CN115760437A - Insurance applicant recommendation method based on safety management measures and insurance letters - Google Patents
Insurance applicant recommendation method based on safety management measures and insurance letters Download PDFInfo
- Publication number
- CN115760437A CN115760437A CN202211181921.3A CN202211181921A CN115760437A CN 115760437 A CN115760437 A CN 115760437A CN 202211181921 A CN202211181921 A CN 202211181921A CN 115760437 A CN115760437 A CN 115760437A
- Authority
- CN
- China
- Prior art keywords
- insurance
- applicant
- value
- safety management
- reliability
- 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
- 238000000034 method Methods 0.000 title claims abstract description 44
- 230000006870 function Effects 0.000 claims description 54
- 238000000605 extraction Methods 0.000 claims description 30
- 239000013598 vector Substances 0.000 claims description 17
- 241001622623 Coeliadinae Species 0.000 claims description 11
- 238000012549 training Methods 0.000 description 18
- 230000002457 bidirectional effect Effects 0.000 description 16
- 230000000873 masking effect Effects 0.000 description 13
- 230000000694 effects Effects 0.000 description 6
- 238000010586 diagram Methods 0.000 description 5
- 238000004590 computer program Methods 0.000 description 4
- 238000010276 construction Methods 0.000 description 4
- 230000014759 maintenance of location Effects 0.000 description 3
- 238000002372 labelling 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
- 230000001360 synchronised effect Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/04—Trading; Exchange, e.g. stocks, commodities, derivatives or currency exchange
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/08—Insurance
Abstract
The invention discloses an insurance applicant recommendation method based on safety management measures and insurance letters, which comprises the steps of obtaining corresponding insurance letter characteristics of a plurality of insurance appliers, and matching the insurance appliers according to the insurance letter characteristics to obtain a plurality of comparison groups; acquiring safety management measures and insurance value respectively corresponding to each insurance applicant, wherein the insurance value corresponding to each insurance applicant is determined based on the safety accident risk and the insurance value corresponding to the insurance applicant; comparing the safety management measures and the insurable value of the insurant in each comparison group in a group to obtain the reliability corresponding to each safety management measure; and determining the target applicant from the applicant according to the reliability corresponding to each safety management measure. The problem of among the prior art recommend the policyholder to the guarantor through the insurance data of aassessment applicant, because insurance data can't indicate policyholder's risk, therefore recommend the policyholder that the insurance value is high easily, and the risk is also high, lead to the guarantor to undertake too high risk is solved.
Description
Technical Field
The invention relates to the field of insurance application, in particular to an applicant recommendation method based on safety management measures and insurance.
Background
The electronic insurance letter for construction engineering is an indispensable ring in the field of construction engineering, and refers to a written credit guarantee voucher issued by banks, insurance companies and guarantee companies to a third party at the request of an applicant. The insurance is mostly issued by insurance companies with strong capital strength, and can be based on credit without the traditional guarantee of mortgage, thereby relieving the capital pressure of enterprises in public resource transaction. Therefore, the insurance has certain economic value. In the prior art, an insurance applicant is recommended to a guarantor by evaluating insurance data of the insurance applicant, and the insurance data cannot prompt risks of the insurance applicant, so that the insurance applicant with high insurance value and high risk is easy to recommend, and the problem that the guarantor needs to bear too high risk is caused.
Thus, there is still a need for improvement and development of the prior art.
Disclosure of Invention
The technical problem to be solved by the invention is to provide an applicant recommendation method based on safety management measures and insurance letters aiming at solving the problem that in the prior art, insurance letters of an applicant are recommended to a guarantor by evaluating insurance letter data of the applicant, and because the insurance letter data cannot prompt the risk of the applicant, the applicant is easy to recommend the insurance letter with high value and high risk, and the risk of the applicant is required to be too high.
The technical scheme adopted by the invention for solving the problems is as follows:
in a first aspect, an embodiment of the present invention provides a method for applicant recommendation based on security management measures and insurance, where the method includes:
acquiring the insurance characteristics corresponding to a plurality of policemen respectively, and matching each policemen according to each insurance characteristic to obtain a plurality of control groups;
acquiring safety management measures and insurance application values corresponding to the insurance applicants respectively, wherein the insurance application value corresponding to each insurance applicant is determined based on the safety accident risk and the insurance letter value corresponding to the insurance applicant;
comparing the safety management measures and the insurance value of the policyholder contained in each comparison group in a group to obtain the reliability corresponding to each safety management measure;
and determining a target applicant from the applicant according to the reliability corresponding to each safety management measure.
In one embodiment, the obtaining of the corresponding insurance characteristics of the plurality of applicant comprises:
acquiring insurance letter data and a preset feature extraction template corresponding to each policyholder respectively;
and extracting the insurance function characteristics corresponding to the applicant from the insurance function data according to the characteristic extraction template.
In one embodiment, said insurance features are vector data, and said matching each said applicant according to each said insurance feature results in a plurality of control groups, including:
determining cosine vector similarity corresponding to each applicant according to the insurance letter characteristics of each applicant;
and determining each comparison group according to the cosine vector similarity corresponding to each applicant, wherein the cosine vector similarity corresponding to each applicant in each comparison group is the highest.
In one embodiment, the process of determining the value of the application corresponding to each of the applicants comprises:
acquiring a safety accident record corresponding to the applicant, and determining the safety accident risk corresponding to the applicant according to the safety accident record;
determining a penalty value according to the safety accident risk corresponding to the applicant;
and acquiring the insurance value corresponding to the applicant, and determining the insurance value corresponding to the applicant according to the difference value of the insurance value and the penalty value.
In one embodiment, said obtaining said value of said insurance policy for said applicant comprises:
inputting the insurance letter data corresponding to the applicant into a pre-trained target extraction model to obtain an insurance letter knowledge graph corresponding to the applicant;
and inputting the insurance function knowledge graph into a pre-trained target prediction model to obtain the value of the insurance function corresponding to the applicant.
In one embodiment, each of the comparison groups includes two policemen, and the performing group comparison on the security management measures and the insurable value of the policemen included in each of the comparison groups to obtain the reliability corresponding to each of the security management measures includes:
determining the initial reliability of the two safety management measures corresponding to each comparison group according to the two insurable values corresponding to each comparison group;
determining a reward and punishment value corresponding to each control group according to the difference value of the two insurance values corresponding to each control group;
performing reliability reward on the safety management measure with the highest initial reliability in the comparison group according to the reward and punishment value to obtain the reliability corresponding to the safety management measure;
and carrying out reliability punishment on the safety management measure with the lowest initial reliability in the comparison group according to the reward and punishment value to obtain the reliability corresponding to the safety management measure.
In one embodiment, the determining a target applicant from each of the applicant according to the reliability corresponding to each of the security management measures comprises:
and taking the applicant corresponding to the safety management measure with the highest reliability as the target applicant.
In a second aspect, an embodiment of the present invention further provides a system for recommending insurance applicants based on security management measures and insurance letters, where the system includes:
the matching module is used for acquiring the insurance function characteristics corresponding to a plurality of insurance applicants respectively, and matching each insurance applicants according to each insurance function characteristic to obtain a plurality of comparison groups;
the acquiring module is used for acquiring safety management measures and insurance application values corresponding to the insurance applicants respectively, wherein the insurance application value corresponding to each insurance applicant is determined based on the safety accident risk and the insurance letter value corresponding to the insurance applicant;
the comparison module is used for carrying out group-in comparison on the safety management measures and the insurance value of the insurance applicant contained in each comparison group to obtain the reliability corresponding to each safety management measure;
and the selection module is used for determining a target applicant from the applicant according to the reliability corresponding to each safety management measure.
In a third aspect, an embodiment of the present invention further provides a terminal, where the terminal includes a storage and one or more processors; the memory stores more than one program; the program contains instructions for performing any of the above described methods of insurance applicant recommendation based on security management measures and insurance coverage; the processor is configured to execute the program.
In a fourth aspect, embodiments of the present invention further provide a computer-readable storage medium having stored thereon a plurality of instructions, wherein the instructions are adapted to be loaded and executed by a processor to perform any of the above-described steps of a method for applicant recommendation based on safety management measures and insurance coverage.
The invention has the beneficial effects that: in the embodiment of the invention, a plurality of comparison groups are obtained by acquiring the insurance function characteristics corresponding to a plurality of insurance applicants respectively and matching each insurance applicant according to each insurance function characteristic; acquiring safety management measures and insurance value respectively corresponding to each insurance applicant, wherein the insurance value corresponding to each insurance applicant is determined based on the safety accident risk and the insurance value corresponding to the insurance applicant; comparing the safety management measures and the insurable value of the insurant in each comparison group in a group to obtain the reliability corresponding to each safety management measure; and determining the target applicant from the applicant according to the reliability corresponding to each safety management measure. The problem of among the prior art recommend the insurant to the guarantor through evaluating the insurance data of the insurant, because the risk of insurant can not be indicateed to the insurance data, therefore recommend the insurant that the insurance value is high, and the risk is also high easily, lead to the guarantor to need to undertake too high risk is solved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments described in the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
FIG. 1 is a flowchart illustrating an applicant recommendation method based on security management measures and insurance coverage according to an embodiment of the present invention.
FIG. 2 is a block diagram of an applicant recommendation system based on security management measures and insurance functions according to an embodiment of the present invention.
Fig. 3 is a schematic block diagram of a terminal according to an embodiment of the present invention.
Detailed Description
The invention discloses an applicant recommendation method based on safety management measures and insurance letters, and in order to make the purposes, technical schemes and effects of the invention clearer and clearer, the invention is further described in detail by referring to the attached drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
As used herein, the singular forms "a", "an", "the" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms "comprises" and/or "comprising," when used in this specification, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof. It will be understood that when an element is referred to as being "connected" or "coupled" to another element, it can be directly connected or coupled to the other element or intervening elements may also be present. Further, "connected" or "coupled" as used herein may include wirelessly connected or wirelessly coupled. As used herein, the term "and/or" includes all or any element and all combinations of one or more of the associated listed items.
It will be understood by those skilled in the art that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. It will be further understood that terms, such as those defined in commonly used dictionaries, should be interpreted as having a meaning that is consistent with their meaning in the context of the prior art and will not be interpreted in an idealized or overly formal sense unless expressly so defined herein.
The electronic insurance letter for construction engineering is an indispensable ring in the field of construction engineering, and refers to a written credit guarantee voucher issued by banks, insurance companies and guarantee companies to a third party at the request of an applicant. The insurance is mostly provided by insurance companies with strong capital strength, and can be based on credit without the traditional mortgage guarantee, thereby relieving the capital pressure of enterprises in public resource transaction. So the insurance policy has a certain economic value. In the prior art, an insurance applicant is recommended to a guarantor by evaluating insurance data of the insurance applicant, and the insurance data cannot prompt risks of the insurance applicant, so that the insurance applicant with high insurance value and high risk is easy to recommend, and the problem that the guarantor needs to bear too high risk is caused.
In view of the above-mentioned drawbacks of the prior art, the present invention provides a method for applicant recommendation based on security management measures and insurances, the method comprising: acquiring the insurance characteristics corresponding to a plurality of policemen respectively, and matching each policemen according to each insurance characteristic to obtain a plurality of control groups; acquiring safety management measures and insurance application values corresponding to the insurance applicants respectively, wherein the insurance application value corresponding to each insurance applicant is determined based on the safety accident risk and the insurance letter value corresponding to the insurance applicant; comparing the safety management measures and the insurance value of the insurance applicant contained in each comparison group in a group to obtain the reliability corresponding to each safety management measure; and determining a target applicant from the applicant according to the reliability corresponding to each safety management measure. The problem of among the prior art recommend the policyholder to the guarantor through the insurance data of aassessment applicant, because insurance data can't indicate policyholder's risk, therefore recommend the policyholder that the insurance value is high easily, and the risk is also high, lead to the guarantor to undertake too high risk is solved.
Exemplary method
As shown in fig. 1, the method includes:
and S100, acquiring the insurance characteristics corresponding to a plurality of insurance applicants respectively, and matching each insurance applicants according to each insurance characteristic to obtain a plurality of comparison groups.
Specifically, the present embodiment first identifies a plurality of applicants who are interested in the guarantor. And acquiring the insurance data corresponding to each applicant, and extracting the insurance characteristics of the applicant according to the insurance data. And then according to the insurance characteristics of the insurance applicants, classifying the insurance applicants with similar insurance characteristics into the same control group to obtain a plurality of control groups.
In one implementation, the step S100 specifically includes:
s101, acquiring insurance letter data and a preset feature extraction template corresponding to each policyholder;
and S102, extracting the insurance function characteristics corresponding to each applicant from each insurance function data according to the characteristic extraction template.
Specifically, in order to extract the feature of the insurance coverage, the present embodiment designs in advance a feature extraction template for different types of insurance applicants, for example, the types of insurance applicants include contractors, suppliers, labor subcontractors, and the like. And determining a characteristic extraction template corresponding to each current applicant, wherein the characteristic extraction template comprises typical variable characteristics of the applicant, such as the establishment years, the registered capital, the revenue data, the personnel scale and the like. The insurance letter characteristics corresponding to each applicant can be extracted by obtaining the insurance letter data of each applicant and using the characteristic extraction template, and the insurance letter characteristics of each applicant are uniformly expressed in a vector form.
In one implementation, the matching of each insurance applicant according to each insurance feature to obtain a plurality of control groups includes:
s103, determining cosine vector similarity corresponding to each policyholder according to the insurance function characteristics of each policyholder;
and S104, determining each comparison group according to the cosine vector similarity corresponding to each policyholder, wherein the cosine vector similarity corresponding to each policyholder in each comparison group is the highest.
Specifically, the present embodiment represents the insurance features of each applicant in vector form. And (3) sequentially calculating cosine vector similarity of the insurance characteristics of the applicant and the insurance characteristics of other people for each applicant, wherein the higher the cosine vector similarity is, the closer the two vectors are, namely the more similar the insurance characteristics of the two applicant are. The other applicant with the highest similarity to the cosine vector is assigned to the same control group. Since the insurance characteristics between two applicant in the same control group are highly similar, the effect of other factors besides the insurance characteristics on the value of the insurance can be effectively observed by each control group.
As shown in fig. 1, the method further comprises:
and S200, acquiring safety management measures and insurance value corresponding to each insurance applicant, wherein the insurance value corresponding to each insurance applicant is determined based on the safety accident risk and the insurance letter value corresponding to the insurance applicant.
Specifically, in order to recommend an applicant of better quality to a guarantor, the present embodiment requires not only consideration of the insurable value of each applicant but also comprehensive consideration of the safety management measures taken by each applicant, respectively, when making an applicant recommendation. In one implementation, the insurable value of each insurant is determined based on the risk of a safety accident and the value of a insurant corresponding to the insurant, thereby excluding high risk insurants.
In one implementation, the process of determining the value of the application corresponding to each of the applicant comprises:
step S201, obtaining a safety accident record corresponding to the applicant, and determining the safety accident risk corresponding to the applicant according to the safety accident record;
step S202, determining a penalty value according to the safety accident risk corresponding to the applicant;
and S203, acquiring the insurance value corresponding to the applicant, and determining the insurance value corresponding to the applicant according to the difference value of the insurance value and the penalty value.
Specifically, for each applicant, a safety accident record of the applicant's company is obtained, and then the safety accident risk of the applicant is predicted according to the safety accident record. When the risk of the safety accident of the applicant is high, the applicant is shown to be a high-risk applicant, a penalty value is given to the applicant, the penalty value is subtracted on the basis of the corresponding insurance function value to generate a final insurance value, the purpose of reducing the insurance value of the high-risk applicant is achieved, and the probability of recommending the high-risk applicant to the applicant is reduced.
In one implementation, the obtaining the value of the insurance corresponding to the applicant comprises:
step S204, inputting the insurance letter data corresponding to the applicant into a pre-trained target extraction model to obtain an insurance letter knowledge graph corresponding to the applicant;
and S205, inputting the insurance function knowledge graph into a pre-trained target prediction model to obtain the insurance value corresponding to the applicant.
Specifically, the present embodiment predicts the value of each applicant's insurance coverage by constructing a knowledge graph. In the embodiment, a target extraction model is constructed in advance, and the target extraction model learns the complex mapping relation between different functional data and functional knowledge maps through mass data in advance. After the insurance data of each applicant is input into the trained target extraction model, the target extraction model can automatically extract the information in the insurance data and generate the corresponding insurance intellectual map of the applicant. In the embodiment, a target prediction model is also constructed in advance, and the target prediction model learns the complex mapping relation between different insurance function knowledge maps and the insurance value through mass data in advance, so that the insurance function knowledge map of the applicant is input into the trained target prediction model, and the target prediction model can output the insurance value corresponding to the applicant.
In one implementation, the specific generation process of the insurance value includes:
inputting the insurance function data into a trained target extraction model to obtain a plurality of triples corresponding to the insurance function data, wherein each triplet is used for reflecting the relationship between two entities in the insurance function data;
generating a corresponding insurance function knowledge graph of the insurance function data according to each triple;
determining an insurance function type corresponding to the insurance function data, and determining a trained target prediction model corresponding to the insurance function data according to the insurance function type;
and inputting the insurance function knowledge graph into the target prediction model to obtain a predicted insurance function value.
In one implementation manner, the training process corresponding to the target extraction model includes:
obtaining historical functional preserving data, and determining a plurality of first training data according to the historical functional preserving data, wherein each first training data comprises a statement in the historical functional preserving data and labeling information corresponding to the statement, and the labeling information is used for reflecting the relationship between an entity contained in the statement and each entity;
acquiring a pre-trained target bidirectional language model, wherein input data of the target bidirectional language model is a functional data statement masked by a mask, and output data of the target bidirectional language model is a predicted word masked by the mask;
adjusting the target bidirectional language model to obtain an extraction model, wherein input data of the extraction model is statements in the historical functional data, and output data of the extraction model is predicted relations between entities and each entity contained in the statements;
and performing iterative training on the extraction model according to the first training data to obtain the target extraction model.
In one implementation manner, the training process corresponding to the target extraction model includes:
obtaining historical functional data, and determining a plurality of first training data according to the historical functional data, wherein each first training data comprises a statement in the historical functional data and marking information corresponding to the statement, and the marking information is used for reflecting the relationship between an entity contained in the statement and each entity;
acquiring a pre-trained target bidirectional language model, wherein input data of the target bidirectional language model is a functional data statement masked by a mask, and output data of the target bidirectional language model is a predicted word masked by the mask;
adjusting the target bidirectional language model to obtain an extraction model, wherein input data of the extraction model is a statement in the historical functional data, and output data of the extraction model is a predicted relation between an entity contained in the statement and each entity;
and performing iterative training on the extraction model according to the first training data to obtain the target extraction model.
In one implementation, the training process corresponding to the target two-way language model includes:
determining a plurality of said functional data statements according to said historical functional data;
masking words in the preserved function data statement through a mask to obtain a masked statement, and generating label information corresponding to the masked statement according to the masked words;
inputting the masking sentences into a bidirectional language model which is not trained completely to obtain predicted words generated by the bidirectional language model based on the masking sentences;
generating a first loss function value corresponding to the bidirectional language model according to the predicted words and the label information;
judging whether the first loss function value is converged to a target value, if not, updating parameters of the bidirectional language model according to the first loss function value to obtain an updated bidirectional language model;
and taking the updated bidirectional language model as the bidirectional language model, and continuously executing masking on words in the functional data sentence through mask to obtain a masked sentence until the first loss function value is converged to the target value to obtain the target bidirectional language model.
In one implementation, the masking words in the functional data statement by masking results in a masked statement, including:
judging whether the first loss function value corresponding to the previous round of training converges to an intermediate value, wherein the intermediate value is an intermediate value between the first loss function value corresponding to the first round of training and the target value;
if the first loss function value corresponding to the previous round of training does not converge to the intermediate value, randomly masking words in the function-preserving data statement through the mask to obtain the masked statement;
if the first loss function value corresponding to the previous round of training converges to the intermediate value, obtaining masking probabilities corresponding to the words in the retention data sentence respectively, wherein the masking probability corresponding to each word is in a direct proportion relation with the contribution degree of the word to the retention value of the historical retention data;
masking the words in the functional data statement to obtain the masking statement through the masking based on the masking probability corresponding to each word in the functional data statement.
In one implementation, the generating, according to each triplet, an insurance-based knowledge graph corresponding to the insurance data includes:
generating nodes in the functional knowledge graph in a one-to-one correspondence mode according to entities contained in each triple;
and connecting the nodes in the insurance function knowledge graph according to the relationship among the entities contained in the triples to obtain the insurance function knowledge graph, wherein the relationships of different types respectively correspond to the connecting lines of different types.
In one implementation, the functional knowledge-graph further includes attention weights corresponding to the nodes, and the determining process of the attention weight corresponding to each node includes:
taking the node as a starting point, performing neighborhood search on the insurance function knowledge graph to obtain all associated nodes corresponding to the node;
determining a relation frame corresponding to each association node according to each association node, wherein the relation frame is a minimum bounding box containing each association node;
and determining the attention weight corresponding to the node according to the size of the relation frame.
In one implementation, the training process corresponding to the target prediction model includes:
acquiring a historical functional knowledge graph and a functional value corresponding to the historical functional knowledge graph, and inputting the historical functional knowledge graph into a prediction model which is not trained to obtain a training functional value; the style of the insurance corresponding to the historical insurance function knowledge graph is the same as the insurance data, and the prediction model is an attention model;
determining a second loss function value corresponding to the prediction model according to the minimum mean square error of the insurance function value and the training insurance function value;
judging whether the second loss function value converges to a preset value, if not, updating parameters of the prediction model according to the second loss function value to obtain an updated prediction model;
and taking the updated prediction model as the prediction model, continuously executing the steps of obtaining a historical function-preserving knowledge graph and a function-preserving value corresponding to the historical function-preserving knowledge graph, and inputting the historical function-preserving knowledge graph into the prediction model which is not trained until the obtained second loss function value converges to the preset value, so as to obtain the target prediction model.
As shown in fig. 1, the method further comprises:
step S300, comparing the safety management measures and the insurance value of the insurance applicant in each comparison group in groups to obtain the reliability corresponding to each safety management measure.
Specifically, because the insurance characteristics of the applicant in the same contrast group are similar, the difference in the application value of the applicant in the same contrast group can only be caused by the difference in the safety management measures. Therefore, in the embodiment, the reliability corresponding to each type of safety management measure can be accurately determined by comparing the safety management measure and the bid value of each applicant in each comparison group.
In one implementation manner, each comparison group includes two insurance applicants, and performing intra-group comparison on the security management measures and the insurance value of the insurance applicants included in each comparison group to obtain the reliability corresponding to each security management measure respectively includes:
step S301, determining the initial reliability of the two safety management measures corresponding to each contrast group according to the two insurable values corresponding to each contrast group;
step S302, determining a reward and punishment value corresponding to each control group according to the difference value of the two insurance values corresponding to each control group;
step S303, carrying out reliability reward on the safety management measure with the highest initial reliability in the comparison group according to the reward and punishment value to obtain the reliability corresponding to the safety management measure;
and S304, carrying out reliability punishment on the safety management measure with the lowest initial reliability in the control group according to the reward punishment value to obtain the reliability corresponding to the safety management measure.
Specifically, because the insurance characteristics of two policemen in the same control group are similar, the only variable causing different insurance values of the two policemen is different safety accident risks caused by safety management measures respectively adopted by the two policemen. For the same contrast group, the higher the insurable value is, the lower the risk of safety accidents of the corresponding insurant is possibly, the higher the insurable value is possibly, the more reliable the adopted safety management measures are, the higher the initial reliability is given to the insurable person; the lower the value of the application, the higher the risk of safety accidents for the corresponding applicant is likely to be, and the lower the value of the insurance letter is likely to be, the less reliable the safety management measures it adopts, and the lower initial reliability it is given to. In addition, the larger the deviation between two insurance application values of the same contrast group is, the more obvious the effect of the safety management measures respectively adopted by the corresponding two insurance appliers is shown to be different, so that the reward and punishment value can be set according to the deviation between the two insurance application values, the reliability of the safety management measures with the obvious good effect can be rewarded, the reliability of the safety management measures with the obvious poor effect can be punished, and the safety management measures with different effects can be better distinguished.
As shown in fig. 1, the method further comprises:
and S400, determining a target applicant from the applicant according to the reliability corresponding to each safety management measure.
Specifically, since the higher the reliability of the security management measures, the lower the risk of a possible future security incident for the applicant, an appropriate target applicant may be recommended for the insurer based on the reliability of each security management measure.
In one implementation, the step S400 specifically includes:
step S401, the policyholder corresponding to the safety management measure with the highest reliability is taken as the target policyholder.
In particular, the applicant with the highest degree of reliability of security management measures has a low probability of developing a risk of a security accident, and thus can be the target applicant.
In one implementation, the insurance function values corresponding to all the insurance applicants can be acquired, the insurance applicants with the insurance function values in the first few positions are used as candidate insurance applicants, and the insurance applicant with the highest reliability of the safety management measures is selected from the candidate insurance applicants to be used as the target insurance applicant. Therefore, the target insurance applicants with high insurance value and low safety accident risk can be accurately screened out.
Based on the above embodiment, the present invention further provides an applicant recommendation system based on security management measures and insurance letters, as shown in fig. 2, the system includes:
the matching module 01 is used for acquiring the insurance characteristics corresponding to a plurality of insurance applicants respectively, and matching each insurance applicants according to each insurance characteristic to obtain a plurality of comparison groups;
an obtaining module 02, configured to obtain security management measures and insurance values corresponding to the insurance applicants, where the insurance value corresponding to each insurance applicant is determined based on the risk of the security accident and the value of the insurance letter corresponding to the insurance applicant;
a comparison module 03, configured to perform intra-group comparison on the security management measures and the insurance value of the insurance applicant included in each comparison group, so as to obtain the reliability corresponding to each security management measure;
and the selecting module 04 is used for determining a target applicant from the applicant according to the reliability corresponding to each safety management measure.
Based on the above embodiment, the present invention further provides a terminal, and a functional block diagram of the terminal may be as shown in fig. 3. The terminal comprises a processor, a memory, a network interface and a display screen which are connected through a system bus. Wherein the processor of the terminal is configured to provide computing and control capabilities. The memory of the terminal comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operating system and the computer program to run on the non-volatile storage medium. The network interface of the terminal is used for connecting and communicating with an external terminal through a network. The computer program when executed by a processor implements a method for applicant recommendation based on security management measures and insurance coverage. The display screen of the terminal can be a liquid crystal display screen or an electronic ink display screen.
It will be appreciated by those skilled in the art that the block diagram of fig. 3 is only a block diagram of a part of the structure associated with the solution of the invention and does not constitute a limitation of the terminal to which the solution of the invention is applied, and that a specific terminal may comprise more or less components than those shown in the figure, or may combine some components, or have a different arrangement of components.
In one implementation, one or more programs are stored in a memory of the terminal and configured to be executed by one or more processors include instructions for performing an applicant recommendation method based on security management measures and insurance functions.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, databases, or other media used in embodiments provided herein may include non-volatile and/or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double Data Rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous Link DRAM (SLDRAM), rambus (Rambus) direct RAM (RDRAM), direct Rambus Dynamic RAM (DRDRAM), and Rambus Dynamic RAM (RDRAM), among others.
In summary, the invention discloses an applicant recommendation method based on safety management measures and insurance letters, which comprises the following steps: acquiring the insurance characteristics corresponding to a plurality of policyholders respectively, and matching each policyholder according to each insurance characteristic to obtain a plurality of comparison groups; acquiring safety management measures and insurance value corresponding to each insurance applicant, wherein the insurance value corresponding to each insurance applicant is determined based on the safety accident risk and the insurance letter value corresponding to the insurance applicant; comparing the safety management measures and the insurance value of the insurance applicant contained in each comparison group in a group to obtain the reliability corresponding to each safety management measure; and determining a target applicant from the applicant according to the reliability corresponding to each safety management measure. The problem of among the prior art recommend the policyholder to the guarantor through the insurance data of aassessment applicant, because insurance data can't indicate policyholder's risk, therefore recommend the policyholder that the insurance value is high easily, and the risk is also high, lead to the guarantor to undertake too high risk is solved.
It is to be understood that the invention is not limited to the examples described above, but that modifications and variations may be effected thereto by those of ordinary skill in the art in light of the foregoing description, and that all such modifications and variations are intended to be within the scope of the invention as defined by the appended claims.
Claims (10)
1. An applicant recommendation method based on security management measures and insurance letters, characterized in that the method comprises:
acquiring the insurance characteristics corresponding to a plurality of policemen respectively, and matching each policemen according to each insurance characteristic to obtain a plurality of control groups;
acquiring safety management measures and insurance value corresponding to each insurance applicant, wherein the insurance value corresponding to each insurance applicant is determined based on the safety accident risk and the insurance letter value corresponding to the insurance applicant;
comparing the safety management measures and the insurance value of the policyholder contained in each comparison group in a group to obtain the reliability corresponding to each safety management measure;
and determining a target applicant from the applicant according to the reliability corresponding to each safety management measure.
2. The insurance applicant recommendation method based on security management measures and insurance functions according to claim 1, wherein the obtaining of the insurance characteristics corresponding to each of a plurality of insurance appliers comprises:
acquiring insurance letter data and a preset feature extraction template corresponding to each policyholder;
and extracting the insurance letter features corresponding to the applicant from the insurance letter data according to the feature extraction template.
3. The method of claim 1 wherein the insurance features are vector data and the matching of each insurance applicant against each insurance feature to obtain a plurality of control groups comprises:
determining cosine vector similarity corresponding to each applicant according to the insurance letter characteristics of each applicant;
and determining each comparison group according to the cosine vector similarity corresponding to each applicant, wherein the cosine vector similarity corresponding to each applicant in each comparison group is the highest.
4. The insurance applicant recommendation method based on security management measures and insurance of claim 1, wherein the process of determining the value of the insurance application corresponding to each insurance applicant comprises:
acquiring a safety accident record corresponding to the applicant, and determining the safety accident risk corresponding to the applicant according to the safety accident record;
determining a penalty value according to the safety accident risk corresponding to the policyholder;
and acquiring the insurance value corresponding to the applicant, and determining the insurance value corresponding to the applicant according to the difference value of the insurance value and the penalty value.
5. The insurance applicant recommendation method based on security management measures and insurance policy according to claim 1, wherein the obtaining of the value of the insurance policy corresponding to the insurance applicant comprises:
inputting the insurance function data corresponding to the applicant into a pre-trained target extraction model to obtain an insurance function knowledge graph corresponding to the applicant;
and inputting the insurance letter knowledge graph into a pre-trained target prediction model to obtain the corresponding insurance value of the applicant.
6. The insurance applicant recommendation method based on security management measures and insurance policies according to claim 1, wherein each of the comparison groups includes two insurance applicants, and the group comparison of the security management measures and the insurance value of the insurance applicants included in each of the comparison groups is performed to obtain the reliability corresponding to each of the security management measures, respectively, includes:
determining the initial reliability of the two safety management measures corresponding to each contrast group according to the two insurable values corresponding to each contrast group;
determining a reward and punishment value corresponding to each control group according to the difference value of the two guarantee values corresponding to each control group;
performing reliability reward on the safety management measure with the highest initial reliability in the comparison group according to the reward and punishment value to obtain the reliability corresponding to the safety management measure;
and carrying out reliability punishment on the safety management measure with the lowest initial reliability in the comparison group according to the reward and punishment value to obtain the reliability corresponding to the safety management measure.
7. The method of claim 1, wherein the determining a target applicant from among the applicant's based on the reliability corresponding to each security management measure comprises:
and taking the applicant corresponding to the safety management measure with the highest reliability as the target applicant.
8. An applicant recommendation system based on security management measures and insurance letters, the system comprising:
the matching module is used for acquiring the insurance function characteristics corresponding to a plurality of insurance applicants respectively, and matching each insurance applicants according to each insurance function characteristic to obtain a plurality of comparison groups;
the acquiring module is used for acquiring the safety management measures and the insurance value corresponding to each insurance applicant, wherein the insurance value corresponding to each insurance applicant is determined based on the safety accident risk and the insurance letter value corresponding to the insurance applicant;
the comparison module is used for carrying out group-in comparison on the safety management measures and the insurance value of the insurance applicant contained in each comparison group to obtain the reliability corresponding to each safety management measure;
and the selection module is used for determining a target applicant from each applicant according to the reliability corresponding to each safety management measure.
9. A terminal is characterized in that the terminal comprises a storage device and more than one processor; the memory stores more than one program; the program comprising instructions for performing the insurance applicant recommendation method based on security management measures and insurance coverage of any one of claims 1-7; the processor is configured to execute the program.
10. A computer readable storage medium having stored thereon a plurality of instructions adapted to be loaded and executed by a processor to perform the steps of the method for applicant recommendation based on security management measures and insurance coverage of any of claims 1-7 above.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211181921.3A CN115760437A (en) | 2022-09-27 | 2022-09-27 | Insurance applicant recommendation method based on safety management measures and insurance letters |
| PCT/CN2022/136946 WO2024066036A1 (en) | 2022-09-27 | 2022-12-06 | Applicant recommendation method based on safety management measures and letters of guarantee |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211181921.3A CN115760437A (en) | 2022-09-27 | 2022-09-27 | Insurance applicant recommendation method based on safety management measures and insurance letters |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115760437A true CN115760437A (en) | 2023-03-07 |
Family
ID=85352005
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211181921.3A Pending CN115760437A (en) | 2022-09-27 | 2022-09-27 | Insurance applicant recommendation method based on safety management measures and insurance letters |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN115760437A (en) |
| WO (1) | WO2024066036A1 (en) |
Family Cites Families (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8095396B1 (en) * | 2008-03-27 | 2012-01-10 | Asterisk Financial Group, Inc. | Computer system for underwriting a personal guaranty liability by utilizing a risk apportionment system |
| CN109377388B (en) * | 2018-09-13 | 2023-08-18 | 深圳平安医疗健康科技服务有限公司 | Medical insurance application method, medical insurance application device, computer equipment and storage medium |
| CN109670974A (en) * | 2018-12-14 | 2019-04-23 | 中国平安人寿保险股份有限公司 | A kind of risk monitoring and control method and device, electric terminal |
| CN110647809A (en) * | 2019-08-15 | 2020-01-03 | 中国平安人寿保险股份有限公司 | AI (Artificial Intelligence) underwriting system and method based on image analysis and computer-readable storage medium |
| CN112182596B (en) * | 2020-03-24 | 2021-12-31 | 工保科技(浙江)有限公司 | Method for bidding and opening bid through electronic insurance policy system and data security method |
| CN113034267A (en) * | 2021-03-26 | 2021-06-25 | 广东中盈盛达数字科技有限公司 | Electronic letter protection system based on supply chain |
| CN114663064A (en) * | 2022-03-30 | 2022-06-24 | 深圳市融锋科技有限公司 | Comprehensive intelligent management system for bidding electronic insurance service |
-
2022
- 2022-09-27 CN CN202211181921.3A patent/CN115760437A/en active Pending
- 2022-12-06 WO PCT/CN2022/136946 patent/WO2024066036A1/en unknown
Also Published As
| Publication number | Publication date |
|---|---|
| WO2024066036A1 (en) | 2024-04-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2019196546A1 (en) | Method and apparatus for determining risk probability of service request event | |
| US11580459B2 (en) | Systems and methods for extracting specific data from documents using machine learning | |
| CN111369342A (en) | Loan approval method, device, equipment and storage medium based on machine learning | |
| US20210089667A1 (en) | System and method for implementing attribute classification for pii data | |
| CN111985207B (en) | Access control policy acquisition method and device and electronic equipment | |
| CN113590823A (en) | Contract approval method and device, storage medium and electronic equipment | |
| US11055200B2 (en) | Systems and methods for validating domain specific models | |
| CN112765980A (en) | Event argument role extraction method and device for alert condition record | |
| CA3169288A1 (en) | Knowledge graph based reasoning recommendation system and method | |
| CN116701734B (en) | Address text processing method and device and computer readable storage medium | |
| US20220036467A1 (en) | Machine learning system and method for quote generation | |
| CN112288279A (en) | Business risk assessment method and device based on natural language processing and linear regression | |
| CN112231556A (en) | User image drawing method, device, equipment and medium based on conversation scene | |
| CN115345727B (en) | Method and device for identifying fraudulent loan application | |
| CN115760437A (en) | Insurance applicant recommendation method based on safety management measures and insurance letters | |
| CN115063145A (en) | Transaction risk factor prediction method and device, electronic equipment and storage medium | |
| CN111401055B (en) | Method and apparatus for extracting context information from financial information | |
| CN112734307A (en) | Online real-time enterprise portrait generation method, device, equipment and storage medium | |
| US20230419401A1 (en) | Methods and systems for loan risk assessment in a smart city based on the internet of things | |
| CN115619573A (en) | Method, system, terminal and storage medium for extracting insurance information and predicting value | |
| CN112347738B (en) | Bidirectional encoder characterization quantity model optimization method and device based on referee document | |
| US20240135103A1 (en) | Training language models and preserving privacy | |
| CN117350547B (en) | Method, device, equipment and storage medium for determining risk processing scheme of order | |
| US11972338B2 (en) | Automated systems for machine learning model development, analysis, and refinement | |
| EP4283537A1 (en) | Automated systems for machine learning model development, analysis, and refinement |
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 |