CN107977864A - A kind of customer insight method and system suitable for financial scenario - Google Patents
A kind of customer insight method and system suitable for financial scenario Download PDFInfo
- Publication number
- CN107977864A CN107977864A CN201711283640.8A CN201711283640A CN107977864A CN 107977864 A CN107977864 A CN 107977864A CN 201711283640 A CN201711283640 A CN 201711283640A CN 107977864 A CN107977864 A CN 107977864A
- Authority
- CN
- China
- Prior art keywords
- data
- user
- finance
- model
- finance data
- 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
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/047—Probabilistic or stochastic networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N7/00—Computing arrangements based on specific mathematical models
- G06N7/01—Probabilistic graphical models, e.g. probabilistic networks
-
- 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
- G06Q40/06—Asset management; Financial planning or analysis
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Finance (AREA)
- Development Economics (AREA)
- Accounting & Taxation (AREA)
- Mathematical Physics (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Strategic Management (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- General Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Biomedical Technology (AREA)
- Computational Linguistics (AREA)
- Entrepreneurship & Innovation (AREA)
- Biophysics (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Economics (AREA)
- Probability & Statistics with Applications (AREA)
- Game Theory and Decision Science (AREA)
- Mathematical Optimization (AREA)
- Human Resources & Organizations (AREA)
- Operations Research (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Analysis (AREA)
- Computational Mathematics (AREA)
- Technology Law (AREA)
- Algebra (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
This application discloses a kind of customer insight method and system suitable for financial scenario, wherein method includes:Obtain the finance data of multiple default dimensions of user;According to pre-set Data Analysis Model, the finance data is analyzed, obtains the assay value of each finance data for presetting dimension in the finance data of the multiple default dimension;According to the finance data and the assay value, analysis result of the output for the user.The Advantageous techniques effect of the application is:By Data Analysis Model, the finance data of the user of acquisition is analyzed, and obtains analysis result, so that the process that user sees clearly is more intelligent, efficient.
Description
Technical field
The present invention relates to computer software fields, smart mobile phone application software field, and in particular to one kind is suitable for finance
The customer insight method and system of scene.
Background technology
In financial field, it will usually introduce the method that some users see clearly, it is therefore an objective to be better understood by a human feelings of client
Condition and assets present situation, so that reach corresponding commercial object, such as:Lift sales achievement, the effective financial risks of financial product
Deng.Traditional user sees clearly, generally use weight scoring method.By design comprising such as:Whether age, income, have investment
The questionnaire of the problems such as experience (stock, fund etc.), then according to situation is filled in, if user is simply classified as dry type, and carries
For respective service.It is small that this method manually seen clearly not only obtains information content, and efficiency is very low.
The content of the invention
In view of the above shortcomings of the prior art, the present invention proposes a kind of customer insight method suitable for financial scenario and is
System, using the teaching of the invention it is possible to provide efficient data processing method.
The invention discloses a kind of customer insight method suitable for financial scenario, including:
Obtain the finance data of multiple default dimensions of user;
According to pre-set Data Analysis Model, the finance data is analyzed, obtains the multiple default dimension
The assay value of the finance data of each default dimension in the finance data of degree;
According to the finance data and the assay value, analysis result of the output for the user.
It is described according to pre-set Data Analysis Model in one embodiment of the disclosure, the finance data is carried out
Analysis, including:
When the finance data, which exists, to be updated, smooth function exploitation is carried out to the finance data of renewal;
The finance data after exploitation is analyzed according to pre-set Data Analysis Model.
In one embodiment of the disclosure, described obtain includes on the finance data of user:One in the following manner
Kind or a variety of obtain the finance data:
The presupposed information in being interacted with the Intelligent dialogue of user is extracted, the presupposed information includes keyword, dialogue duration;
The scene information that record user is logged in, the scene information bury point data, User Status, track including the page;
The operation data of user in the application is gathered, the operation data includes the page residence time, holds production
Product, state;
Record answer information of the user for questionnaire survey, the answer information includes the content of questionnaire, answer duration,
Every;
The default dimension include it is following in it is multiple:Financial strength, investment experiences, risk partiality, earnings target, personality
Feature and investment time limit.
It is described according to pre-set Data Analysis Model in one embodiment of the disclosure, the finance data is carried out
Analysis obtains analysis result, including:Using the one or more in data below analysis model, the finance data is divided
Analysis obtains analysis result:NLP is accompanied and is seen clearly model, neutral net, weighted decision tree-model, priori Bayesian model and row
For finance model.
It is described that the user is directed to according to the finance data and the assay value, output in one embodiment of the disclosure
Analysis result, including:
According to default dimension and the corresponding weighted value of each default dimension, the assay value is patterned;
The assay value of output pattern.
The present invention also provides a kind of customer insight system suitable for financial scenario, including:
Acquisition module, the finance data of multiple default dimensions for obtaining user;
Analysis module, for according to pre-set Data Analysis Model, analyzing the finance data, obtaining institute
State the assay value of the finance data of each default dimension in the finance data of multiple default dimensions;
Output module, for according to the finance data and the assay value, analysis result of the output for the user.
In one embodiment of the disclosure, the analysis module is additionally operable to:
When the finance data, which exists, to be updated, smooth function exploitation is carried out to the finance data of renewal;
The finance data after exploitation is analyzed according to pre-set Data Analysis Model.
In one embodiment of the disclosure, the one or more during the acquisition module is additionally operable in the following manner obtain
The finance data:
The presupposed information in being interacted with the Intelligent dialogue of user is extracted, the presupposed information includes keyword, dialogue duration;
The scene information that record user is logged in, the scene information bury point data, User Status, track including the page;
The operation data of user in the application is gathered, the operation data includes the page residence time, holds production
Product, state;
Record answer information of the user for questionnaire survey, the answer information includes the content of questionnaire, answer duration,
Every;
The default dimension include it is following in it is multiple:Financial strength, investment experiences, risk partiality, earnings target, personality
Feature and investment time limit.
In one embodiment of the disclosure, the analysis module is additionally operable to:Using one kind in data below analysis model or
It is a variety of, analysis is carried out to the finance data and obtains analysis result:NLP is accompanied and is seen clearly model, neutral net, weighted decision tree mould
Type, priori Bayesian model and behavior finance model.
In one embodiment of the disclosure, the output module, is additionally operable to:
According to default dimension and the corresponding weighted value of each default dimension, the assay value is patterned;
The assay value of output pattern.
The Advantageous techniques effect of the application is:By Data Analysis Model, the finance data of the user of acquisition is carried out
Analysis, and analysis result is obtained, so that the process that user sees clearly is more intelligent, efficient.
Brief description of the drawings
Fig. 1 show the flow signal of the customer insight method suitable for financial scenario of one embodiment of disclosure offer
Figure.
Fig. 2 show the flow signal for the customer insight method suitable for financial scenario that another embodiment of the disclosure provides
Figure.
Fig. 3 show the computation model structure diagram of neural network model.
Fig. 4 show the schematic diagram of priori Bayesian model.
Fig. 5 show perceived probability and the schematic diagram of effectiveness.
Fig. 6 show the visual analyzing result of front end output.
Fig. 7 show the visual analyzing result of rear end output.
Fig. 8 show the structural representation of the customer insight system suitable for financial scenario of one embodiment of disclosure offer
Figure.
Embodiment
In order to make the purpose , technical scheme and advantage of the present invention be clearer, with reference to the accompanying drawings and embodiments, it is right
The present invention is further elaborated.It should be appreciated that specific embodiment described herein is only to explain the present invention, not
For limiting the present invention.
Shown in refer to the attached drawing 1, wherein Fig. 1 show the customer insight method suitable for financial scenario of disclosure offer
Flow diagram.The customer insight method suitable for financial scenario that the disclosure provides comprises the following steps:
Step S101:Obtain the finance data of multiple default dimensions of user;
Step S102:According to pre-set Data Analysis Model, finance data is analyzed, obtains multiple default dimensions
The assay value of the finance data of each default dimension in the finance data of degree;
Step S103:According to finance data and assay value, analysis result of the output for user.
The embodiment of the present disclosure is analyzed the finance data of the user of acquisition, and obtain by Data Analysis Model
Analysis result, so that the process that user sees clearly is more intelligent, efficient.
It is described according to pre-set Data Analysis Model in one embodiment of the disclosure, the finance data is carried out
Analysis, obtains the assay value of each finance data for presetting dimension in the finance data of the multiple default dimension, including:
When the finance data, which exists, to be updated, smooth function exploitation is carried out to the finance data of renewal;
The finance data after exploitation is analyzed according to pre-set Data Analysis Model.
In one embodiment of the disclosure, the finance data of the multiple default dimensions for obtaining user includes:By following
One or more in mode obtain the finance data:
The presupposed information in being interacted with the Intelligent dialogue of user is extracted, the presupposed information includes keyword, dialogue duration;
The scene information that record user is logged in, the scene information bury point data, User Status, track including the page;
The operation data of user in the application is gathered, the operation data includes the page residence time, holds production
Product, state;
Record answer information of the user for questionnaire survey, the content of the answer information including questionnaire, answer duration,
Every;
The default dimension include it is following in it is multiple:Financial strength, investment experiences, risk partiality, earnings target, personality
Feature and investment time limit.
It is described according to pre-set Data Analysis Model in one embodiment of the disclosure, the finance data is carried out
Analysis obtains analysis result, including:Using the one or more in data below analysis model, the finance data is divided
Analysis obtains analysis result:NLP (Natural Language Processing, natural language processing) is accompanied and is seen clearly model, nerve
Network, weighted decision tree-model, priori Bayesian model and behavior finance model.
It is described that the user is directed to according to the finance data and the assay value, output in one embodiment of the disclosure
Analysis result, including:
According to default dimension and the corresponding weighted value of each default dimension, the assay value is patterned;
The assay value of output pattern.
In one embodiment of the disclosure, the method further includes:According to default analysis result with suggest correspondence,
There is provided the analysis result corresponding suggestion for the user.
Above-mentioned all optional technical solutions, can form the alternative embodiment of the disclosure according to any combination, and the disclosure is real
Example is applied no longer to repeat this one by one.
The flow chart of the customer insight method suitable for financial scenario of embodiment of the present disclosure offer, bag are provided
Include following steps:
Step S201:Obtain the finance data of multiple default dimensions of user.
Wherein, default dimension can include it is following in it is multiple:Financial strength, investment experiences, risk partiality, income mesh
Mark, character trait and investment time limit.The finance data of each default dimension can gradually increase with the deep contact of user.Both may be used
To see active user's last state, i.e., so-called " user's portrait ";Again it can be seen that the process that this user persistently changes, i.e., so-called
" user's video recording ".
Can in the following manner in one or more obtain finance data:
Extract the presupposed information in being interacted with the Intelligent dialogue of user:In being interacted by extraction with the Intelligent dialogue of user
Presupposed information, obtains user in sides such as financial strength, investment experiences, risk partiality, earnings target, character trait, investment time limits
The finance data in face, presupposed information include keyword, dialogue duration.
The scene information that record user is logged in:By recording decision-making of the user in concrete application scene, user is obtained
Data in terms of risk partiality, character trait.Scene is defined in specific application than wide, such as:User is a certain
Fund faces " the investment scene " how to invest, and user is facing " studying and judging when whether needing additional investment or reducing investment
Scene ", user are influenced to need to obtain " the help scene " of the consulting of consultant's formula, etc. be subject to macro market environment.Scene information bag
Include the page and bury point data, User Status, track.
Gather the operation data of user in the application:By gathering user's page access in the application, stopping
Stay the time, hold the operation datas such as product, operation note, obtain user in terms of investment experiences, risk partiality, character trait
Finance data.For example, when user more pays close attention to " conservative-low-risk " type, or access currency fund page number
More (and the residence time is longer), these operation datas can be as the financial strength of the user, risk partiality, character traits
The finance data of aspect.
Record answer information of the user for questionnaire survey:By concrete intelligence dialogue interaction in ingenious embedded questionnaire,
The problem of by being pre-designed, obtain user in financial strength, investment experiences, risk partiality, earnings target, character trait, throwing
The finance data in terms of the time limit is provided, answer information includes the content of questionnaire, answers duration, interval.
In terms of data source, finance data can be divided into unstructured data and structural data.Unstructured data
Including extracted in being interacted from Intelligent dialogue presupposed information, the operation data of user within a particular application is (such as including foot
Mark and behavioral data (containing page access, residence time, operation note etc.)).Structural data includes obtaining from questionnaire survey
Data.
Step S202:When finance data, which exists, to be updated, smooth function exploitation is carried out to the finance data of renewal.
When finance data is analyzed in the embodiment of the present disclosure, time dimension is introduced, when the data of certain each dimension have more
When new, this dimension data source (bottom index) last update time can be referred to, is contrasted afterwards with this renewal time.
After updating the data every time, smooth function exploitation can be passed through.Detailed process is:The access away from last time is taken to join
Number, using increment size as follow-up basis for estimation, and is stored in this dimension and sees clearly value.Smooth function formula is as follows:
Cn=a*C(n-1)+(1+a)C
Wherein, what C referred to this some dimension sees clearly result.I.e. this sees clearly final result, can combine this hole
Examine result and history is seen clearly result calculating and got.A is parameter, the value in (0,1) scope, by login number of days Separation control.When
Be spaced number of days it is longer when (be no more than 12 months, more than when by 12 months calculate), then a values are closer to 0.When being spaced n months,
A values are exactly (12-n)/12.
Step S203:The finance data after exploitation is analyzed according to pre-set Data Analysis Model, is obtained
Take the assay value of the finance data of each default dimension in the finance data of multiple default dimensions.
Wherein, the data analysis module used can include:NLP is accompanied and is seen clearly model, neural network model, weighted decision
Tree-model, priori Bayesian model, behavior finance model etc..
The computation model structure of neural network model is as shown in Figure 3.
Wherein, the w connected on side is weights, and f is activation primitive, and specific calculation formula is as follows:
Zj=∑I=1,2 ..., nwI, jxI, j-bj
In above formula, ajFor model prediction as a result, ZjTo consider input factor xI, j, weight wI, jAnd offset term bjAfterwards
Input variable.
, it is necessary to update the error of loss function generation, carry out backpropagation in specific training process, and to connection weight
Value is modified and is optimized.The computational methods of weights are:
Wj+1=Wj-σWbj+1=bj-σb
Wherein, ∑ is mean square error function, σW、σbRepresent the gradient of weight W and offset term b.
On the structure of deep neural network structure, each layer of parameter information can be carried out visualizing and configure.With
The increase of training iterations, loss can be greatly decreased.When more than 15 times iteration, less than 0.2 is substantially at.Moreover it is possible to
Intuitively find very much iterations and model accuracy walks power curve, and with the increase of training iterations, accuracy rate can be big
Width increase.When more than 15 times iteration, accuracy rate can be more than 95%.
Weighted decision tree-model mainly applies to six big dimensions seeing clearly data as priori Bayesian model
Input.The user being introduced below sees clearly index end model, i.e., a kind of weighted decision tree-model.
Priori Bayesian model can be updated with the time, can cover user characteristics gamut, suitable for a variety of
Type feature data, it is specific as shown in Figure 4.
The user behavior record of behavior finance model Behavior-based control finance, including:Intelligent dialogue interaction data, questionnaire tune
Data, behavior gathered data are looked into, formula is as follows:
In above formula, α, β, γ, r0Four parameters, distribution represent risk partiality (risk aversion), the risk partiality of user
(loss aversion), probability right, reference 4 features of earning rate, by solving optimization problem, user and financial asset are carried out
Matching.V is client's total utility function;S is scene number;ws(p;It is γ) that client is for Probability p in scene s, in probability right
Perceived probability under γ;v(xs) it is that client obtains x in scene ssSituation of Profit under effectiveness, figure illustrate as figure 5 illustrates.
User sees clearly index end model, which is actually derived from weighted decision tree-model, tree structure first layer knot
Point is that a represents financial strength respectively, and b represents investment experiences, and c represents risk partiality, and d represents earnings target, and e represents personality spy
Sign, f represent the investment time limit.In addition, the key element weight more new model of the model, the following institute of situation of matching value Optimized model
Show:
Key element weight more new model and parameter:
P(Wi+1)=P (Wi)*P(Wi+1|Wi)
The model is based on model-naive Bayesian, wherein, W is initial weight, and weight in a program can progressive alternate.P is
Bayesian prior and posterior probability.
Match value Optimized model and parameter:
MaX strategies fraction=f (v1+v2+v3+v4+v5+v6)
Wherein, v1、v2、v3、v4、v5、v6Parameter, it is as shown in the table:
Parameter name | |
v1 | Expected revenus |
v2 | It is expected to fluctuate |
v3 | Earning rate is less than r0Probability |
v4 | Expecting profit (earning rate is the conditional mean of timing) |
v5 | Expected loss (earning rate is the conditional mean of timing) |
v6 | Maximum is withdrawn |
It is the effect for independently seeing clearly user tag that user, which sees clearly index end model, can will see clearly result and show user.
User sees clearly the data input for the product that index end model also can be downstream, and the wake-up of matching, old user such as recommended products, use
Family experience optimization, displaying of customer manager's instrument etc..
Step S204:According to finance data and assay value, visual analysis result of the output for user in front end.
It is the visual analyzing exported according to user's finance data and assay value as shown in Figure 6 as a result, user can be helped
Carry out decision-making Investment & Financing behavior.
Step S205:According to default dimension and the corresponding weighted value of each default dimension, assay value is patterned,
The assay value of output pattern.
On program backstage, the embodiment of the present disclosure is set up one " seeing clearly account " for each user, output result displaying six
Dimension default greatly, (by taking some user as an example) as shown in Figure 7.Figure relies on six big dimension structures, and the shape of figure is by each dimension
The different weights of degree point determine.Adjusted in addition, analysis result can do dynamic with time change, such as:When the capital quantity of user
During there occurs changing, his financial strength, risk partiality, earnings target, investment time limit may all occur accordingly to change;When with
Family also more understands that oneself is desired assorted through after a period of time, he may grasp more Investment & Financing knowledge by study
, his investment experiences, earnings target may all occur accordingly to change.These change, and can all be embodied in the analysis knot of output
Among fruit.
Step S206:According to default analysis result and the correspondence suggested, the analysis result pair is provided to the user
The suggestion answered.
It can plan suitable product according to default analysis result and the correspondence of suggestion for user, provide individual character
Change financing to suggest.
The embodiment of the present disclosure, by the dimension default greatly of integrated use six, realization is more fully seen clearly;By Intelligent dialogue,
The modes such as scene record, behavior collection, questionnaire test and appraisal, realize more thoughtful see clearly;Accompanied by NLP and see clearly model, nerve net
Network, weighted decision tree-model, priori Bayesian model, behavior finance model etc., realize seeing clearly more efficiently;Pass through
Introduce time dimension and design matching algorithm, realize more accurate, more helpful to user see clearly.When different user's query phases
The problem of same, or same user different time inquire same problem when, answer have it is dramatically different so that user's body
Test more preferably, user's evaluation is more intelligent;Help more users to go planning life from the angle of financial asset, obtain more free
Wealth is experienced.
As shown in figure 8, the present invention also provides a kind of customer insight system suitable for financial scenario, including:
Acquisition module 1301, the finance data of multiple default dimensions for obtaining user;
Analysis module 1302, for according to pre-set Data Analysis Model, analyzing the finance data, obtaining
Take the assay value of the finance data of each default dimension in the finance data of the multiple default dimension;
Output module 1303, for according to the finance data and the assay value, analysis of the output for the user
As a result.
In one embodiment of the disclosure, the analysis module 1302 is additionally operable to:
When the finance data, which exists, to be updated, smooth function exploitation is carried out to the finance data of renewal;
The finance data after exploitation is analyzed according to pre-set Data Analysis Model.
In one embodiment of the disclosure, the one or more during the acquisition module 1301 is additionally operable in the following manner are come
Obtain the finance data:
Extraction interacted with the Intelligent dialogue of user in presupposed information, including refer to keyword (sentence), talk with duration etc.;
The scene information that record user is entered, including the page bury point data, User Status, track etc.;
Gather user's operation data in the application, including the page residence time, hold product, state etc.;With
And
Record answer information of the user for questionnaire survey, including the content of questionnaire, answer duration, interval etc.;
The default dimension includes following multiple:Financial strength, investment experiences, risk partiality, earnings target, character trait
With the investment time limit.And each dimension finance data can gradually increase with the deep contact of user.Both it can see that active user was newest
State, i.e., so-called " user's portrait ";Again it can be seen that the process that this user persistently changes, i.e., so-called " user's video recording ".
In one embodiment of the disclosure, the analysis module 1302 is additionally operable to:Use one in data below analysis model
Kind is a variety of, and analysis is carried out to the finance data and obtains analysis result:NLP is accompanied and is seen clearly model, neutral net, weighted decision
Tree-model, priori Bayesian model and behavior finance model.
In one embodiment of the disclosure, the output module 1303 is additionally operable to:
According to default dimension and the corresponding weighted value of each default dimension, the assay value is patterned;
The assay value of output pattern.
In one embodiment of the disclosure, the system also includes suggestion module, for according to default analysis result with building
The correspondence of view, provides the analysis result corresponding suggestion for the user.
Embodiment described above only expresses embodiments of the present invention, its description is more specific and detailed, but can not
Therefore it is interpreted as the limitation to the scope of the claims of the present invention.It should be pointed out that for those of ordinary skill in the art,
Without departing from the inventive concept of the premise, various modifications and improvements can be made, these belong to the protection model of the present invention
Enclose.Therefore, the protection domain of patent of the present invention should be determined by the appended claims.
Claims (10)
- A kind of 1. customer insight method suitable for financial scenario, it is characterised in that including:Obtain the finance data of multiple default dimensions of user;According to pre-set Data Analysis Model, the finance data is analyzed, obtains the multiple default dimension The assay value of the finance data of each default dimension in finance data;According to the finance data and the assay value, analysis result of the output for the user.
- 2. according to the method described in claim 1, it is characterized in that, described according to pre-set Data Analysis Model, to institute Finance data is stated to be analyzed, including:When finance data, which exists, to be updated, smooth function exploitation is carried out to the finance data of renewal;The finance data after exploitation is analyzed according to pre-set Data Analysis Model.
- 3. the according to the method described in claim 1, it is characterized in that, finance data of the multiple default dimensions for obtaining user Including:One or more in the following manner obtain the finance data:The presupposed information in being interacted with the Intelligent dialogue of user is extracted, the presupposed information includes keyword, dialogue duration;The scene information that record user is logged in, the scene information bury point data, User Status, track including the page;The operation data of user in the application is gathered, the operation data includes the page residence time, holds product, shape State;Answer information of the user for questionnaire survey is recorded, the answer information includes the content of questionnaire, answers duration, interval;The default dimension include it is following in it is multiple:Financial strength, investment experiences, risk partiality, earnings target, character trait With the investment time limit.
- 4. according to the method described in claim 1, it is characterized in that, described according to pre-set Data Analysis Model, to institute State finance data and carry out analysis acquisition analysis result, including:Using the one or more in data below analysis model, to described Finance data carries out analysis and obtains analysis result:Natural language processing accompanies and sees clearly model, neural network model, weighted decision tree Model, priori Bayesian model and behavior finance model.
- It is 5. defeated according to the method described in claim 1, it is characterized in that, described according to the finance data and the assay value Go out to be directed to the analysis result of the user, including:According to default dimension and the corresponding weighted value of each default dimension, the assay value is patterned;The assay value of output pattern.
- A kind of 6. customer insight system suitable for financial scenario, it is characterised in that including:Acquisition module, the finance data of multiple default dimensions for obtaining user;Analysis module, for according to pre-set Data Analysis Model, analyzing the finance data, obtaining described more The assay value of the finance data of each default dimension in the finance data of a default dimension;Output module, for according to the finance data and the assay value, analysis result of the output for the user.
- 7. system according to claim 6, it is characterised in that the analysis module is additionally operable to:When the finance data, which exists, to be updated, smooth function exploitation is carried out to the finance data of renewal;The finance data after exploitation is analyzed according to pre-set Data Analysis Model.
- 8. system according to claim 6, it is characterised in that the acquisition module be additionally operable in the following manner in one Kind or a variety of obtain the finance data:The presupposed information in being interacted with the Intelligent dialogue of user is extracted, the presupposed information includes keyword, dialogue duration;The scene information that record user is logged in, the scene information bury point data, User Status, track including the page;The operation data of user in the application is gathered, the operation data includes the page residence time, holds product, shape State;Answer information of the user for questionnaire survey is recorded, the answer information includes the content of questionnaire, answers duration, interval;The default dimension include it is following in it is multiple:Financial strength, investment experiences, risk partiality, earnings target, character trait With the investment time limit.
- 9. system according to claim 6, it is characterised in thatThe analysis module is additionally operable to:Using the one or more in data below analysis model, the finance data is carried out Analysis obtains analysis result:Natural language processing company is seen clearly model, neural network model, weighted decision tree-model, priori and is known Know Bayesian model and behavior finance model.
- 10. system according to claim 6, it is characterised in thatThe output module, is additionally operable to:According to default dimension and the corresponding weighted value of each default dimension, the assay value is patterned;The assay value of output pattern.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711283640.8A CN107977864A (en) | 2017-12-07 | 2017-12-07 | A kind of customer insight method and system suitable for financial scenario |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711283640.8A CN107977864A (en) | 2017-12-07 | 2017-12-07 | A kind of customer insight method and system suitable for financial scenario |
Publications (1)
Publication Number | Publication Date |
---|---|
CN107977864A true CN107977864A (en) | 2018-05-01 |
Family
ID=62009479
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201711283640.8A Pending CN107977864A (en) | 2017-12-07 | 2017-12-07 | A kind of customer insight method and system suitable for financial scenario |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107977864A (en) |
Cited By (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108776932A (en) * | 2018-05-22 | 2018-11-09 | 深圳壹账通智能科技有限公司 | Determination method, storage medium and the server of customer investment type |
CN109144640A (en) * | 2018-08-08 | 2019-01-04 | 互通金融科技有限公司 | Customized Intelligent rolling column system and method |
CN109214923A (en) * | 2018-08-02 | 2019-01-15 | 阿里巴巴集团控股有限公司 | A kind of financing task creating method and device |
CN109360106A (en) * | 2018-09-27 | 2019-02-19 | 北京京东金融科技控股有限公司 | Portrait construction method, system, medium and computer system |
CN109614509A (en) * | 2018-10-29 | 2019-04-12 | 山东中创软件工程股份有限公司 | Ship portrait construction method, device, equipment and storage medium |
CN109614982A (en) * | 2018-10-18 | 2019-04-12 | 平安科技(深圳)有限公司 | Product analysis method, apparatus, computer equipment and storage medium |
CN111242779A (en) * | 2020-01-03 | 2020-06-05 | 湖南工商大学 | Financial data characteristic selection and prediction method, device, equipment and storage medium |
CN113780806A (en) * | 2021-09-10 | 2021-12-10 | 未鲲(上海)科技服务有限公司 | Broker matching method, device, equipment and storage medium based on decision tree |
CN113781233A (en) * | 2021-07-29 | 2021-12-10 | 河北洋绣球网络科技有限公司 | Financial data service analysis system based on big data |
CN113781239A (en) * | 2021-09-10 | 2021-12-10 | 未鲲(上海)科技服务有限公司 | Policy determination method and device, electronic equipment and storage medium |
WO2022141877A1 (en) * | 2020-12-31 | 2022-07-07 | 平安科技(深圳)有限公司 | Financial data processing method, apparatus and device, and storage medium |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105824912A (en) * | 2016-03-15 | 2016-08-03 | 平安科技(深圳)有限公司 | Personalized recommending method and device based on user portrait |
US20170083965A1 (en) * | 2014-06-10 | 2017-03-23 | Huawei Technologies Co., Ltd. | Item Recommendation Method and Apparatus |
CN106991611A (en) * | 2017-03-27 | 2017-07-28 | 北京贝塔智投科技有限公司 | A kind of intelligence financing investment consultant's robot system and its method of work |
CN107077489A (en) * | 2015-06-29 | 2017-08-18 | 微软技术许可有限责任公司 | Automatic for multidimensional data is seen clearly |
CN107423442A (en) * | 2017-08-07 | 2017-12-01 | 火烈鸟网络(广州)股份有限公司 | Method and system, storage medium and computer equipment are recommended in application based on user's portrait behavioural analysis |
-
2017
- 2017-12-07 CN CN201711283640.8A patent/CN107977864A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20170083965A1 (en) * | 2014-06-10 | 2017-03-23 | Huawei Technologies Co., Ltd. | Item Recommendation Method and Apparatus |
CN107077489A (en) * | 2015-06-29 | 2017-08-18 | 微软技术许可有限责任公司 | Automatic for multidimensional data is seen clearly |
CN105824912A (en) * | 2016-03-15 | 2016-08-03 | 平安科技(深圳)有限公司 | Personalized recommending method and device based on user portrait |
CN106991611A (en) * | 2017-03-27 | 2017-07-28 | 北京贝塔智投科技有限公司 | A kind of intelligence financing investment consultant's robot system and its method of work |
CN107423442A (en) * | 2017-08-07 | 2017-12-01 | 火烈鸟网络(广州)股份有限公司 | Method and system, storage medium and computer equipment are recommended in application based on user's portrait behavioural analysis |
Cited By (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108776932A (en) * | 2018-05-22 | 2018-11-09 | 深圳壹账通智能科技有限公司 | Determination method, storage medium and the server of customer investment type |
CN109214923A (en) * | 2018-08-02 | 2019-01-15 | 阿里巴巴集团控股有限公司 | A kind of financing task creating method and device |
CN109144640A (en) * | 2018-08-08 | 2019-01-04 | 互通金融科技有限公司 | Customized Intelligent rolling column system and method |
CN109360106B (en) * | 2018-09-27 | 2021-03-30 | 京东数字科技控股有限公司 | Sketch construction method, system, medium and computer system |
CN109360106A (en) * | 2018-09-27 | 2019-02-19 | 北京京东金融科技控股有限公司 | Portrait construction method, system, medium and computer system |
CN109614982A (en) * | 2018-10-18 | 2019-04-12 | 平安科技(深圳)有限公司 | Product analysis method, apparatus, computer equipment and storage medium |
CN109614509A (en) * | 2018-10-29 | 2019-04-12 | 山东中创软件工程股份有限公司 | Ship portrait construction method, device, equipment and storage medium |
CN111242779A (en) * | 2020-01-03 | 2020-06-05 | 湖南工商大学 | Financial data characteristic selection and prediction method, device, equipment and storage medium |
CN111242779B (en) * | 2020-01-03 | 2023-08-18 | 湖南工商大学 | Financial data characteristic selection and prediction method, device, equipment and storage medium |
WO2022141877A1 (en) * | 2020-12-31 | 2022-07-07 | 平安科技(深圳)有限公司 | Financial data processing method, apparatus and device, and storage medium |
CN113781233A (en) * | 2021-07-29 | 2021-12-10 | 河北洋绣球网络科技有限公司 | Financial data service analysis system based on big data |
CN113780806A (en) * | 2021-09-10 | 2021-12-10 | 未鲲(上海)科技服务有限公司 | Broker matching method, device, equipment and storage medium based on decision tree |
CN113781239A (en) * | 2021-09-10 | 2021-12-10 | 未鲲(上海)科技服务有限公司 | Policy determination method and device, electronic equipment and storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107977864A (en) | A kind of customer insight method and system suitable for financial scenario | |
Ładyżyński et al. | Direct marketing campaigns in retail banking with the use of deep learning and random forests | |
An | Modeling human decisions in coupled human and natural systems: Review of agent-based models | |
Rambachan et al. | An economic approach to regulating algorithms | |
Sohn et al. | Decision tree based on data envelopment analysis for effective technology commercialization | |
Wilson et al. | Towards a generic artificial neural network model for dynamic predictions of algal abundance in freshwater lakes | |
Palmer et al. | Designing an artificial neural network for forecasting tourism time series | |
Sexton et al. | Employee turnover: a neural network solution | |
US7930196B2 (en) | Model-based and data-driven analytic support for strategy development | |
Gao et al. | Identifying preferred management options: An integrated agent-based recreational fishing simulation model with an AHP-TOPSIS evaluation method | |
Pan et al. | BIM log mining: Exploring design productivity characteristics | |
Ocampo-Duque et al. | A neural-fuzzy approach to classify the ecological status in surface waters | |
Alshmrany | Adaptive learning style prediction in e-learning environment using levy flight distribution based CNN model | |
Holt et al. | Using the theory of planned behavior to understand family forest owners’ intended responses to invasive forest insects | |
Robati et al. | Inflation rate modeling: adaptive neuro-fuzzy inference system approach and particle swarm optimization algorithm (ANFIS-PSO) | |
da Fonseca et al. | Agro 4.0: A data science-based information system for sustainable agroecosystem management | |
Czajkowski et al. | Steering the interpretability of decision trees using lasso regression-an evolutionary perspective | |
Gonçalves et al. | The role of attribute non-attendance on consumer decision-making: Theoretical insights and empirical evidence | |
Osoba et al. | Policy-focused agent-based modeling using RL behavioral models | |
Silva et al. | Information spreading in a population modeled by continuous asynchronous probabilistic cellular automata | |
Wang et al. | A new interactive model for improving the learning performance of back propagation neural network | |
Lee et al. | An agent-based cognitive mapping system for sales opportunity analysis | |
Widiputra et al. | Dynamic interaction networks versus local trend models for multiple time-series prediction | |
Wang | A net premium model for life insurance under a sort of generalized uncertain interest rates | |
Sang et al. | Graph convolution with topology refinement for automatic reinforcement learning |
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: 20180501 |
|
RJ01 | Rejection of invention patent application after publication |