CN106934480A - Insure grade analysis method, server and terminal - Google Patents
Insure grade analysis method, server and terminal Download PDFInfo
- Publication number
- CN106934480A CN106934480A CN201511020547.9A CN201511020547A CN106934480A CN 106934480 A CN106934480 A CN 106934480A CN 201511020547 A CN201511020547 A CN 201511020547A CN 106934480 A CN106934480 A CN 106934480A
- Authority
- CN
- China
- Prior art keywords
- group
- data
- insuring
- analysis
- default
- 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
- 238000004458 analytical method Methods 0.000 title claims abstract description 129
- 238000000034 method Methods 0.000 claims abstract description 26
- 238000006467 substitution reaction Methods 0.000 claims abstract description 7
- 238000012549 training Methods 0.000 claims description 21
- 238000012706 support-vector machine Methods 0.000 claims description 10
- 238000012360 testing method Methods 0.000 claims description 9
- 238000012545 processing Methods 0.000 claims description 2
- 241000894007 species Species 0.000 abstract description 24
- 239000000463 material Substances 0.000 abstract description 6
- 238000000605 extraction Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 238000004891 communication Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
Classifications
-
- 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
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- 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
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Finance (AREA)
- Development Economics (AREA)
- Physics & Mathematics (AREA)
- Accounting & Taxation (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Game Theory and Decision Science (AREA)
- Technology Law (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
Abstract
One kind is insured grade analysis method, including:The insurance data of the customer insured based on multiple groups of having insured, according to default model create-rule, generates an analysis forecast model;When receiving new group's grade analysis of insuring and instructing, the insurance data of group of insuring to be analyzed is obtained;According to presupposition analysis rule, the insurance data species of the group of insuring to be analyzed to obtaining is analyzed treatment, to obtain the analyze data of each default species data of the group of insuring to be analyzed;And by the analysis forecast model of the analyze data above-mentioned generation of substitution of the group of insuring to be analyzed, analyze the grade of the group of insuring of the analysis.The present invention also provides a kind of server and terminal suitable for the above method.The present invention can quickly analyze the level data of group of insuring, it is to avoid the spent a large amount of human and material resources of Traditional Man assessment.
Description
Technical field
The present invention relates to data assessment technical field, particularly one kind insure grade analysis method, server and
Terminal.
Background technology
At present, many insurance kinds for corporate customer are developed very much to insurance company, the group that accepts insurance is being treated
Client accept insurance when assessing, and typically insurance company staff analyzes group to be accepted insurance according to personal experience
The risk covered of body client, accuracy and promptness be difficult to ensure that, meanwhile, this assessment mode expends very big
Human and material resources.
The content of the invention
In view of the foregoing, it is necessary to one kind is provided and is insured grade analysis method, server and terminal, can be with
The quick level data for analyzing group of insuring, it is to avoid the spent a large amount of human and material resources of Traditional Man assessment.
One kind is insured grade analysis method, including:
The insurance data of the customer insured based on multiple groups of having insured, according to default model create-rule,
One analysis forecast model of generation;
When receiving new group's grade analysis of insuring and instructing, the insurance data of group of insuring to be analyzed is obtained;
According to presupposition analysis rule, the insurance data species of the group of insuring to be analyzed to obtaining is analyzed place
Reason, to obtain the analyze data of each default species data of the group of insuring to be analyzed;And
By in the analysis forecast model of the analyze data above-mentioned generation of substitution of the group of insuring to be analyzed, divide
Separate out the grade of the group of insuring of the analysis.
Preferably, the analysis forecast model is supporting vector machine model.
Preferably, the default model create-rule is:By the insurance data of the multiple group of having insured
In preset species data treatment be analyzed according to presupposition analysis rule, acquisition insured group each is default
The analyze data of species data;Grade classification is carried out according to default judgment rule to each group of having insured, will
Different grades of analyze data point is put into different files, and each extraction first is pre- under different files
If the analyze data of each preset kind of ratio is used as training data, to be analyzed the training of forecast model,
Again never with corresponding point of the analyze data of each preset kind of each remaining second preset ratio under file
Data are analysed as test data, to the Classification and Identification effect of assessment models;If the analysis forecast model of generation
Accuracy rate is less than or equal to default accuracy rate, then increase the acquisition quantity of the insurance data of group of having insured, and repeats
The generating process of above-mentioned model, until the analysis forecast model accuracy rate for generating is more than default accuracy rate.
Preferably, the presupposition analysis rule includes:Data class for serial number calculates average value,
Accounting calculating is carried out for the non-data class for serial number.
Preferably, the default judgment rule includes:Profit of insuring is more than first threshold and profit margin more than the
The group of insuring of two threshold values is for high-quality group;Profit of insuring is less than the 3rd threshold value and profit margin is less than the 4th threshold value
Group of insuring for poor group;Other situations are general group, carry out grade classification.
In view of the foregoing, there is a need to a kind of server suitable for the above method of offer, it can be quick
Analyze the level data of group of insuring, it is to avoid the spent a large amount of human and material resources of Traditional Man assessment.
A kind of server suitable for the above method, the server includes storage device and processor, wherein:
The memory cell, for storing a grade analysis system of insuring;
The processor, for calling and performs the grade analysis system of insuring, to perform following steps:
The insurance data of the customer insured based on multiple groups of having insured, according to default model create-rule,
One analysis forecast model of generation;
When receiving new group's grade analysis of insuring and instructing, the insurance data of group of insuring to be analyzed is obtained;
According to presupposition analysis rule, the insurance data species of the group of insuring to be analyzed to obtaining is analyzed place
Reason, to obtain the analyze data of each default species data of the group of insuring to be analyzed;And
By in the analysis forecast model of the analyze data above-mentioned generation of substitution of the group of insuring to be analyzed, divide
Separate out the grade of the group of insuring of the analysis.
Preferably, the default model create-rule is:By the insurance data of the multiple group of having insured
In preset species data treatment be analyzed according to presupposition analysis rule, acquisition insured group each is default
The analyze data of species data;Grade classification is carried out according to default judgment rule to each group of having insured, will
Different grades of analyze data point is put into different files, and each extraction first is pre- under different files
If the analyze data of each preset kind of ratio is used as training data, to be analyzed the training of forecast model,
Again never with corresponding point of the analyze data of each preset kind of each remaining second preset ratio under file
Data are analysed as test data, to the Classification and Identification effect of assessment models;If the analysis forecast model of generation
Accuracy rate is less than or equal to default accuracy rate, then increase the acquisition quantity of the insurance data of group of having insured, and repeats
The generating process of above-mentioned model, until the analysis forecast model accuracy rate for generating is more than default accuracy rate.
Preferably, wherein:
The presupposition analysis rule includes:Data class for serial number calculates average value, is for non-
The data class of serial number carries out accounting calculating;And
The default judgment rule includes:Profit of insuring is more than first threshold and profit margin is more than Second Threshold
Group insure for high-quality group;Profit of insuring is less than the group of insuring of the 4th threshold value less than the 3rd threshold value and profit margin
Body is poor group;Other situations are general group, carry out grade classification.
In view of the foregoing, there is a need to a kind of personal terminal suitable for the above method of offer, it can be fast
Speed analyzes the level data of group of insuring, it is to avoid the spent a large amount of human and material resources of Traditional Man assessment.
A kind of terminal suitable for the above method, the terminal includes storage device and processor, wherein:
The memory cell, for the grade analysis system of insuring that is stored with;
The processor, for calling and performs the grade analysis system of insuring, to perform following steps:
The memory cell, for storing a grade analysis system of insuring;
The processor, for calling and performs the grade analysis system of insuring, to perform following steps:
The insurance data of the customer insured based on multiple groups of having insured, according to default model create-rule,
One analysis forecast model of generation;
When receiving new group's grade analysis of insuring and instructing, the insurance data of group of insuring to be analyzed is obtained;
According to presupposition analysis rule, the insurance data species of the group of insuring to be analyzed to obtaining is analyzed place
Reason, to obtain the analyze data of each default species data of the group of insuring to be analyzed;And
By in the analysis forecast model of the analyze data above-mentioned generation of substitution of the group of insuring to be analyzed, divide
Separate out the grade of the group of insuring of the analysis.
Preferably, the default model create-rule is:By the insurance data of the multiple group of having insured
In preset species data treatment be analyzed according to presupposition analysis rule, acquisition insured group each is default
The analyze data of species data;Grade classification is carried out according to default judgment rule to each group of having insured, will
Different grades of analyze data point is put into different files, and each extraction first is pre- under different files
If the analyze data of each preset kind of ratio is used as training data, to be analyzed the training of forecast model,
Again never with corresponding point of the analyze data of each preset kind of each remaining second preset ratio under file
Data are analysed as test data, to the Classification and Identification effect of assessment models;If the analysis forecast model of generation
Accuracy rate is less than or equal to default accuracy rate, then increase the acquisition quantity of the insurance data of group of having insured, and repeats
The generating process of above-mentioned model, until the analysis forecast model accuracy rate for generating is more than default accuracy rate.
Grade analysis method of insuring of the present invention, server and terminal, by the assessment for setting up analysis prediction
Model, the quick level data for analyzing group of insuring, it is to avoid the spent a large amount of manpowers of Traditional Man assessment,
Material resources.
Brief description of the drawings
Fig. 1 is the hardware environment figure of grade analysis system first embodiment of insuring of the invention.
Fig. 2 is the hardware environment figure of grade analysis system second embodiment of insuring of the invention.
Fig. 3 is the functional block diagram of grade analysis system preferred embodiment of insuring of the invention.
Fig. 4 is the method implementing procedure figure of grade analysis method preferred embodiment of insuring of the invention.
Specific embodiment
Refering to the hardware environment figure for shown in Fig. 1, being grade analysis system first embodiment of insuring of the invention.
During grade analysis system 2 of being insured described in the present embodiment can be installed and run on a server 1.Institute
Stating server 1 can be connected by communication module (not shown) with least one communication of personal terminal 3, institute
It can be the equipment such as PC, smart mobile phone, panel computer to state personal terminal 3.The personal terminal 3
Including input equipment 30 and display device 31.
The server 1 can include processor and storage device (not shown).The processor is
The arithmetic core (Core Unit) and control core (Control Unit) of server 1, calculate for explaining
Machine instructs and processes the data in computer software.The storage device can be that one or more is non-easily
The property lost memory cell, such as ROM, EPROM or Flash Memory (flash memory cell).It is described to deposit
Storage equipment can be with built-in or be external in server 1.
In the present embodiment, the grade analysis system 2 of insuring can be a kind of computer software, and it includes meter
The executable program instruction code of calculation machine, the program instruction code can be stored in the storage device,
Under the execution of the processor, following function is realized:Obtained from the database 4 being connected with the server 1
The insurance data of the customer insured of multiple groups of having insured is taken, based on the insurance data of the group of having insured,
According to default model create-rule, an analysis forecast model is generated.
The default model create-rule is:To be preset in the insurance data of the multiple group of having insured and planted
Class data, for example, age, sex, occupation, position, owned enterprise's property, including government bodies' unit,
State-owned enterprise, private enterprise, foreign enterprise etc., treatment is analyzed according to presupposition analysis rule, obtains each of group of having insured
The analyze data of individual default species data.Grade is carried out to each group of having insured according to default judgment rule to draw
Point, different grades of analyze data point is put into different files, respectively extracted under different files
First preset ratio, for example, 70%, each preset kind analyze data as training data, to carry out
The training of forecast model is analyzed, then never with each remaining second preset ratio under file, for example, 30%,
Each preset kind the corresponding analyze data of analyze data as test data, to assessment models point
Class recognition effect;If the analysis forecast model accuracy rate of generation is less than or equal to default accuracy rate, for example, 99%,
Then increase the acquisition quantity of the insurance data of group of having insured, repeat the generating process of above-mentioned model, Zhi Daosheng
Into analysis forecast model accuracy rate be more than default accuracy rate, for example, 99%.
The presupposition analysis rule is analyzed treatment to be included, for example, for the data class of serial number,
For example, " age " calculates average value;For it is non-be the data class of serial number, for example, " sex ", " duty
Industry ", " position ", " owned enterprise's property " provide corresponding operation result by default treatment logic.Default treatment
Logic is:Accounting calculating is carried out to the non-data class for serial number, for example, male accounts in group of insuring
Than, management duty accounting, state-owned enterprise's work accounting etc., research staff's accounting etc..Therefore, resulting default kind
The analyze data of class data includes, for example, the accounting of the average value at age, sex, the accounting of position, institute
Belong to accounting of enterprise nature etc..
The default judgment rule includes, for example, insuring, profit is more than first threshold and profit margin is more than second
The group of insuring of threshold value is for high-quality group;Profit of insuring is less than the 3rd threshold value and profit margin is less than the 4th threshold value
Group insure for poor group;Other situations are general group, carry out grade classification, obtain grade including excellent
Matter group body, poor group and general group.
In the present embodiment, it is described analysis forecast model be SVMs (Support Vector Machine,
SVM) model.The SVMs is built upon the VC dimensions of Statistical Learning Theory
(Vapnik-Chervonenkis Dimension) theoretical and Structural risk minization basis on, according to having
The sample information of limit model complexity (i.e. to the study precision of specific training sample) and learning ability is (i.e.
Without error recognize arbitrary sample ability) between seek optimal compromise, to obtain best Generalization Ability.
In new group's grade analysis of insuring that the user for receiving personal terminal 3 is sent by its input equipment 30
During instruction, the grade analysis system 2 of insuring obtains the insurance data of group of insuring to be analyzed, according to upper
State presupposition analysis rule, to obtain group of insuring to be analyzed insurance data species, such as example, the age,
Sex, occupation, position, owned enterprise's property, are analyzed treatment, to obtain the group of insuring to be analyzed
Each default species data analyze data, such as the average value at age, the accounting of sex, the accounting of position,
Accounting of owned enterprise's property etc., above-mentioned generation is substituted into by the analyze data of the group of insuring to be analyzed
In analysis forecast model, the grade of the group of insuring of the analysis is analyzed.The group of insuring for analyzing
Grade can send personal terminal 3 to, and be displayed on its display device 31.
In other embodiments of the invention, as shown in Fig. 2 described grade analysis system 2 of insuring also may be used
To install and run on personal terminal 3, the program code of the grade analysis system 2 of insuring can be stored
In the storage device (not shown) of the personal terminal 3, and the processor in personal terminal 3 execution
Under, realize the function of foregoing description.
Refering to the functional block diagram for shown in Fig. 3, being grade analysis system preferred embodiment of insuring of the invention.
The program code of the grade analysis system 2 of insuring can be divided into multiple according to its different function
Functional module.In present pre-ferred embodiments, the grade analysis system 2 of insuring can be set up including model
Module 20, acquisition module 21, pretreatment module 22 and grade analysis module 23.
The model building module 20 is used for the insurance data of the customer insured based on multiple groups of having insured, and presses
According to default model create-rule, an analysis forecast model is generated.
The default model create-rule is:To be preset in the insurance data of the multiple group of having insured and planted
Class data, for example, age, sex, occupation, position, owned enterprise's property, including government bodies' unit,
State-owned enterprise, private enterprise, foreign enterprise etc., treatment is analyzed according to presupposition analysis rule, obtains each of group of having insured
The analyze data of individual default species data.Grade is carried out to each group of having insured according to default judgment rule to draw
Point, different grades of analyze data point is put into different files, respectively extracted under different files
First preset ratio, for example, 70%, each preset kind analyze data as training data, to carry out
The training of forecast model is analyzed, then never with each remaining second preset ratio under file, for example, 30%,
Each preset kind the corresponding analyze data of analyze data as test data, to assessment models point
Class recognition effect;If the analysis forecast model accuracy rate of generation is less than or equal to default accuracy rate, for example, 99%,
Then increase the acquisition quantity of the insurance data of group of having insured, repeat the generating process of above-mentioned model, Zhi Daosheng
Into analysis forecast model accuracy rate be more than default accuracy rate, for example, 99%.
The presupposition analysis rule includes, for example, for the data class of serial number, for example, " age "
Calculate average value;For it is non-be the data class of serial number, for example, " sex ", " occupation ", " position ",
" owned enterprise's property " provides corresponding operation result by default treatment logic.Presetting treatment logic is:It is to non-
The data class of serial number carries out accounting calculating, for example, male's accounting in group of insuring, management duty accounting,
State-owned enterprise's work accounting etc., research staff's accounting etc..Therefore, the analyze data of resulting default species data
Including for example, the accounting of the average value at age, sex, the accounting of position, the accounting of owned enterprise's property
Deng.
The default judgment rule includes, for example, insuring, profit is more than first threshold and profit margin is more than second
The group of insuring of threshold value is for high-quality group;Profit of insuring is less than the 3rd threshold value and profit margin is less than the 4th threshold value
Group insure for poor group;Other situations are general group, carry out grade classification, obtain grade including excellent
Matter group body, poor group and general group.
In the present embodiment, it is described analysis forecast model be SVMs (Support Vector Machine,
SVM) model.The SVMs is built upon the VC dimensions of Statistical Learning Theory
(Vapnik-Chervonenkis Dimension) theoretical and Structural risk minization basis on, according to having
The sample information of limit model complexity (i.e. to the study precision of specific training sample) and learning ability is (i.e.
Without error recognize arbitrary sample ability) between seek optimal compromise, to obtain best Generalization Ability.
The acquisition module 21 is used for what is sent by its input equipment 30 in the user for receiving personal terminal 3
When new group's grade analysis of insuring are instructed, the insurance data of group of insuring to be analyzed is obtained.
The pretreatment module 22 is used for according to above-mentioned presupposition analysis rule, to the group of insuring to be analyzed for obtaining
Insurance data species, such as example, age, sex, occupation, position, owned enterprise's property, are divided
Analysis is processed, to obtain the analyze data of each default species data of the group of insuring to be analyzed, such as the age
Average value, the accounting of sex, the accounting of position, accounting of owned enterprise's property etc..
The grade analysis module 23 is used to for the analyze data of the group of insuring to be analyzed to substitute into above-mentioned life
Into analysis forecast model in, analyze the grade of the group of insuring of the analysis.The group of insuring for analyzing
The grade of body can send personal terminal 3 to, and be displayed on its display device 31.
Refering to the method implementing procedure figure for shown in Fig. 4, being grade analysis method preferred embodiment of insuring of the invention.
Grade analysis method of being insured described in the present embodiment is not limited to step shown in flow chart, in addition institute in flow chart
In showing step, some steps can be omitted, the order between step can change.
Step S10, model building module 20 is based on the insurance data of the customer insured of multiple groups of having insured,
According to default model create-rule, an analysis forecast model is generated.
The default model create-rule is:To be preset in the insurance data of the multiple group of having insured and planted
Class data, for example, age, sex, occupation, position, owned enterprise's property, including government bodies' unit,
State-owned enterprise, private enterprise, foreign enterprise etc., treatment is analyzed according to presupposition analysis rule, obtains each of group of having insured
The analyze data of individual default species data.Grade is carried out to each group of having insured according to default judgment rule to draw
Point, different grades of analyze data point is put into different files, respectively extracted under different files
First preset ratio, for example, 70%, each preset kind analyze data as training data, to carry out
The training of forecast model is analyzed, then never with each remaining second preset ratio under file, for example, 30%,
Each preset kind the corresponding analyze data of analyze data as test data, to assessment models point
Class recognition effect;If the analysis forecast model accuracy rate of generation is less than or equal to default accuracy rate, for example, 99%,
Then increase the acquisition quantity of the insurance data of group of having insured, repeat the generating process of above-mentioned model, Zhi Daosheng
Into analysis forecast model accuracy rate be more than default accuracy rate, for example, 99%.
The presupposition analysis rule is analyzed treatment to be included, for example, for the data class of serial number,
For example, " age " calculates average value;For it is non-be the data class of serial number, for example, " sex ", " duty
Industry ", " position ", " owned enterprise's property " provide corresponding operation result by default treatment logic.Default treatment
Logic is:Accounting calculating is carried out to the non-data class for serial number, for example, male accounts in group of insuring
Than, management duty accounting, state-owned enterprise's work accounting etc., research staff's accounting etc..Therefore, resulting default kind
The analyze data of class data includes, for example, the accounting of the average value at age, sex, the accounting of position, institute
Belong to accounting of enterprise nature etc..
The default judgment rule includes, for example, insuring, profit is more than first threshold and profit margin is more than second
The group of insuring of threshold value is for high-quality group;Profit of insuring is less than the 3rd threshold value and profit margin is less than the 4th threshold value
Group insure for poor group;Other situations are general group, carry out grade classification, obtain grade including excellent
Matter group body, poor group and general group.
In the present embodiment, it is described analysis forecast model be SVMs (Support Vector Machine,
SVM) model.The SVMs is built upon the VC dimensions of Statistical Learning Theory
(Vapnik-Chervonenkis Dimension) theoretical and Structural risk minization basis on, according to having
The sample information of limit model complexity (i.e. to the study precision of specific training sample) and learning ability is (i.e.
Without error recognize arbitrary sample ability) between seek optimal compromise, to obtain best Generalization Ability.
Step S11, in the new group of insuring that the user for receiving personal terminal 3 is sent by its input equipment 30
When body grade analysis are instructed, acquisition module 21 obtains the insurance data of group of insuring to be analyzed.
Step S12, pretreatment module 22 is regular according to above-mentioned presupposition analysis, to the group of insuring to be analyzed for obtaining
The insurance data species of body, such as example, age, sex, occupation, position, owned enterprise's property, are carried out
Analyzing and processing, to obtain the analyze data of each default species data of the group of insuring to be analyzed, such as age
Average value, the accounting of sex, the accounting of position, the accounting of owned enterprise's property etc..
Step S13, grade analysis module 23 substitutes into the analyze data of the group of insuring to be analyzed above-mentioned
In the analysis forecast model of generation, the grade of the group of insuring of the analysis is analyzed.It is described analyze insure
The grade of group can send personal terminal 3 to, and be displayed on its display device 31.
It should be noted last that, the above embodiments are merely illustrative of the technical solutions of the present invention and it is unrestricted,
Although being described in detail to the present invention with reference to preferred embodiment, one of ordinary skill in the art should manage
Solution, can modify or equivalent, without deviating from technical solution of the present invention to technical scheme
Spirit and scope.
Claims (10)
1. one kind is insured grade analysis method, it is characterised in that the method includes:
The insurance data of the customer insured based on multiple groups of having insured, according to default model create-rule,
One analysis forecast model of generation;
When receiving new group's grade analysis of insuring and instructing, the insurance data of group of insuring to be analyzed is obtained;
According to presupposition analysis rule, the insurance data species of the group of insuring to be analyzed to obtaining is analyzed place
Reason, to obtain the analyze data of each default species data of the group of insuring to be analyzed;And
By in the analysis forecast model of the analyze data above-mentioned generation of substitution of the group of insuring to be analyzed, divide
Separate out the grade of the group of insuring of the analysis.
2. the method for claim 1, it is characterised in that the analysis forecast model is SVMs mould
Type.
3. method as claimed in claim 2, it is characterised in that the default model create-rule is:By institute
Species data being preset in the insurance data for stating multiple groups of having insured, place is analyzed according to presupposition analysis rule
Reason, obtains the analyze data of each default species data of group of having insured;To each insured group according to
Default judgment rule carries out grade classification, and different grades of analyze data point is put into different files,
The analyze data of each each preset kind for extracting the first preset ratio is used as training number under different files
According to be analyzed the training of forecast model, then never with each remaining second preset ratio under file
The corresponding analyze data of analyze data of each preset kind as test data, to the classification of assessment models
Recognition effect;If the analysis forecast model accuracy rate of generation is less than or equal to default accuracy rate, increases and insured
The acquisition quantity of the insurance data of group, repeats the generating process of above-mentioned model, until the analysis prediction for generating
Model accuracy rate is more than default accuracy rate.
4. method as claimed in claim 3, it is characterised in that the presupposition analysis rule includes:For continuous
The data class of numerical value calculates average value, and accounting calculating is carried out for the non-data class for serial number.
5. method as claimed in claim 3, it is characterised in that the default judgment rule includes:Insure profit
Group of insuring more than first threshold and profit margin more than Second Threshold is for high-quality group;Profit of insuring is less than the
Three threshold values and profit margin less than the 4th threshold value group of insuring for poor group;Other situations are general group,
Carry out grade classification.
6. a kind of server suitable for claim 1 to 5 any one methods described, it is characterised in that the service
Device includes storage device and processor, wherein:
The memory cell, for storing a grade analysis system of insuring;
The processor, for calling and performs the grade analysis system of insuring, to perform following steps:
The insurance data of the customer insured based on multiple groups of having insured, according to default model create-rule,
One analysis forecast model of generation;
When receiving new group's grade analysis of insuring and instructing, the insurance data of group of insuring to be analyzed is obtained;
According to presupposition analysis rule, the insurance data species of the group of insuring to be analyzed to obtaining is analyzed place
Reason, to obtain the analyze data of each default species data of the group of insuring to be analyzed;And
By in the analysis forecast model of the analyze data above-mentioned generation of substitution of the group of insuring to be analyzed, divide
Separate out the grade of the group of insuring of the analysis.
7. server as claimed in claim 6, it is characterised in that wherein, the default model create-rule
For:Carried out species data are preset in the insurance data of the multiple group of having insured according to presupposition analysis rule
Analyzing and processing, obtains the analyze data of each default species data of group of having insured;To each group of having insured
Body carries out grade classification according to default judgment rule, and different grades of analyze data point is put into different files
Underedge, the analyze data conduct of each each preset kind for extracting the first preset ratio under different files
Training data is to be analyzed the training of forecast model then never default with file each remaining second
The corresponding analyze data of analyze data of each preset kind of ratio as test data, to assessment models
Classification and Identification effect;If the analysis forecast model accuracy rate of generation is less than or equal to default accuracy rate, increase
Insured group insurance data acquisition quantity, repeat the generating process of above-mentioned model, until generation point
Analysis forecast model accuracy rate is more than default accuracy rate.
8. server as claimed in claim 6, it is characterised in that wherein:
The presupposition analysis rule includes:Data class for serial number calculates average value, is for non-
The data class of serial number carries out accounting calculating;And
The default judgment rule includes:Profit of insuring is more than first threshold and profit margin is more than Second Threshold
Group insure for high-quality group;Profit of insuring is less than the group of insuring of the 4th threshold value less than the 3rd threshold value and profit margin
Body is poor group;Other situations are general group, carry out grade classification.
9. a kind of terminal suitable for claim 1 to 5 any one methods described, it is characterised in that the terminal bag
Storage device and processor are included, wherein:
The memory cell, for the grade analysis system of insuring that is stored with;
The processor, for calling and performs the grade analysis system of insuring, to perform following steps:
The memory cell, for storing a grade analysis system of insuring;
The processor, for calling and performs the grade analysis system of insuring, to perform following steps:
The insurance data of the customer insured based on multiple groups of having insured, according to default model create-rule,
One analysis forecast model of generation;
When receiving new group's grade analysis of insuring and instructing, the insurance data of group of insuring to be analyzed is obtained;
According to presupposition analysis rule, the insurance data species of the group of insuring to be analyzed to obtaining is analyzed place
Reason, to obtain the analyze data of each default species data of the group of insuring to be analyzed;And
By in the analysis forecast model of the analyze data above-mentioned generation of substitution of the group of insuring to be analyzed, divide
Separate out the grade of the group of insuring of the analysis.
10. terminal as claimed in claim 9, it is characterised in that the default model create-rule is:By institute
Species data being preset in the insurance data for stating multiple groups of having insured, place is analyzed according to presupposition analysis rule
Reason, obtains the analyze data of each default species data of group of having insured;To each insured group according to
Default judgment rule carries out grade classification, and different grades of analyze data point is put into different files,
The analyze data of each each preset kind for extracting the first preset ratio is used as training number under different files
According to be analyzed the training of forecast model, then never with each remaining second preset ratio under file
The corresponding analyze data of analyze data of each preset kind as test data, to the classification of assessment models
Recognition effect;If the analysis forecast model accuracy rate of generation is less than or equal to default accuracy rate, increases and insured
The acquisition quantity of the insurance data of group, repeats the generating process of above-mentioned model, until the analysis prediction for generating
Model accuracy rate is more than default accuracy rate.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201511020547.9A CN106934480A (en) | 2015-12-29 | 2015-12-29 | Insure grade analysis method, server and terminal |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201511020547.9A CN106934480A (en) | 2015-12-29 | 2015-12-29 | Insure grade analysis method, server and terminal |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106934480A true CN106934480A (en) | 2017-07-07 |
Family
ID=59441508
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201511020547.9A Pending CN106934480A (en) | 2015-12-29 | 2015-12-29 | Insure grade analysis method, server and terminal |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106934480A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108961069A (en) * | 2018-05-24 | 2018-12-07 | 中国平安人寿保险股份有限公司 | Electronic device, personal insurance application data processing method and computer storage medium |
CN109712000A (en) * | 2018-11-27 | 2019-05-03 | 平安科技(深圳)有限公司 | Repeat insure method for limiting, device, equipment and readable storage medium storing program for executing |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103678659A (en) * | 2013-12-24 | 2014-03-26 | 焦点科技股份有限公司 | E-commerce website cheat user identification method and system based on random forest algorithm |
CN104616198A (en) * | 2015-02-12 | 2015-05-13 | 哈尔滨工业大学 | P2P (peer-to-peer) network lending risk prediction system based on text analysis |
-
2015
- 2015-12-29 CN CN201511020547.9A patent/CN106934480A/en active Pending
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103678659A (en) * | 2013-12-24 | 2014-03-26 | 焦点科技股份有限公司 | E-commerce website cheat user identification method and system based on random forest algorithm |
CN104616198A (en) * | 2015-02-12 | 2015-05-13 | 哈尔滨工业大学 | P2P (peer-to-peer) network lending risk prediction system based on text analysis |
Non-Patent Citations (5)
Title |
---|
刘军丽,等: "基于决策树的个人住房贷款信用风险评估模型", 《计算机工程》 * |
刘晓葳: "基于数据挖掘的保险客户风险-贡献评级管理", 《保险研究》 * |
彭岩,等: "基于机器学习的风险预测方法研究", 《计算机科学》 * |
方匡南,等: "基于随机森林的保险客户利润贡献度研究", 《数理统计与管理》 * |
隋学深,等: "基于支持向量机的贷款风险等级", 《审计研究》 * |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108961069A (en) * | 2018-05-24 | 2018-12-07 | 中国平安人寿保险股份有限公司 | Electronic device, personal insurance application data processing method and computer storage medium |
CN109712000A (en) * | 2018-11-27 | 2019-05-03 | 平安科技(深圳)有限公司 | Repeat insure method for limiting, device, equipment and readable storage medium storing program for executing |
WO2020107591A1 (en) * | 2018-11-27 | 2020-06-04 | 平安科技(深圳)有限公司 | Double insurance limiting method, apparatus, device, and readable storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110443657B (en) | Client flow data processing method and device, electronic equipment and readable medium | |
CN109284372B (en) | User operation behavior analysis method, electronic device and computer readable storage medium | |
CN112528025A (en) | Text clustering method, device and equipment based on density and storage medium | |
CN107844836A (en) | A kind of system and learning method based on machine learning | |
CN112328909B (en) | Information recommendation method and device, computer equipment and medium | |
CN111161804B (en) | Query method and system for species genomics database | |
CN111931809A (en) | Data processing method and device, storage medium and electronic equipment | |
CN114090601A (en) | Data screening method, device, equipment and storage medium | |
CN106934480A (en) | Insure grade analysis method, server and terminal | |
CN112839185B (en) | Method, apparatus, device and medium for processing image | |
CN112560480A (en) | Task community discovery method, device, equipment and storage medium | |
CN115757075A (en) | Task abnormity detection method and device, computer equipment and storage medium | |
CN114863994A (en) | Pollution assessment method, device, electronic equipment and storage medium | |
CN114565470A (en) | Financial product recommendation method based on artificial intelligence and related equipment thereof | |
CN112069807A (en) | Text data theme extraction method and device, computer equipment and storage medium | |
Zhang et al. | On the cost of interactions in interactive visual machine learning | |
CN113190154A (en) | Model training method, entry classification method, device, apparatus, storage medium, and program | |
CN111767735A (en) | Method, apparatus and computer readable storage medium for executing task | |
CN112906723A (en) | Feature selection method and device | |
CN116340864B (en) | Model drift detection method, device, equipment and storage medium thereof | |
CN110796381B (en) | Modeling method and device for wind control model, terminal equipment and medium | |
CN112396445B (en) | Method and device for identifying user identity information | |
CN115965817A (en) | Training method and device of image classification model and electronic equipment | |
CN105095531A (en) | Method and device for realizing device analysis | |
CN115761445A (en) | Method, device, equipment and medium for training chromosome analysis model |
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 | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170707 |