and device | |
| CN115115369A (en) | Data processing method, device, equipment and storage medium | |
| CN115603955B (en) | Abnormal access object identification method, device, equipment and medium | |
| CN111507850A (en) | Authority guaranteeing method and related device and equipment | |
| CN116402546A (en) | Store risk attribution method and device, equipment, medium and product thereof | |
| CN117764742A (en) | Method and device for determining insurance health index | |
| CN114170000A (en) | Credit card user risk category identification method, device, computer equipment and medium | |
| CN114925275A (en) | Product recommendation method and device, computer equipment and storage medium | |
| CN113627997A (en) | Data processing method and device, electronic equipment and storage medium | |
| CN113239126A (en) | Business activity information standardization scheme based on BOR method | |
| CN113269179A (en) | Data processing method, device, equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |