CN108197742A - Continuation of insurance behavior prediction method, system and the computer readable storage medium of user - Google Patents
Continuation of insurance behavior prediction method, system and the computer readable storage medium of user Download PDFInfo
- Publication number
- CN108197742A CN108197742A CN201711497554.7A CN201711497554A CN108197742A CN 108197742 A CN108197742 A CN 108197742A CN 201711497554 A CN201711497554 A CN 201711497554A CN 108197742 A CN108197742 A CN 108197742A
- Authority
- CN
- China
- Prior art keywords
- insurance
- user
- continuation
- condition
- decision
- 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
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)
- Strategic Management (AREA)
- Economics (AREA)
- Theoretical Computer Science (AREA)
- Finance (AREA)
- Human Resources & Organizations (AREA)
- Accounting & Taxation (AREA)
- Marketing (AREA)
- Development Economics (AREA)
- General Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Tourism & Hospitality (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Entrepreneurship & Innovation (AREA)
- Game Theory and Decision Science (AREA)
- Technology Law (AREA)
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses the continuation of insurance behavior prediction method of user a kind of, including step:It obtains the corresponding user data of user for having purchased insurance and the corresponding benchmark continuation of insurance probability value of user is calculated according to preset benchmark algorithm;The predicted condition that multiple continuation of insurance behaviors to user in different levels cause good effect is obtained, wherein, each level correspondence includes at least one predicted condition;According to user data and the corresponding benchmark continuation of insurance probability value of the user, by decision-tree model, according to the sequence of the level, using the predicted condition as decision node, the decision node of selection corresponding level is classified successively, to obtain the corresponding continuation of insurance probability value of the user.The present invention also provides the continuation of insurance behavior prediction systems and computer readable storage medium of a kind of user.The present invention solves the problems, such as that salesman predicts that user's continuation of insurance behavior accuracy is not high by personal experience, convenient for analyzing the continuation of insurance behavior of a large number of users.
Description
Technical field
The present invention relates to the continuation of insurance behavior prediction method of data analysis technique field more particularly to a kind of user, system and
Computer readable storage medium.
Background technology
With the improvement of people's living standards, people also gradually put forward the guarantee consciousness of personal safety and life and health
Height, and one of measure for wherein improving safety and health guarantee is purchase insurance.For insurance company, how to maintain
Existing user ensures certain continuation of insurance number, is one of the basic means that insurance company keeps profit, among these with regard to needing to make
It is gone to predict which user can continue insurance with means of numerical analysis.Before this, the method predicted for the continuation of insurance behavior of user
Salesman is mainly based upon according to limited information and personal experience to judge whether user can continue insurance, but for a large number of users and
Speech, this mode accuracy artificially judged be not high and unstable.
Invention content
It is a primary object of the present invention to provide the continuation of insurance behavior prediction method of user a kind of, system and computer-readable deposit
Storage media, it is intended to solve the problems, such as that the accuracy of user's continuation of insurance behavior prediction is not high, improve the stability of prediction.
To achieve the above object, the continuation of insurance behavior prediction method of user, includes the following steps:
Acquisition has purchased the corresponding user data of user of insurance and according to the corresponding preset base of insurance business to be predicted
Quasi- algorithm calculates the corresponding benchmark continuation of insurance probability value of the user;
The predicted condition that multiple continuation of insurance behaviors to the user in different levels cause good effect is obtained,
In, each level correspondence includes at least one predicted condition;
According to the user data and the user corresponding benchmark continuation of insurance probability value, by decision-tree model, according to
The sequence of the level using the predicted condition as decision node, selects the decision node of corresponding level to classify successively,
To obtain the continuation of insurance probability value of the corresponding insurance business to be predicted of the user.
To achieve the above object, the present invention also provides the continuation of insurance behavior prediction system of user a kind of, including:
Acquisition module, for obtaining the corresponding user data of user for having purchased insurance;
Computing module, for calculating the corresponding benchmark continuation of insurance probability value of the user according to preset benchmark algorithm;
The acquisition module is additionally operable to multiple continuation of insurance behavior to the user of the acquisition in different levels and causes actively
The predicted condition of effect, wherein, each level correspondence includes at least one predicted condition;
Selecting module, for being continued insurance probability value according to the user data and the corresponding benchmark of the user, by certainly
Plan tree-model according to the sequence of the level, using the predicted condition as decision node, selects the decision of corresponding level successively
Node is classified, to obtain the continuation of insurance probability value of the corresponding insurance business to be predicted of the user.
To achieve the above object, the present invention also provides the continuation of insurance behavior prediction system of user a kind of, including communication module, place
The computer program managed device, memory and storage on a memory and can run on a processor;Described in the processor performs
The step of continuation of insurance behavior prediction method of user as described above is realized during computer program.
To achieve the above object, the present invention also provides a kind of computer readable storage medium, the computer-readable storages
Computer program is stored on medium, the computer program realizes the continuation of insurance row of user as described above when being executed by processor
The step of for Forecasting Methodology.
The present invention is by obtaining the corresponding user data of user for having purchased insurance and pair according to insurance business to be predicted
The preset benchmark algorithm answered calculates the corresponding benchmark continuation of insurance probability value of the user;It obtains multiple to institute in different levels
The continuation of insurance behavior for stating user causes the predicted condition of good effect, wherein, each level correspondence includes at least one described
Predicted condition;According to the user data and the user corresponding benchmark continuation of insurance probability value, by decision-tree model, according to
The sequence of the level using the predicted condition as decision node, selects the decision node of corresponding level to classify successively,
To obtain the corresponding continuation of insurance probability value of the user.The more of good effect are caused since the present invention is provided with behavior of continuing insurance to user
A predicted condition, and decision-tree model has been used to classify, then obtain the continuation of insurance probability value of user so that continue insurance in user
During behavior prediction, a large amount of user data is incorporated, realizes the continuation of insurance for manual operation being replaced to carry out user with machine
The prediction of behavior, forecasting accuracy is high, and stability is good.
Description of the drawings
Fig. 1 is the schematic network structure of the continuation of insurance behavior prediction system of the user in the present invention;
Fig. 2 is the flow diagram of the first embodiment of the continuation of insurance behavior prediction method of user of the present invention;
Refinement flow signals of the Fig. 3 for step S30 in the second embodiment of the continuation of insurance behavior prediction method of user of the present invention
Figure;
Fig. 4 is the flow diagram of the 3rd embodiment of the continuation of insurance behavior prediction method of user of the present invention;
Refinement flow signals of the Fig. 5 for step S10 in the fourth embodiment of the continuation of insurance behavior prediction method of user of the present invention
Figure;
Refinement flow signals of the Fig. 6 for step S30 in the 5th embodiment of the continuation of insurance behavior prediction method of user of the present invention
Figure;
Fig. 7 is the high-level schematic functional block diagram of the continuation of insurance behavior prediction system of user of the present invention.
The embodiments will be further described with reference to the accompanying drawings for the realization, the function and the advantages of the object of the present invention.
Specific embodiment
It should be appreciated that specific embodiment described herein is not intended to limit the present invention only to explain the present invention.
In subsequent description, using for representing that the suffix of such as " module ", " component " or " unit " of element is only
Be conducive to the explanation of the present invention, itself there is no a specific meaning.Therefore, " module ", " component " or " unit " can mix
Ground uses.
Fig. 1 is please referred to, Fig. 1 is the structure diagram of the continuation of insurance behavior prediction system of user in each embodiment of the present invention.
The continuation of insurance behavior prediction system of the user can be individually used for the plateform system of behavior prediction analysis or as number
According to the total server integrated is collected, the continuation of insurance behavior prediction system of the user can dispose beyond the clouds, can also be deployed in this
Ground.
The continuation of insurance behavior prediction system of the user includes the components such as communication module 10, memory 20 and processor 30.Its
In, the processor 30 is connect respectively with the memory 20 and the communication module 10, and meter is stored on the memory 20
Calculation machine program, the computer program are performed simultaneously by processor 30.
Communication module 10 can be connected by network and external equipment, such as electric terminal.Communication module 10 can receive outer
The request that portion's communication apparatus is sent out, also transmittable event, instruction and information to the external equipment and/or other servers.Institute
It can be the electronic equipments such as scanner, keyboard, mobile phone, computer and picture pick-up device to state external communications equipment.
Memory 20, available for storage software program and various data.Memory 20 can mainly include storing program area
And storage data field, wherein, storing program area can storage program area etc.;Storage data field can store the continuation of insurance row according to user
The data created or information etc., such as the asset data of user are used by forecasting system.In addition, memory 20 can include
High-speed random access memory can also include nonvolatile memory, for example, at least disk memory, a flash memories
Part or other volatile solid-state parts.
Processor 30 is the control centre of the continuation of insurance behavior prediction system of user, whole using various interfaces and connection
The various pieces of the continuation of insurance behavior prediction system of a user, by running or performing the software program being stored in memory 20
And/or module and calling are stored in the data in memory 20, perform the various functions of the continuation of insurance behavior prediction system of user
With processing data.Processor 30 may include one or more processing units;Preferably, processor 30 can integrate application processor and
Modem processor, wherein, the main processing operation system of application processor, user interface and application program etc., modulation /demodulation
Processor mainly handles wireless communication.It is understood that above-mentioned modem processor can not also be integrated into processor 30
In.
Although Fig. 1 is not shown, the continuation of insurance behavior prediction system of above-mentioned user can also include circuit control module, be used for
It is connect with power supply, ensures normal work of other component etc..
It will be understood by those skilled in the art that the continuation of insurance behavior that the system structure shown in Fig. 1 is not formed to user is pre-
The restriction of examining system can include either combining certain components or different component cloth than illustrating more or fewer components
It puts.
Based on above-mentioned hardware configuration, the first embodiment of the continuation of insurance behavior prediction method of user of the present invention is proposed, referring to figure
2, in the present embodiment, the method includes the steps:
Step S10 is obtained and has been purchased the corresponding user data of user of insurance and corresponding according to insurance business to be predicted
Preset benchmark algorithm calculates the corresponding benchmark continuation of insurance probability value of the user;
In the present embodiment, the user for having purchased insurance can be conglomerate client or individual client, relate to
And the kind of insurance can be property insurance towards personal and/or enterprise or for personal health insurance or people
Body safety insurance etc..
Processor can obtain the corresponding user data of user for having purchased insurance, and the user data can be user's purchase
The data of all insurances, the assets information of user and the basic identity information of user, such as can include purchased insurance production
The name of an article claims, insure type, business personnel, phase pay premium, gross premium, user's assets information, age of user information, user have purchased guarantor
The claim number and user's the last time of danger buy the time of insurance.
User data can also be pre-processed after the corresponding user data of user is obtained, such as:According to the use
The title of all insurance products of family purchase, the entire quantity that user is bought to insurance products are added to obtain user and hold insurance production
The phase that the sum of product, user buy these insurance products pays premium the how many, phase is respectively needed to pay premium how many is amounted to and all unites
Meter, i.e., according to obtaining subsequently needing the condition data judged after original data processing.
For the different corresponding preset benchmark algorithms of insurance business to be predicted can be it is identical can also be different
, for example, having bought the user of commercial medical insurance, the continuation of insurance probability of the commercial medical insurance similar to homogeneity may be handed over
Bottom, and the continuation of insurance probability insured to personal accidental death and injury insurance or other financing types may be higher, for different insurances to be predicted
Different predictive algorithms may be used in corresponding preset benchmark algorithm of being engaged in.
In addition, in order to predict the continuation of insurance behavior of user, it is necessary to obtain the continuation of insurance probability of the insurance business to be predicted of user
Value, and the continuation of insurance probability value in order to obtain then need a benchmark continuation of insurance probability value as reference.It for example, can be in prediction point
The Satisfaction of product to having purchased insurance is issued the user with before analysis, directly user can be allowed to give a mark, can also according to
The respective option of family selection calculates score, it is possible to understand that ground is, different users can test and assess out different scores, test and assess out
Score can be converted to a reference value continuation of insurance probability value, can not also convert continuation of insurance probability value corresponding directly as user;Alternatively,
The different insurances that can also be bought according to user set different benchmark continuation of insurance probability values.
Step S20 obtains the prediction that multiple continuation of insurance behaviors to the user in different levels cause good effect
Condition, wherein, each level correspondence includes at least one predicted condition;
Since user is when considering whether continuation of insurance, physical condition and the Assets of itself of insurant etc. can be combined
Aspect considers, and therefore, the predicted condition of different levels can be set based on this, each predicted condition is that user is continued
Guarantor's behavior causes good effect.It is understood that the predicted condition can also be set as to the continuation of insurance to the user
Behavior causes the predicted condition of negative influence, then correspondingly the step of subsequent execution and judging result are contrary, herein no longer
It repeats.
It should be noted that different weights can be set for the corresponding predicted condition of different levels, such as first layer is pre-
The corresponding weight of survey condition is 0.462, and the corresponding weight of second layer predicted condition is 0.254;It can also be according to preset sequence
The predicted condition of different levels is ranked up, for example, being arranged according to the sequence of weight from high to low.Wherein, for pre-
The setting of the weight of survey condition is mainly obtained by the continuation of insurance ratio of the existing client of insurance company and existing client, may be used also
With with the propulsion of time, constantly to company, the continuation of insurance ratio of existing client and existing client are updated, so as in real time
Update the corresponding weight of predicted condition.
Step S30 according to the user data and the corresponding benchmark continuation of insurance probability value of the user, passes through decision tree mould
Type, according to the sequence of the level, using the predicted condition as decision node, select successively the decision node of corresponding level into
Row classification, to obtain the continuation of insurance probability value of the corresponding insurance business to be predicted of the user.
A upper decision node can pass through decision branch and current decision node derived from the classification situation of corresponding predicted condition
Connection, current decision node can be connected by decision branch derived from the classification situation of corresponding predicted condition with next decision node
It connects.The classification of decision node is then the sequence progress according to level, and the classification of each decision node can influence the continuation of insurance of user
The corresponding continuation of insurance probability value of behavior when all hierarchical classifications are completed, that is, obtains the corresponding continuation of insurance probability value of user.
Same level can be corresponded to including at least one predicted condition, when same level includes two or more
Predicted condition, the corresponding weight of predicted condition of these same level can be the same or different.Below with same level
It is illustrated including two predicted conditions, such as the corresponding predicted condition of last layer decision node is user Yi Gou Pingan Insurances
Sum is more than 5.5, if the sum of user Yi Gou Pingan Insurances is more than 5.5, the corresponding prediction item of current decision node
Part has purchased H706 products for user;If the sum of user Yi Gou Pingan Insurances is less than or equal to 5.5, current decision section
The corresponding predicted condition of point has purchased H705 products for user.
Classify to obtain the continuation of insurance probability of the insurance business to be predicted of user in addition, also performing decision-tree model with three levels
Value is illustrated, it is assumed that it is 10 that A client, which holds each series of products sum of safety, and the phase pays premium and adds up to 2000 yuan, and purchase
Product H706 is bought.First first layer when judge that user holds the product sum of Pingan Insurance and whether holds more than 5.5, A clients
The product sum of Pingan Insurance continues to judge the second layer after meeting more than 5.5, and the predicted condition of the second layer is paid premium by a definite date and added up to
Premium is paid the phase more than 1885.0, A clients adds up to 1885.0 yuan of satisfaction;Then third layer is judged again, and the predicted condition of third layer is
User has purchased product H706, and A client has purchased product H706, then analysis system judges that the continuation of insurance probability of A client is automatically
85.64%.
The present embodiment is based on decision-tree model, according to the multiple to the user of benchmark continuation of insurance probability value combination different levels
Continuation of insurance behavior cause the predicted condition of good effect that the decision node of corresponding level is selected to classify successively, so as to obtain
The continuation of insurance probability value of the corresponding insurance business to be predicted of user is stated, so as to provide a kind of user based on big data algorithm model
Continuation of insurance behavior prediction method, solve salesman and judge that the continuation of insurance behavior accuracy of user is not high by personal experience and ask
Topic improves the stability of prediction.Furthermore this programme only needs to obtain user data at user or salesman, suitable for big
Measure the behavior prediction of client.
Referring to Fig. 3, the first embodiment of the continuation of insurance behavior prediction method based on user of the present invention proposes user's of the present invention
The second embodiment for behavior prediction method of continuing insurance, in the present embodiment, the step S30 includes:
Step S31, according to the user data and the corresponding benchmark continuation of insurance probability value of the user, by the prediction item
Part selects the decision node of corresponding level according to preset sequence, and sentence according to the user data successively as decision node
Whether the predicted condition is true under the disconnected decision node;In the decision node of selection corresponding level successively and determine the decision section
When the predicted condition is set up under point, step S32 is performed;In the decision node of selection corresponding level successively and determine the decision section
When the predicted condition is invalid under point, step S33 is performed;
Step S32 promotes the continuation of insurance probability of the user according to the corresponding weight of the predicted condition;
Step S33 reduces the continuation of insurance probability of the user according to the corresponding weight of the predicted condition;
Cause the predicted condition of good effect that can promote the continuation of insurance probability of user the continuation of insurance behavior of user;On the contrary not to
The continuation of insurance behavior at family causes good effect or even causes the predicted condition of negative effect, can reduce the continuation of insurance probability of user, because
This when to user continuation of insurance behavior generate good effect predicted condition set up when, can the continuation of insurance based on the user measured it is general
Rate promotes the continuation of insurance probability of user according to the corresponding weight of predicted condition;When the continuation of insurance behavior generation good effect to user
When predicted condition is invalid, can the continuation of insurance probability based on the user measured, reduce and use according to the corresponding weight of predicted condition
The continuation of insurance probability at family.
For example, level sum can be set as M (M=1,2,3..., M), only there are one predicted conditions for each level correspondence, divide
First layer is not corresponded to as M1, second layer M2, until MM, benchmark continuation of insurance probability value be N, the corresponding weight of each layer of predicted condition
The probability lifting values being converted into are NM(M=1,2,3 ..., M), the probability value for needing the user obtained are P.When proceeding by use
During the continuation of insurance behavior prediction at family, the corresponding predicted condition M of first layer is first determined whether1It is whether true, as predicted condition M1During establishment,
The continuation of insurance probability currently measured after then judging is N1+N;Then judge whether the second layer corresponds to predicted condition true, when pre-
Survey condition M2When invalid, then the continuation of insurance probability that is measured after judging is N+N1-N2, until determining MMAfter layer, survey
The continuation of insurance probability obtained is the continuation of insurance probability value P of the insurance business to be predicted of the user finally obtained.
Step S34, after the completion of decision node selection, the continuation of insurance for obtaining the insurance business to be predicted of the user is general
The result of calculation of rate value.
The opportunity that decision node selection is completed can be by setting number of plies counter to be confirmed, the corresponding prediction of each layer
Condition stub is completed, and counter adds one, and until counter values equal to until all numbers of plies, will finally classify, the continuation of insurance that measures is general
Continuation of insurance probability value of the rate as the insurance business to be predicted of user.This programme describes continuation of insurance row of the decision-tree model for user
For the process of prediction, the computational methods of the continuation of insurance probability value of the corresponding insurance business to be predicted of user are given, solve sale
Member predicts the problem of user's continuation of insurance behavior accuracy is not high by personal experience, convenient for dividing the continuation of insurance behavior of a large number of users
Analysis.
Referring to Fig. 4, the first embodiment of the continuation of insurance behavior prediction method based on user of the present invention proposes user's of the present invention
The 3rd embodiment for behavior prediction method of continuing insurance in the present embodiment, further includes after the step S30:
Step S40, obtain described in purchased insurance user's reality continuation of insurance information;
It, can be with and in order to verify whether prediction result is accurate since this programme is the prediction of the continuation of insurance behavior for user
The corresponding user of all users purchased insurance all have exceeded collect after the continuation of insurance time limit statistics purchased insurance user it is practical
Continuation of insurance information.Whether practical continuation of insurance information can generate continuation of insurance behavior, the corresponding insurance of continuation of insurance behavior letter in detail including user
Breath etc..
Step S50 according to the continuation of insurance information of user's reality for having purchased insurance, judges described preset to the user
Continuation of insurance behavior cause the predicted condition of good effect whether true;
Step S60 is performed while step S50 is performed,
Step S60 according to the corresponding user data of the user for having purchased insurance, judges described preset to the user
Continuation of insurance behavior cause the predicted condition of good effect whether true;
Determine that the predicted condition is set up, but according to described in the continuation of insurance information of user's reality for having purchased insurance according to
When having purchased the corresponding user data of user of insurance and determining that the predicted condition is invalid, step S61 is performed;
Step S61, it is false negative decision condition to determine the predicted condition;
Determine that the predicted condition is set up, and according to described in the continuation of insurance information of user's reality for having purchased insurance according to
When having purchased the corresponding user data of user of insurance and determining that the predicted condition is set up, step S62 is performed;
Step S62, it is real decision condition to determine the predicted condition;
Determine that the predicted condition is invalid, and according to institute in the continuation of insurance information of user's reality for having purchased insurance according to
It states when having purchased the corresponding user data of user of insurance and determining that the predicted condition is invalid, performs step S63;
Step S63, it is very negative decision condition to determine the predicted condition;
Determine that the predicted condition is invalid, but according to institute in the continuation of insurance information of user's reality for having purchased insurance according to
It states when having purchased the corresponding user data of user of insurance and determining that the predicted condition is set up, performs step S64;
Step S64, it is false positive decision condition to determine the predicted condition.
Processor is according to the practical continuation of insurance information for having purchased Insurance User of collection, it may be determined that user is under truth
No generation continuation of insurance behavior.The true continuation of insurance situation of existing subscriber's data classification results and user can be combined by processor,
Determine the authenticity of predicted condition.
Wherein, the continuation of insurance behavior to the user causes the predicted condition of good effect that can include in the following conditions
At least one:
The insurance total quantity of purchase of the user is more than preset amount threshold;
The phase of the user pays gross premium more than preset premium threshold value;
The insurance products of purchase of the user include pre-set product and number corresponding insurance products;
Described user's the last time buys the time of insurance in preset time range;
The age of the user is in preset the range of age;
The claim number for having purchased insurance of the user is in preset frequency threshold value;
The asset transition rate of the user is more than preset rate of change.
In addition to this it is possible to newly-increased others predicted condition and the corresponding weight of the predicted condition.
In logical algorithm subject, if predicted condition, which judges to set up, represents it is positive conditions or positive condition;It is if pre-
Survey condition judgment is invalid to represent it is negative condition or negative condition.But if predicted condition is correct, table when actually occurring
It is true condition to show the predicted condition;If predicted condition is wrong, it is false condition to represent the predicted condition.The two is predicted and is tied
Altogether, that is, will appear four kinds of situations, correspond to respectively false negative decision condition, real decision condition, very negative decision condition and
False positive decision condition.
Further, after the situation for confirming predicted condition, following steps can also be performed:
Step S70 records all false negative decision conditions, real decision condition, false positive decision condition and very negative judgement item
The corresponding number of part;
It, can be by all event types pair after actual conditions and the corresponding type of previous data validation predicted condition is combined
The number answered preserves in memory.
Step S80, according to the negative decision condition of vacation, real decision condition, false positive decision condition and very negative decision condition pair
The number answered calculates the accuracy rate of the continuation of insurance behavior of user by following algorithm;
Wherein, Accuracy is accuracy rate, TP is the corresponding number of real decision condition, TN
Be the corresponding number of very negative decision condition, ∑ Total be false negative decision condition, real decision condition, false positive decision condition and
Very the sum of corresponding number of negative decision condition.
Accuracy rate is the percentage that the accurate predicted condition of classification accounts for all predicted condition sums, real decision condition and very
Negative decision condition addition is the accurate predicted condition of classification, and the corresponding condition number addition of all possibilities is all pre-
Survey condition sum.The accuracy situation that can obtain the continuation of insurance behavior prediction model of user by calculating accuracy rate, convenient point
Analysis personnel adjust the parameter, predicted condition and corresponding weight of model according to accuracy rate result immediately.
Further, it is also possible to pass through algorithmError rate is calculated, wherein Error rate represent wrong
Accidentally rate, FP represent the corresponding number of false positive decision condition, and FN represents the corresponding number of false negative decision condition, and ∑ Total is false negative
Decision condition, real decision condition, vacation just decision condition and very the sum of corresponding number of negative decision condition.Further, it is wrong
Accidentally rate is also equal to 1 and subtracts accuracy rate.
Can also by algorithm sensitive=TP/P meter sensitivities, represent in all positive predicted conditions by point to
Thus ratio, can weigh out recognition capability of the continuation of insurance behavior prediction to positive predicted condition of user, wherein sensitive is represented
Sensitivity, TP represent real decision condition, and P represents the sum of real decision condition number corresponding with false just decision condition.
Can also special efficacy degree be calculated by algorithm specificity=TN/N, represent in all negative predicted conditions by point pair
Ratio, can thus weigh out user continuation of insurance behavior prediction to bear predicted condition recognition capability, wherein specificity
Represent special efficacy degree, TN represents very negative decision condition, and P represents that the very negative corresponding number of decision condition is corresponding with the negative decision condition of vacation
The sum of number.
Comprehensive evaluation index and robustness of evaluation processing missing values and exceptional value ability etc. can also be thus obtained to comment
Valency index can also obtain ROC (Receiver Operating Characteristic) curves and PR (Precision-
Recall) curve.
Referring to Fig. 5, the first embodiment of the continuation of insurance behavior prediction method based on user of the present invention proposes user's of the present invention
The fourth embodiment for behavior prediction method of continuing insurance, in the present embodiment, the step S10 includes:
Step S11 obtains the corresponding user data of user for having purchased insurance;
Step S12 obtains the Satisfaction information for having purchased insurance of the user;
The present embodiment is the corresponding user data of user of insurance have been purchased to being obtained in first embodiment and according to default
Benchmark algorithm calculate the further improvement of the corresponding benchmark continuation of insurance probability value of the user, the use of insurance has been purchased for acquisition
The corresponding user data in family has no improvement, the present embodiment and first embodiment, and difference lies in the present embodiment further provides
The preset benchmark algorithm for obtaining the corresponding benchmark continuation of insurance probability value of user.Analysis system or salesman can continue insurance in user
Prompting user fills in Satisfaction table before time limit terminates, and the satisfaction to have purchased insurance is given a mark, and is then generated and has been purchased guarantor
The Satisfaction information of danger is to processor.
Step S13, according to the Satisfaction information for having purchased insurance of the user, it is corresponding that measuring and calculating obtains the user
Benchmark continuation of insurance probability value.
After processor receives the Satisfaction information for having purchased insurance, it can be corresponded to according to every information in test and appraisal information table
Score value calculate user's corresponding benchmark continuation of insurance probability value.For example, user fills in is Satisfaction table for prompting, altogether 10
Multiple-choice question is divided per problem 5, and after the completion of user A is filled in, the score value for having 8 problems is 3 points, and 2 problems are 5 points, then fractional bits of testing and assessing are
8*3+2*5=34 points, it can be translated into baseline probability value in proportion.
The continuation of insurance probability value of user is obtained by the Satisfaction information combination decision-tree model of user, can more be pasted
The true continuation of insurance situation of nearly different user improves the accuracy of continuation of insurance behavior prediction.
Referring to Fig. 6, the first embodiment of the continuation of insurance behavior prediction method based on user of the present invention proposes user's of the present invention
5th embodiment of behavior prediction method of continuing insurance, in the present embodiment, the step S30 includes:
Step S35 is repeatedly performed according to the user data and the corresponding benchmark continuation of insurance probability value of the user,
By decision-tree model, according to the sequence of the level, using the predicted condition as decision node, corresponding level is selected successively
Decision node the step of being classified, to obtain multiple decision trees continuation of insurance probability values, wherein, perform the prediction item of step every time
Part is not exactly the same;
The user for only being classified and being predicted by a decision tree continues insurance probability may be less accurate, improves accurate
True method can repeatedly be verified, and ensure that the predicted condition of different levels classified every time is endless
Exactly the same, it is all independent without being associated with to make the classification between every decision tree, while more decision trees constitute one at random
Forest.
Step S36, is voted based on Random Forest model, to select to throw from the multiple decision tree continuation of insurance probability value
Continuation of insurance probability value of the most decision tree continuation of insurance probability value of ticket as the corresponding insurance business to be predicted of the user.
When getting the corresponding user data of the user for having purchased insurance, the more decision trees allowed in random forest are sentenced
It is disconnected, using most result of voting as the continuation of insurance probability value of final user, improve correct probability.
The present invention also provides the continuation of insurance behavior prediction system of user a kind of, referring to Fig. 7, in one embodiment, including:
Acquisition module 10, for obtaining the corresponding user data of user for having purchased insurance;
Computing module 20 calculates the user couple for the corresponding preset benchmark algorithm according to insurance business to be predicted
The benchmark continuation of insurance probability value answered;
The acquisition module 10 is additionally operable to multiple continuation of insurance behavior to the user of the acquisition in different levels and causes to accumulate
The predicted condition of pole effect, wherein, each level correspondence includes at least one predicted condition;
Selecting module 30, for according to the user data and the corresponding benchmark continuation of insurance probability value of the user, passing through
Decision-tree model according to the sequence of the level, using the predicted condition as decision node, selects determining for corresponding level successively
Plan node is classified, to obtain the continuation of insurance probability value of the corresponding insurance business to be predicted of the user.
In another embodiment, the selecting module 30 includes judging unit 31, processing unit 32 and acquiring unit 33;
The judging unit 31, for being continued insurance probability value according to the user data and the corresponding benchmark of the user,
Using the predicted condition as decision node, the decision node of corresponding level is selected successively according to preset sequence, and according to institute
It states user data and judges whether the predicted condition is true under the decision node;And in the decision node for selecting corresponding level successively
And when determining that the predicted condition is set up under the decision node, the processing unit 32 is triggered according to the corresponding power of the predicted condition
Bring up again the continuation of insurance probability for rising the user;It is described pre- under the decision node of selection corresponding level successively and the determining decision node
When survey condition is invalid, the continuation of insurance for triggering the processing unit 32 according to the predicted condition corresponding weight reduction user is general
Rate;
The acquiring unit 33, for after the completion of decision node selection, obtaining the insurance to be predicted of the user
The result of calculation of the continuation of insurance probability value of business.
In another embodiment, judgment module 40 and determining module 50 are further included;
The acquisition module 10, for obtaining the continuation of insurance information of the user's reality for having purchased insurance;
The judgment module 40 for having purchased the continuation of insurance information of user's reality of insurance according to, judges described default
Cause the predicted condition of good effect whether true the continuation of insurance behavior of the user;
Determine that the predicted condition is set up, but according to described in the continuation of insurance information of user's reality for having purchased insurance according to
When having purchased the corresponding user data of user of insurance and determining that the predicted condition is invalid, triggering that the determining module 50 determines should
Predicted condition is false negative decision condition;
Determine that the predicted condition is set up, and according to described in the continuation of insurance information of user's reality for having purchased insurance according to
When having purchased the corresponding user data of user of insurance and determining that the predicted condition is set up, trigger the determining module 50 and determine that this is pre-
Survey condition is real decision condition;
Determine that the predicted condition is invalid, and according to institute in the continuation of insurance information of user's reality for having purchased insurance according to
It states when having purchased the corresponding user data of user of insurance and determining that the predicted condition is invalid, triggers the determining module 50 and determine
The predicted condition is very negative decision condition;
Determine that the predicted condition is invalid, but according to institute in the continuation of insurance information of user's reality for having purchased insurance according to
It states when having purchased the corresponding user data of user of insurance and determining that the predicted condition is set up, triggering that the determining module 50 determines should
Predicted condition is false positive decision condition.
In another embodiment, logging modle 60 is further included;
The logging modle 60, for record all false negative decision conditions, real decision condition, false positive decision condition and
The very corresponding number of negative decision condition;
The computing module 20 is additionally operable to according to the negative decision condition of vacation, real decision condition, false positive decision condition and true
The corresponding number of negative decision condition calculates the accuracy rate of the continuation of insurance behavior of user by following algorithm;
Wherein, Accuracy is accuracy rate, TP is the corresponding number of real decision condition, TN
Be the corresponding number of very negative decision condition, ∑ Total be false negative decision condition, real decision condition, false positive decision condition and
Very the sum of corresponding number of negative decision condition.
In another embodiment, the continuation of insurance behavior to the user causes the predicted condition of good effect to include following
At least one of condition:
The insurance total quantity of purchase of the user is more than preset amount threshold;
The phase of the user pays gross premium more than preset premium threshold value;
The insurance products of purchase of the user include pre-set product and number corresponding insurance products;
Described user's the last time buys the time of insurance in preset time range;
The age of the user is in preset the range of age;
The claim number for having purchased insurance of the user is in preset frequency threshold value;
The asset transition rate of the user is more than preset rate of change.
In another embodiment, the computing module 20 is additionally operable to obtain the Satisfaction for having purchased insurance of the user
Information;And according to the Satisfaction information for having purchased insurance of the user, measuring and calculating obtains the corresponding benchmark continuation of insurance of the user
Probability value.
Selecting module 30 includes acquiring unit 33 and selecting unit 34;
The acquiring unit 33 is performed for being repeated as many times according to the user data and the corresponding benchmark of the user
It continues insurance probability value, by decision-tree model, according to the sequence of the level, using the predicted condition as decision node, successively
The step of decision node of selection corresponding level is classified, to obtain multiple decision tree continuation of insurance probability values, wherein, it performs every time
The predicted condition of step is not exactly the same;
The selecting unit 34, votes for being based on Random Forest model, with general from the continuation of insurance of the multiple decision tree
Select the decision tree for voting most continuation of insurance probability value general as the continuation of insurance of the corresponding insurance business to be predicted of the user in rate value
Rate value.
Please continue to refer to Fig. 1, the present invention also provides the continuation of insurance behavior prediction system of user a kind of, including communication module 10,
Memory 20, processor 30 and it is stored in the computer program that can be run on memory 20 and on processor 30;The processing
Device 30 realizes the step of continuation of insurance behavior prediction method of user as described above when performing the computer program.
The present invention also provides a kind of computer readable storage medium, calculating is stored on the computer readable storage medium
Machine program, the computer program realize the step of the continuation of insurance behavior prediction method of user as described above when being executed by processor
Suddenly.
In the description of this specification, reference term " embodiment ", " another embodiment ", " other embodiment " or "
The description of one embodiment~X embodiment " etc. mean to combine the specific features that the embodiment or example describe, structure, material or
Person's feature is contained at least one embodiment of the present invention or example.In the present specification, to the schematic table of above-mentioned term
It states and may not refer to the same embodiment or example.Moreover, specific features, structure, material, method and step or the spy of description
Point can in an appropriate manner combine in any one or more embodiments or example.
It should be noted that herein, term " comprising ", "comprising" or its any other variant are intended to non-row
His property includes, so that process, method, article or device including a series of elements not only include those elements, and
And it further includes other elements that are not explicitly listed or further includes intrinsic for this process, method, article or device institute
Element.In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that including this
Also there are other identical elements in the process of element, method, article or device.
The embodiments of the present invention are for illustration only, do not represent the quality of embodiment.
By the description of above embodiment, those skilled in the art can be understood that above-described embodiment method
The mode of required general hardware platform can be added to realize by software, naturally it is also possible to by hardware, but in many cases before
Person is more preferably embodiment.Based on such understanding, technical scheme of the present invention substantially makes tribute to the prior art in other words
The part offered can be embodied in the form of software product, which is stored in a computer-readable storage
In medium (such as ROM/RAM, magnetic disc, CD), used including some instructions so that a station terminal (can be mobile phone, computer takes
Be engaged in device, air conditioner or the network equipment etc.) perform method described in each embodiment of the present invention.
The embodiment of the present invention is described above in conjunction with attached drawing, but the invention is not limited in above-mentioned specific
Embodiment, above-mentioned specific embodiment is only schematical rather than restricted, and those of ordinary skill in the art are at this
Under the enlightenments of invention, present inventive concept and scope of the claimed protection are not being departed from, can also make many forms,
These are belonged within the protection of the present invention.
Claims (10)
1. the continuation of insurance behavior prediction method of a kind of user, which is characterized in that including step:
It obtains the corresponding user data of user for having purchased insurance and is calculated according to the corresponding preset benchmark of insurance business to be predicted
Method calculates the corresponding benchmark continuation of insurance probability value of the user;
The predicted condition that multiple continuation of insurance behaviors to the user in different levels cause good effect is obtained, wherein, institute
It states each level correspondence and includes at least one predicted condition;
According to the user data and the corresponding benchmark continuation of insurance probability value of the user, by decision-tree model, according to described
The sequence of level using the predicted condition as decision node, selects the decision node of corresponding level to classify, to obtain successively
Obtain the continuation of insurance probability value of the corresponding insurance business to be predicted of the user.
2. the continuation of insurance behavior prediction method of user as described in claim 1, which is characterized in that described according to the user data
And the corresponding benchmark continuation of insurance probability value of the user, by decision-tree model, according to the sequence of the level, by the prediction
Condition selects the decision node of corresponding level to classify as decision node successively, with obtain the user it is corresponding treat it is pre-
The step of surveying the continuation of insurance probability value of insurance business, including:
According to the user data and the corresponding benchmark continuation of insurance probability value of the user, using the predicted condition as decision section
Point selects the decision node of corresponding level according to preset sequence, and judges the decision node according to the user data successively
Under the predicted condition it is whether true;
It is pre- according to this when the predicted condition is set up under the decision node of selection corresponding level successively and the determining decision node
The corresponding weight of survey condition promotes the continuation of insurance probability of the user;
When predicted condition described under selecting the decision node of corresponding level successively and determining the decision node is invalid, according to this
The corresponding weight of predicted condition reduces the continuation of insurance probability of the user;
After the completion of decision node selection, the calculating knot of the continuation of insurance probability value of the insurance business to be predicted of the user is obtained
Fruit.
3. the continuation of insurance behavior prediction method of user as described in claim 1, which is characterized in that described according to the user data
And the corresponding benchmark continuation of insurance probability value of the user, by decision-tree model, according to the sequence of the level, by the prediction
Condition selects the decision node of corresponding level to classify as decision node successively, with obtain the user it is corresponding treat it is pre-
After the step of surveying the continuation of insurance probability value of insurance business, further include:
The continuation of insurance information of user's reality of insurance has been purchased described in obtaining;
According to the continuation of insurance information of user's reality for having purchased insurance, judge that the preset continuation of insurance behavior to the user is made
Whether the predicted condition into good effect is true;
Determine that the predicted condition is set up, but has been purchased according to described in the continuation of insurance information of user's reality for having purchased insurance according to
When the corresponding user data of user of insurance determines that the predicted condition is invalid, it is false negative judgement item to determine the predicted condition
Part;
Determine that the predicted condition is set up, and has been purchased according to described in the continuation of insurance information of user's reality for having purchased insurance according to
When the corresponding user data of user of insurance determines that the predicted condition is set up, it is real decision condition to determine the predicted condition;
According to described in purchased insurance the continuation of insurance information of user's reality determine that the predicted condition is invalid, and according to it is described
When the corresponding user data of user of purchase insurance determines that the predicted condition is invalid, it is very negative judgement item to determine the predicted condition
Part;
According to described in purchased insurance the continuation of insurance information of user's reality determine that the predicted condition is invalid, but according to it is described
When the corresponding user data of user of purchase insurance determines that the predicted condition is set up, it is false positive judgement item to determine the predicted condition
Part.
4. the continuation of insurance behavior prediction method of user as claimed in claim 3, which is characterized in that also wrapped after all steps
It includes:
It records all false negative decision conditions, real decision condition, false positive decision condition and really bears the corresponding number of decision condition;
The corresponding number of decision condition is born according to the negative decision condition of vacation, real decision condition, false positive decision condition and really, passed through
Following algorithm calculates the accuracy rate of the continuation of insurance behavior of user;
Wherein, Accuracy is accuracy rate, TP is the corresponding number of real decision condition, TN is true
Bear the corresponding number of decision condition, ∑ Total is false negative decision condition, real decision condition, false positive decision condition and is really born
The sum of corresponding number of decision condition.
5. the continuation of insurance behavior prediction method of user as described in claim 1, which is characterized in that the continuation of insurance to the user
Behavior causes the predicted condition of good effect to include at least one of the following conditions:
The insurance total quantity of purchase of the user is more than preset amount threshold;
The phase of the user pays gross premium more than preset premium threshold value;
The insurance products of purchase of the user include pre-set product and number corresponding insurance products;
Described user's the last time buys the time of insurance in preset time range;
The age of the user is in preset the range of age;
The claim number for having purchased insurance of the user is in preset frequency threshold value;
The asset transition rate of the user is more than preset rate of change.
6. the continuation of insurance behavior prediction method of user as described in claim 1, which is characterized in that further include step:
Obtain the Satisfaction information for having purchased insurance of the user;
According to the Satisfaction information for having purchased insurance of the user, measuring and calculating obtains the corresponding benchmark continuation of insurance probability of the user
Value.
7. the continuation of insurance behavior prediction method of user as described in claim 1, which is characterized in that described according to the user data
And the corresponding benchmark continuation of insurance probability value of the user, by decision-tree model, according to the sequence of the level, by the prediction
Condition selects the decision node of corresponding level to classify as decision node successively, with obtain the user it is corresponding treat it is pre-
The step of continuation of insurance probability value for surveying insurance business, includes:
It repeatedly performs according to the user data and the corresponding benchmark continuation of insurance probability value of the user, passes through decision tree mould
Type, according to the sequence of the level, using the predicted condition as decision node, select successively the decision node of corresponding level into
The step of row classification, to obtain multiple decision tree continuation of insurance probability values, wherein, each incomplete phase of predicted condition for performing step
Together;
It is voted based on Random Forest model, with the most decision of the selection ballot from the multiple decision tree continuation of insurance probability value
Continuation of insurance probability value of the tree continuation of insurance probability value as the corresponding insurance business to be predicted of the user.
8. a kind of continuation of insurance behavior prediction system of user, which is characterized in that including:
Acquisition module, for obtaining the corresponding user data of user for having purchased insurance;
Computing module, for calculating the corresponding benchmark continuation of insurance probability value of the user according to preset benchmark algorithm;
The acquisition module is additionally operable to multiple continuation of insurance behavior to the user of the acquisition in different levels and causes good effect
Predicted condition, wherein, each level correspondence includes at least one predicted condition;
Selecting module, for according to the user data and the corresponding benchmark continuation of insurance probability value of the user, passing through decision tree
Model according to the sequence of the level, using the predicted condition as decision node, selects the decision node of corresponding level successively
Classify, to obtain the continuation of insurance probability value of the corresponding insurance business to be predicted of the user.
9. the continuation of insurance behavior prediction system of a kind of user, which is characterized in that including communication module, processor, memory and storage
On a memory and the computer program that can run on a processor;The processor is realized such as when performing the computer program
The step of continuation of insurance behavior prediction method of claim 1-7 any one of them users.
10. a kind of computer readable storage medium, which is characterized in that be stored with computer on the computer readable storage medium
Program realizes the continuation of insurance behavior such as claim 1-7 any one of them users when the computer program is executed by processor
The step of Forecasting Methodology.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711497554.7A CN108197742A (en) | 2017-12-29 | 2017-12-29 | Continuation of insurance behavior prediction method, system and the computer readable storage medium of user |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711497554.7A CN108197742A (en) | 2017-12-29 | 2017-12-29 | Continuation of insurance behavior prediction method, system and the computer readable storage medium of user |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108197742A true CN108197742A (en) | 2018-06-22 |
Family
ID=62587557
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201711497554.7A Pending CN108197742A (en) | 2017-12-29 | 2017-12-29 | Continuation of insurance behavior prediction method, system and the computer readable storage medium of user |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108197742A (en) |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109447824A (en) * | 2018-09-26 | 2019-03-08 | 平安健康保险股份有限公司 | Declaration form data processing method, device, computer equipment and storage medium |
CN109785076A (en) * | 2019-01-28 | 2019-05-21 | 中国平安财产保险股份有限公司 | Acquisition methods, device, computer equipment and the storage medium of information |
CN110136011A (en) * | 2019-05-09 | 2019-08-16 | 泰康保险集团股份有限公司 | Insure intelligent price quoting method, device, medium and electronic equipment |
CN110659922A (en) * | 2018-06-28 | 2020-01-07 | 马上消费金融股份有限公司 | Client screening method, device, server and computer readable storage medium |
CN113177801A (en) * | 2020-03-04 | 2021-07-27 | 阿里巴巴集团控股有限公司 | Information processing method and device, electronic equipment and storage medium |
CN113837889A (en) * | 2021-11-26 | 2021-12-24 | 太平金融科技服务(上海)有限公司深圳分公司 | Service data generation method and device, computer equipment and storage medium |
CN117834780A (en) * | 2024-03-05 | 2024-04-05 | 济南云上电子科技有限公司 | Intelligent outbound customer intention prediction analysis system |
-
2017
- 2017-12-29 CN CN201711497554.7A patent/CN108197742A/en active Pending
Cited By (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110659922A (en) * | 2018-06-28 | 2020-01-07 | 马上消费金融股份有限公司 | Client screening method, device, server and computer readable storage medium |
CN110659922B (en) * | 2018-06-28 | 2021-01-26 | 马上消费金融股份有限公司 | Client screening method, device, server and computer readable storage medium |
CN109447824A (en) * | 2018-09-26 | 2019-03-08 | 平安健康保险股份有限公司 | Declaration form data processing method, device, computer equipment and storage medium |
CN109447824B (en) * | 2018-09-26 | 2024-06-28 | 平安健康保险股份有限公司 | Policy data processing method, device, computer equipment and storage medium |
CN109785076A (en) * | 2019-01-28 | 2019-05-21 | 中国平安财产保险股份有限公司 | Acquisition methods, device, computer equipment and the storage medium of information |
CN110136011A (en) * | 2019-05-09 | 2019-08-16 | 泰康保险集团股份有限公司 | Insure intelligent price quoting method, device, medium and electronic equipment |
CN113177801A (en) * | 2020-03-04 | 2021-07-27 | 阿里巴巴集团控股有限公司 | Information processing method and device, electronic equipment and storage medium |
CN113837889A (en) * | 2021-11-26 | 2021-12-24 | 太平金融科技服务(上海)有限公司深圳分公司 | Service data generation method and device, computer equipment and storage medium |
CN113837889B (en) * | 2021-11-26 | 2022-03-18 | 太平金融科技服务(上海)有限公司深圳分公司 | Service data generation method and device, computer equipment and storage medium |
CN117834780A (en) * | 2024-03-05 | 2024-04-05 | 济南云上电子科技有限公司 | Intelligent outbound customer intention prediction analysis system |
CN117834780B (en) * | 2024-03-05 | 2024-05-14 | 济南云上电子科技有限公司 | Intelligent outbound customer intention prediction analysis system |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108197742A (en) | Continuation of insurance behavior prediction method, system and the computer readable storage medium of user | |
CN107910068A (en) | Insure health risk Forecasting Methodology, device, equipment and the storage medium of user | |
CN108257027B (en) | Policy data auditing method and device, computer equipment and storage medium | |
CN110313009A (en) | The trust model and risk tolerance of learning object carry out calculation risk score | |
CN110417721A (en) | Safety risk estimating method, device, equipment and computer readable storage medium | |
CN107437198A (en) | Determine method, information recommendation method and the device of consumer's risk preference | |
CN112559900B (en) | Product recommendation method and device, computer equipment and storage medium | |
CN110009449A (en) | Business trust appraisal procedure, server and storage medium | |
CN109670852A (en) | User classification method, device, terminal and storage medium | |
CN109242573A (en) | Evaluation method, device, equipment and the storage medium of APP | |
CN108109066A (en) | A kind of credit scoring model update method and system | |
CN111773732A (en) | Target game user detection method, device and equipment | |
CN111160783A (en) | Method and system for evaluating digital asset value and electronic equipment | |
CN111695831B (en) | Open source code use risk assessment method and device and electronic equipment | |
CN110517071A (en) | Information forecasting method, device, equipment and storage medium based on machine mould | |
CN109636565A (en) | Processing method, device, equipment and the computer readable storage medium of risk data | |
CN115062734A (en) | Wind control modeling method, device, equipment and medium capable of outputting explanatory information | |
CN112449002B (en) | Method, device and equipment for pushing object to be pushed and storage medium | |
CN107908673A (en) | The real relationship match method, apparatus and readable storage medium storing program for executing of social platform user | |
JP6706397B1 (en) | Learning system, learning method, and program | |
CN107832925A (en) | Internet content risk evaluating method, device and server | |
US20080016058A1 (en) | Multi-candidate, multi-criteria decision-making method | |
CN115115322A (en) | Target group identification method, risk assessment method, apparatus, device and medium | |
CN111652492A (en) | Administrator evaluation system, method, electronic device and storage medium | |
CN112232945A (en) | Method and device for determining personal customer credit |
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 |
Application publication date: 20180622 |
|
RJ01 | Rejection of invention patent application after publication |