CN110197301A - A kind of prediction technique of disposable income, device, server and storage medium - Google Patents
A kind of prediction technique of disposable income, device, server and storage medium Download PDFInfo
- Publication number
- CN110197301A CN110197301A CN201910447918.3A CN201910447918A CN110197301A CN 110197301 A CN110197301 A CN 110197301A CN 201910447918 A CN201910447918 A CN 201910447918A CN 110197301 A CN110197301 A CN 110197301A
- Authority
- CN
- China
- Prior art keywords
- data
- user
- probability value
- prediction model
- related information
- 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
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
-
- 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
- G06Q40/12—Accounting
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Development Economics (AREA)
- General Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Physics & Mathematics (AREA)
- Marketing (AREA)
- General Physics & Mathematics (AREA)
- Educational Administration (AREA)
- Finance (AREA)
- Accounting & Taxation (AREA)
- Game Theory and Decision Science (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Technology Law (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The embodiment of the invention discloses a kind of prediction technique of disposable income, device, server and storage mediums, this method comprises: obtaining destination-related information associated with target user；Wherein, the destination-related information includes the essential information of target user, the behavioral data of user and third party's data relevant to the user；Based on at least two target prediction models that preparatory training obtains, probability value corresponding with the destination-related information is determined respectively；Wherein, at least two target predictions model includes the first prediction model and the second prediction model；According to the probability value, obtain it is corresponding with disposable income dominate probability value, and according to the disposable income for dominating probability value and determining target user.The technical solution of the embodiment of the present invention realizes the disposable income of convenient, efficient prediction target user, and then the behavior of user can be better anticipated in terms of air control, to reduce the technical effect of risk.
Description
Technical field
The present embodiments relate to field of computer technology more particularly to a kind of prediction technique of disposable income, device,
Server and storage medium.
Background technique
In the prior art, the risk for judging user is mainly to be determined by the income of user.The income of user can be with
It is determined by the taxable amount of user.
Under normal conditions, there are huge differences between the income that the income of user and user can dominate, therefore pass through
The income of user is difficult the risk to determine the user.And the disposable income of user comes trade marketing and risk profile
It says, is a critically important data.
Currently, determining in the prior art according only to the tax revenue of user, the income of user, and then determine the risk of user,
In the presence of the technical problem of assessment inaccuracy.
Summary of the invention
The embodiment of the invention provides a kind of prediction technique of disposable income, device, server and storage mediums, with reality
The technical effect of now accurate, efficient and convenient determining target user's disposable income.
In a first aspect, the embodiment of the invention provides a kind of prediction techniques of disposable income, this method comprises:
Obtain destination-related information associated with target user；Wherein, the destination-related information includes target user
Essential information, user behavioral data and third party's data relevant to the user；
Based on at least two target prediction models that preparatory training obtains, determination is opposite with the destination-related information respectively
The probability value answered；Wherein, at least two target predictions model includes the first prediction model and the second prediction model；
According to the probability value, obtain it is corresponding with disposable income dominate probability value, and dominated according to described
Probability value determines the disposable income of target user.
Second aspect, the embodiment of the invention also provides a kind of prediction meanss of disposable income, the devices
Target user's related information obtains module, for obtaining destination-related information associated with target user；Wherein,
The destination-related information includes the essential information of target user, the behavioral data of user and third relevant to the user
Number formulary evidence；
Probability value determining module, at least two target prediction models for being obtained based on preparatory training, determine respectively with
The corresponding probability value of the destination-related information；Wherein, at least two target predictions model includes the first prediction model
With the second prediction model；
Disposable income determining module, for obtain it is corresponding with disposable income dominate probability value, and according to institute
The disposable income that probability value determines target user can be dominated by stating.
The third aspect, the embodiment of the invention also provides a kind of server, the server includes:
One or more processors；
Storage device, for storing one or more programs,
When one or more of programs are executed by one or more of processors, so that one or more of processing
Device realizes the prediction technique of the disposable income as described in the embodiment of the present invention is any.
Fourth aspect, it is described the embodiment of the invention also provides a kind of storage medium comprising computer executable instructions
Computer executable instructions by computer processor when being executed for executing dominating as described in the embodiment of the present invention is any
The prediction technique of income.
The technical solution of the embodiment of the present invention, by obtaining destination-related information associated with target user；Wherein, mesh
Marking related information includes the essential information of target user, the behavioral data of user and third party's data related to user；Base
In at least two target prediction models that preparatory training obtains, probability value corresponding with destination-related information is determined respectively；Its
In, at least two target prediction models include the first prediction model and the second prediction model；According to probability value, obtains and can dominate
Take in it is corresponding dominate probability value, and according to the disposable income that can dominate probability value and determine target user, solve existing
There is the disposable income that target user can not be determined in technology, realize convenient, efficient prediction target user dominates receipts
Enter, and then the behavior of user can be better anticipated in terms of air control, to reduce the technical effect of risk.
Detailed description of the invention
In order to more clearly illustrate the technical scheme of the exemplary embodiment of the present invention, below to required in description embodiment
The attached drawing to be used does a simple introduction.Obviously, the attached drawing introduced is present invention a part of the embodiment to be described
Attached drawing, rather than whole attached drawings without creative efforts, may be used also for those of ordinary skill in the art
To obtain other attached drawings according to these attached drawings.
Fig. 1 is a kind of prediction technique flow diagram of disposable income provided by the embodiment of the present invention one；
Fig. 2 is the flow diagram of training objective model provided by the embodiment of the present invention two；
Fig. 3 is a kind of another flow diagram of the prediction technique of disposable income provided by the embodiment of the present invention three；
Fig. 4 is the interference generated at default value provided by the embodiment of the present invention three to model.
Fig. 5 is the probability schematic diagram obtained provided by the embodiment of the present invention three using one of target prediction model；
Fig. 6 is the probability distribution signal obtained provided by the embodiment of the present invention three using at least two target prediction models
Figure；
Fig. 7 is a kind of user's normally refund rate and comparison signal for predicting refund rate provided by the embodiment of the present invention three
Figure；
Fig. 8 is a kind of prediction meanss structural schematic diagram of disposable income provided by the embodiment of the present invention four；
Fig. 9 is a kind of server architecture schematic diagram provided by the embodiment of the present invention five.
Specific embodiment
The present invention is described in further detail with reference to the accompanying drawings and examples.It is understood that this place is retouched
The specific embodiment stated is used only for explaining the present invention rather than limiting the invention.It also should be noted that in order to just
Only the parts related to the present invention are shown in description, attached drawing rather than entire infrastructure.
Embodiment one
Fig. 1 is a kind of prediction technique flow diagram of disposable income provided by the embodiment of the present invention one, this implementation
Example is applicable to the case where being managed during idle time to storage chip, this method can by the management system of storage chip Lai
It executes, which can be realized by way of software and/or hardware.
As described in Figure 1, the method for the present embodiment includes:
S110, acquisition destination-related information associated with target user.
Wherein, target user takes in situation it is to be understood that needing to assess user, and true according to the income situation of user
The user of its fixed risk.Related information may include the essential information of user, the behavioral data of user, and with target user's phase
Associated third party's data.
Wherein, the essential information of user can be, name, gender, age, work of user etc., the behavioral data of user
It can be the consumption data by some application program, credit card repayment data etc., third party's data can be consolidating for user
Surely data are paid, optionally, the data such as loan.
Specifically, can first obtain association letter associated with target user when determining the disposable income of target user
Breath.
S120, at least two target prediction models obtained based on preparatory training, are determined and destination-related information phase respectively
Corresponding probability value.
Wherein, target prediction model is that preparatory training obtains.The quantity of at least two target prediction models can be two
A prediction model is the first prediction model and the second prediction model respectively.First prediction model is for obtaining target user as richness
The probability of abundant user, the second prediction model are used to determine that target user to be the probability of poor user.Target prediction model is used for root
Probability value corresponding with target user is determined according to the related information of target user.
Specifically, will destination-related information associated with target user, be input to the first prediction model, obtain and target
Corresponding first probability value of user, the as probability value of affluent user；Meanwhile it will target association associated with target user
Information input obtains the second probability value corresponding with target user into the second prediction model, i.e. target user is poor uses
The probability value at family.
S130, according to probability value, obtain it is corresponding with disposable income dominate probability value, and according to probability can be dominated
It is worth the disposable income for determining target user.
Wherein, can dominate probability value can determine according to the first probability value obtained in S120 and the second probability value.
Optionally, after default value being subtracted second probability value, in addition first probability value obtains median；It will
The median obtains corresponding with institute's disposable income dominating probability value divided by 2.
Wherein, default value can be 1.1 subtract the second probability value obtain by target user the second prediction model item
The rich probability value obtained under part.First probability value obtains centre plus the rich probability value determined under the second prediction model
Value.Median obtains the probability value of the disposable income to match with target user divided by 2.According to the probability of disposable income
Value, determines the disposable income of target user, and then the information such as risk that can determine target user.
The technical solution of the embodiment of the present invention, by obtaining destination-related information associated with target user；Wherein, mesh
Marking related information includes the essential information of target user, the behavioral data of user and third party's data related to user；Base
In at least two target prediction models that preparatory training obtains, probability value corresponding with destination-related information is determined respectively；Its
In, at least two target prediction models include the first prediction model and the second prediction model；According to probability value, obtains and can dominate
Take in it is corresponding dominate probability value, and according to the disposable income that can dominate probability value and determine target user, solve existing
There is the disposable income that target user can not be determined in technology, realize convenient, efficient prediction target user dominates receipts
Enter, and then the behavior of user can be better anticipated in terms of air control, to reduce the technical effect of risk.
Embodiment two
Before dominating probability value according to the determination of the destination-related information of target user is corresponding with target user, also
Need at least two target prediction models of training.
S210, at least three groups of associated datas associated with the user are obtained as first sample data；Wherein, first sample
It include first kind data and Second Type data in data, the quantity of data is greater than Second Type in first kind data
The quantity of data in data.
It should be noted that obtaining the related information of multiple users in advance.The essential information of i.e. multiple users, behavioral data
Information and third party's data corresponding with each user.Optionally, the data at ten general-purpose families are obtained.Multiple users are divided
Two classes, one kind is used as training sample data, another kind of to be used as test sample data.Training sample data are used for training pattern, survey
Sample notebook data is used to the accuracy of Knowledge Verification Model.
Wherein, by training sample data according to preset rules, optionally, according to the height of wage income, by training sample
Data are divided into first kind data and Second Type data.Test sample data can be divided at least three grades,
For further verifying the accuracy of model.That is, the content of data is simultaneously in training sample data and test sample data
It is not exactly the same.
Optionally, when the Revenue for according to the related information of multiple users, determining user being higher than the first default Revenue,
Using the associated data of the user as first kind data；By according to the related information of multiple users, determine that the income of user is low
When the second default Revenue, using the associated data of the user as Second Type data.Wherein, the first default Revenue can be with
It is 8000 yuan, the second default Revenue can be 5000 yuan.
That is, first kind data are the subscriber association information for including at least one higher than the first default Revenue,
Second Type number is the subscriber association information that at least one is lower than the second default Revenue；Wherein the first default Revenue is higher than the
Two default Revenues.
, can be using the user of first kind data as booming income user for training pattern and Knowledge Verification Model, and mark
It is denoted as 1；The user of Second Type data is labeled as 0 as low income user.
Wherein, first sample data can be understood as participating in the data of model training.Specifically, available multiple first
Categorical data and Second Type data, wherein the quantity of first kind data is much larger than the quantity of Second Type data.Its
In, the quantity of data is it is to be understood that store the number of how many associated data information corresponding with user.
Illustratively, the quantity of first kind data is 20,000 in first sample data, and the quantity of Second Type data is
500.As training sample data, the first prediction model of Lai Xunlian.
In the present embodiment, in order to enable to model concentrate in the rich ranking for distinguishing user, therefore, mainly with the
Based on the subscriber association information of one categorical data, the model obtained at this time is denoted as the first prediction model.That is, first is pre-
Survey what model was mainly trained so that booming income crowd is target group.
S220, the first sample data are trained using gradient promotion decision making algorithm, obtain first prediction
Model.
Decision making algorithm is promoted using gradient, the first sample data got are trained, it is rich to obtain characterization user
The model of degree, i.e. the first prediction model.
Accordingly, it is also necessary to the second prediction model is obtained using identical method.May refer to Fig. 2 S230 and
S240.
S230, at least three groups of associated datas associated with the user are obtained as the second sample data；Wherein, the second sample
It include first kind data and Second Type data in data, the quantity of data is less than Second Type data in first kind data
The quantity of middle data.
Second prediction model in order to obtain, available multiple Second Type data and first kind data, wherein the
The quantity of one categorical data is much smaller than Second Type data bulk.Optionally, the quantity of first kind data be 5000, second
The quantity of categorical data is 50,000 etc., and the number of certain actual selection more may be only to illustrate, be not specifically limited herein.
Wherein, details are not described herein for first kind data and Second Type data.
Wherein, the second sample data refers to, sample data required for the second prediction model of training.In order to obtain with low receipts
The corresponding model of access customer, in the second sample data, the quantity of low income user is much larger than booming income number of users.
S240, second sample data is trained using gradient promotion decision making algorithm, obtains second prediction
Model.
Using identical algorithm, the second sample data is trained, obtains the second prediction model.
First prediction model is used to determine the probability that user is booming income user, and the second prediction model is for determining that user is
The probability of low income user.
It should be noted that above-mentioned steps S210-S220 and S230-S240 can be executed sequentially, it is only necessary to ensure
After being respectively trained, the first prediction model and the second prediction model are obtained.
It based on the above technical solution, can be with after obtaining the first prediction model and the second prediction model
The first prediction model and the second prediction model are verified by test sample data.
Optionally, at least one set of test sample data are separately input into first prediction model and described second in advance
Model is surveyed, distribution curve corresponding with the test sample data is obtained；If the distribution curve meets preset condition,
Using first prediction model and second preset model as target prediction model.
It should be noted that also user is marked in advance in test sample data.Wherein it is possible to which user is divided into
At least three grades optionally will be less than the user of the second default Revenue as low income user, in the first default Revenue
To the user between the second default Revenue as medium income value user, the user of the first default Revenue will be above as high
Take in user.By in test sample data, the related information of user is separately input into the first prediction model that training obtains, and
Second prediction model.The probability value exported respectively to first prediction model and second prediction model is handled,
Obtain probability distribution curve corresponding with different brackets user.When the error of distribution curve and default distribution curve is in default model
Within enclosing, it is determined that the first prediction model and the second prediction model are target prediction model.
The technical solution of the embodiment of the present invention, by obtaining destination-related information associated with target user；Wherein, mesh
Marking related information includes the essential information of target user, the behavioral data of user and third party's data related to user；Base
In at least two target prediction models that preparatory training obtains, probability value corresponding with destination-related information is determined respectively；Its
In, at least two target prediction models include the first prediction model and the second prediction model；According to probability value, obtains and can dominate
Take in it is corresponding dominate probability value, and according to the disposable income that can dominate probability value and determine target user, solve existing
There is the disposable income that target user can not be determined in technology, realize convenient, efficient prediction target user dominates receipts
Enter, and then the behavior of user can be better anticipated in terms of air control, to reduce the technical effect of risk.
Embodiment three
As a preferred embodiment of above-described embodiment, Fig. 3 is that one kind provided by the embodiment of the present invention three can dominate receipts
Another flow diagram of the prediction technique entered, as shown in Figure 3:
S301, the first training data is handled using preset algorithm.
In the present embodiment, can the related information in advance to multiple users handle, optionally, it is pre- to will be above first
If the user's mark of Revenue be 1, can using labeled as 1 subscriber association information as in first kind data；It will be less than
The user's mark of two default Revenues be 2, can using labeled as 0 subscriber association information as in Second Type data.Also
It is to say, the quantity of first kind data and Second Type data can have multiple.
Wherein, the first default Revenue is higher than the second default Revenue.The advantages of this arrangement are as follows: the income of user is
Continuous variable, if taking one of diacritical point of the value as high low income, that takes the data around score value will be to model
Differentiation generate greatly interference, referring to fig. 4, therefore during training prediction model, using the user of intermediate income
As sample data, but use the data of booming income user and low income user.That is, the obtained model of training can be with
Preferably according to the related information of user, the high low income of user is distinguished.
It wherein, include first kind data and Second Type data in first sample data.Obtain multiple first kind
Data obtain the related information of different booming income users, optionally, the related information of 50,000 booming income users；Correspondingly,
Multiple Second Type data are obtained, that is, the related information of different low income users are obtained, as the part in first sample data
Data, optionally, the related information of 5000 users.
It should be noted that the quantity in first kind data is much larger than Second Type number when the first prediction model of training
According to quantity.When the second prediction model of training, the quantity in first kind data is much smaller than the quantity of Second Type data.
Specifically, decision making algorithm can be promoted using gradient, first sample data are trained.
S302, the second training data is handled using preset algorithm.
In the present embodiment, can the related information in advance to multiple users handle, optionally, it is pre- to will be above first
If the user's mark of Revenue be 1, can using labeled as 1 subscriber association information as in first kind data；It will be less than
The user's mark of two default Revenues be 0, can using labeled as 0 subscriber association information as Second Type data.
It wherein, include first kind data and Second Type data in the second sample data.Obtain multiple Second Types
Data, the i.e. related information of low income user, optionally, the related information of 50,000 low income users；Correspondingly, from the first kind
The related information of selected section user in type data, the i.e. related information of booming income user, as the portion in the second sample data
Divide sample data, optionally, the related information of 5000 low income users.
It should be noted that the quantity in Second Type data is much larger than the first kind number when the second prediction model of training
According to quantity.
Specifically, decision making algorithm can be promoted using gradient, the second sample data is trained.
S303, after handling the first training data, the first prediction model is obtained.
After promoting decision making algorithm to first sample data training managing using gradient, the first prediction model is obtained.
S304, after handling the second training data, the second prediction model is obtained.
After promoting decision making algorithm to the second sample data training managing using gradient, the second prediction model has been obtained.
S305, associated data associated with target user is input to the first prediction model.
Obtain the related information of target user, optionally, the essential information of target user, target user behavioral data,
And third party's data corresponding with target user.
Specifically, related information corresponding with target user is input in the first prediction model.
S306, associated data associated with target user is input to the second prediction model.
Obtain the related information of target user, optionally, the essential information of target user, target user behavioral data,
And third party's data corresponding with target user.
Specifically, related information corresponding with target user is input in the second prediction model.
S307, the first prediction model export the first score.
Wherein, the first prediction model, for determining the wealth index of target user.First score is it is to be understood that target
User is the probability of affluent user, it is understood that is that target user's income is higher than the probability of the first default Revenue, the probability
Value is within the scope of 0-1.
Specifically, after the destination-related information of target user is input to the first prediction model, available first score,
The score can react the wealth of target user.
S308, the second prediction model export the second score.
Wherein, the second prediction model, for determining the poor index of target user.Second score is it is to be understood that target
User is the probability of poor user, it is understood that is that target user's income is lower than the probability of the second default Revenue, the probability
Value is within the scope of 0-1.
Specifically, after the destination-related information of target user is input to the second prediction model, available second score,
The score can react the poor degree of target user.
S309, the first score and the second score are handled to obtain disposable income score.
It should be noted that the related information of the user result that can be retrodicted, but obtained by the tax revenue of user with
There are a certain distance for the disposable income of user, therefore can use for reference the concept of Signal averaging and Fusion Model.It can be logical
Two different models of focus are crossed, the first prediction model and the second prediction model is can be, goes to calculate from least two dimensions
The income difference of user, and result is combined, complete the differentiation to target user's disposable income.
Wherein, disposable income score can be understood as that probability value can be dominated.
Specifically, can be using formula: disposable income score can dominate probability value=[the first score+(1-
Two scores)]/2.
Based on the above technical solution, the sample data more than 200,000 can be tested, if using tradition
Single model is predicted, i.e., when the related information of user being input in traditional Individual forecast model, available signal
Figure, as shown in Figure 5.At this time, it is not easy to distinguish the user of high low income.And when using the present embodiment technical solution, to user
It is as shown in Figure 6 to input separating capacity.Referring to Fig. 6, it can be seen that judge that user takes in how many abilities and is obviously improved.Accordingly
, after at least two target prediction models, compared with using single model, the risk of user is more accurately embodied,
Referring to Fig. 7.
The technical solution of the embodiment of the present invention, by obtaining destination-related information associated with target user；Wherein, mesh
Marking related information includes the essential information of target user, the behavioral data of user and third party's data related to user；Base
In at least two target prediction models that preparatory training obtains, probability value corresponding with destination-related information is determined respectively；Its
In, at least two target prediction models include the first prediction model and the second prediction model；According to probability value, obtains and can dominate
Take in it is corresponding dominate probability value, and according to the disposable income that can dominate probability value and determine target user, solve existing
There is the disposable income that target user can not be determined in technology, realize convenient, efficient prediction target user dominates receipts
Enter, and then the behavior of user can be better anticipated in terms of air control, to reduce the technical effect of risk.
Example IV
Fig. 8 is a kind of structural schematic diagram of the prediction meanss for disposable income that the embodiment of the present invention four provides；Such as Fig. 8 institute
Show, described device includes: that target user's related information obtains module 810, probability value determining module 820 and disposable income
Determining module 830.
Wherein, target user's related information obtains module 810, for obtaining target association letter associated with target user
Breath；Wherein, the destination-related information include the essential information of target user, user behavioral data and with user's phase
Third party's data of pass；Probability value determining module 820, at least two target prediction models for being obtained based on preparatory training,
Probability value corresponding with the destination-related information is determined respectively；Wherein, at least two target predictions model includes the
One prediction model and the second prediction model；Disposable income determining module 830, for obtain it is corresponding with disposable income can
Probability value is dominated, and according to the disposable income for dominating probability value and determining target user.
The technical solution of the embodiment of the present invention, by obtaining destination-related information associated with target user；Wherein, mesh
Marking related information includes the essential information of target user, the behavioral data of user and third party's data related to user；Base
In at least two target prediction models that preparatory training obtains, probability value corresponding with destination-related information is determined respectively；Its
In, at least two target prediction models include the first prediction model and the second prediction model；According to probability value, obtains and can dominate
Take in it is corresponding dominate probability value, and according to the disposable income that can dominate probability value and determine target user, solve existing
There is the disposable income that target user can not be determined in technology, realize convenient, efficient prediction target user dominates receipts
Enter, and then reduces the technical effect of risk.
Based on the above technical solution, described device further includes the first prediction model determining module: described first is pre-
Surveying model determining module includes: first sample data capture unit and the first training unit；
The first sample data capture unit, for obtaining at least three groups of associated datas associated with the user as the
One sample data；Wherein, in the first sample data include first kind data and Second Type data, described first
The quantity of data is greater than the quantity of data in the Second Type data in categorical data；First training unit, for adopting
Decision making algorithm is promoted with gradient to be trained the first sample data, obtains first prediction model；Wherein, described
One prediction model is used to generate the first probability value corresponding with the related information based on the related information.
Based on the above technical solution, described device further includes the second prediction model determining module: described second is pre-
Surveying model determining module includes: the second sample data acquiring unit and the second training unit；
The second sample data acquiring unit, for obtaining at least three groups of associated datas associated with the user as the
Two sample datas；It wherein, include first kind data and Second Type data, the first kind in second sample data
The quantity of data is less than the quantity of data in the Second Type data in data；Second training unit, for using ladder
Degree promotes decision making algorithm and is trained to second sample data, obtains second prediction model；Wherein, described second is pre-
Model is surveyed to be used to generate the second probability value corresponding with the related information based on the related information.
It include that at least one is higher than the first default receipts on the basis of above-mentioned each technical solution, in the first kind data
Enter the subscriber association information of value, Second Type data include the subscriber association information that at least one is lower than the second default Revenue；
Wherein the described first default Revenue is higher than the described second default Revenue.
On the basis of above-mentioned each technical solution, the first prediction model determining module and second prediction model
Determining module is also used to after obtaining first prediction model or second prediction model:
At least one set of test sample data are separately input into first prediction model and second prediction model,
Obtain distribution curve corresponding with the test sample data；
If the distribution curve meets preset condition, first prediction model and second preset model are made
For target prediction model.
On the basis of above-mentioned each technical solution, the disposable income determining module is also used to:
After default value is subtracted second probability value, in addition first probability value obtains median；In described
Between value obtain corresponding with institute's disposable income dominating probability value divided by 2.
The prediction meanss of disposable income provided by the embodiment of the present invention can be performed any embodiment of that present invention and be provided
Disposable income prediction technique, have the corresponding functional module of execution method and beneficial effect.
It is worth noting that, each unit included by above-mentioned apparatus and module are only divided according to function logic
, but be not limited to the above division, as long as corresponding functions can be realized；In addition, the specific name of each functional unit
Title is also only for convenience of distinguishing each other, and is not intended to restrict the invention the protection scope of embodiment.
Embodiment five
Fig. 9 is a kind of structural schematic diagram for server that the embodiment of the present invention five provides.Fig. 9, which is shown, to be suitable for being used to realizing
The block diagram of the exemplary servers 90 of embodiment of the embodiment of the present invention.The server 90 that Fig. 9 is shown is only an example, no
The function and use scope for coping with the embodiment of the present invention bring any restrictions.
As shown in figure 9, server 90 is showed in the form of general-purpose computations server.The component of server 90 may include but
Be not limited to: one or more processor or processing unit 901, system storage 902, connect different system components (including
System storage 902 and processing unit 901) bus 903.
Bus 903 indicates one of a few class bus structures or a variety of, including memory bus or Memory Controller,
Peripheral bus, graphics acceleration port, processor or the local bus using any bus structures in a variety of bus structures.It lifts
For example, these architectures include but is not limited to industry standard architecture (ISA) bus, microchannel architecture (MAC)
Bus, enhanced isa bus, Video Electronics Standards Association (VESA) local bus and peripheral component interconnection (PCI) bus.
Server 90 typically comprises a variety of computer system readable media.These media can be and any can be serviced
The usable medium that device 90 accesses, including volatile and non-volatile media, moveable and immovable medium.
System storage 902 may include the computer system readable media of form of volatile memory, such as deposit at random
Access to memory (RAM) 904 and/or cache memory 905.Server 90 may further include it is other it is removable/can not
Mobile, volatile/non-volatile computer system storage medium.Only as an example, storage system 906 can be used for reading and writing not
Movably, non-volatile magnetic media (Fig. 9 do not show, commonly referred to as " hard disk drive ").It, can be with although being not shown in Fig. 9
The disc driver for reading and writing to removable non-volatile magnetic disk (such as " floppy disk ") is provided, and non-volatile to moving
The CD drive of CD (such as CD-ROM, DVD-ROM or other optical mediums) read-write.In these cases, each driving
Device can be connected by one or more data media interfaces with bus 903.Memory 902 may include at least one program
Product, the program product have one group of (for example, at least one) program module, these program modules are configured to perform the present invention
The function of each embodiment.
Program/utility 908 with one group of (at least one) program module 907, can store in such as memory
In 902, such program module 907 includes but is not limited to operating system, one or more application program, other program modules
And program data, it may include the realization of network environment in each of these examples or certain combination.Program module 907
Usually execute the function and/or method in embodiment described in the invention.
Server 90 can also be with one or more external servers 909 (such as keyboard, sensing equipment, display 910
Deng) communication, can also enable a user to the server communication interacted with the server 90 with one or more, and/or with make
Any server that the server 90 can be communicated with one or more of the other calculation server (such as network interface card, modulation /demodulation
Device etc.) communication.This communication can be carried out by input/output (I/O) interface 911.Also, server 90 can also pass through
Network adapter 912 and one or more network (such as local area network (LAN), wide area network (WAN) and/or public network, such as
Internet) communication.As shown, network adapter 912 is communicated by bus 903 with other modules of server 90.It should be bright
It is white, although being not shown in Fig. 9, other hardware and/or software module can be used in conjunction with server 0, including but not limited to: micro- generation
Code, server driver, redundant processing unit, external disk drive array, RAID system, tape drive and data backup
Storage system etc..
Processing unit 901 by the program that is stored in system storage 902 of operation, thereby executing various function application with
And data processing, such as realize the prediction technique of disposable income provided by the embodiment of the present invention.
Embodiment six
The embodiment of the present invention six also provides a kind of storage medium comprising computer executable instructions, and the computer can be held
Row instruction by computer processor when being executed for executing the prediction technique of disposable income.
The described method includes:
Obtain destination-related information associated with target user；Wherein, the destination-related information includes target user
Essential information, user behavioral data and third party's data relevant to the user；
Based on at least two target prediction models that preparatory training obtains, determination is opposite with the destination-related information respectively
The probability value answered；Wherein, at least two target predictions model includes the first prediction model and the second prediction model；
According to the probability value, obtain it is corresponding with disposable income dominate probability value, and dominated according to described
Probability value determines the disposable income of target user.
The computer storage medium of the embodiment of the present invention, can be using any of one or more computer-readable media
Combination.Computer-readable medium can be computer-readable signal media or computer readable storage medium.It is computer-readable
Storage medium for example may be-but not limited to-the system of electricity, magnetic, optical, electromagnetic, infrared ray or semiconductor, device or
Device, or any above combination.The more specific example (non exhaustive list) of computer readable storage medium includes: tool
There are electrical connection, the portable computer diskette, hard disk, random access memory (RAM), read-only memory of one or more conducting wires
(ROM), erasable programmable read only memory (EPROM or flash memory), optical fiber, portable compact disc read-only memory (CD-
ROM), light storage device, magnetic memory device or above-mentioned any appropriate combination.In this document, computer-readable storage
Medium can be any tangible medium for including or store program, which can be commanded execution system, device or device
Using or it is in connection.
Computer-readable signal media may include in a base band or as carrier wave a part propagate data-signal,
Wherein carry computer-readable program code.The data-signal of this propagation can take various forms, including but unlimited
In electromagnetic signal, optical signal or above-mentioned any appropriate combination.Computer-readable signal media can also be that computer can
Any computer-readable medium other than storage medium is read, which can send, propagates or transmit and be used for
By the use of instruction execution system, device or device or program in connection.
The program code for including on computer-readable medium can transmit with any suitable medium, including --- but it is unlimited
In wireless, electric wire, optical cable, RF etc. or above-mentioned any appropriate combination.
Can with one or more programming languages or combinations thereof come write for execute the embodiment of the present invention operation
Computer program code, described program design language include object oriented program language-such as Java,
Smalltalk, C++ further include conventional procedural programming language --- such as " C " language or similar program design language
Speech.Program code can be executed fully on the user computer, partly be executed on the user computer, as an independence
Software package execute, part on the user computer part execute on the remote computer or completely in remote computer or
It is executed on server.In situations involving remote computers, remote computer can pass through the network of any kind --- packet
It includes local area network (LAN) or wide area network (WAN)-is connected to subscriber computer, or, it may be connected to outer computer (such as benefit
It is connected with ISP by internet).
Note that the above is only a better embodiment of the present invention and the applied technical principle.It will be appreciated by those skilled in the art that
The invention is not limited to the specific embodiments described herein, be able to carry out for a person skilled in the art it is various it is apparent variation,
It readjusts and substitutes without departing from protection scope of the present invention.Therefore, although being carried out by above embodiments to the present invention
It is described in further detail, but the present invention is not limited to the above embodiments only, without departing from the inventive concept, also
It may include more other equivalent embodiments, and the scope of the invention is determined by the scope of the appended claims.
Claims (10)
1. a kind of prediction technique of disposable income characterized by comprising
Obtain destination-related information associated with target user；Wherein, the destination-related information includes the base of target user
This information, the behavioral data of user and third party's data relevant to the user；
Based on at least two target prediction models that preparatory training obtains, determination is corresponding with the destination-related information respectively
Probability value；Wherein, at least two target predictions model includes the first prediction model and the second prediction model；
According to the probability value, obtain it is corresponding with disposable income dominate probability value, and dominate probability according to described
It is worth the disposable income for determining target user.
2. the method according to claim 1, wherein further include:
At least three groups of associated datas associated with the user are obtained as first sample data；Wherein, the first sample data
In include first kind data and Second Type data, the quantity of data is greater than described second in the first kind data
The quantity of data in categorical data；
Decision making algorithm is promoted using gradient to be trained the first sample data, obtains first prediction model；
Wherein, first prediction model is used to generate corresponding with the related information first based on the related information general
Rate value.
3. the method according to claim 1, wherein further include:
At least three groups of associated datas associated with the user are obtained as the second sample data；Wherein, second sample data
In include first kind data and Second Type data, the quantity of data is less than the Second Type in the first kind data
The quantity of data in data；
Decision making algorithm is promoted using gradient to be trained second sample data, obtains second prediction model；
Wherein, second prediction model is used to generate corresponding with the related information second based on the related information general
Rate value.
4. according to the method in claim 2 or 3, which is characterized in that include that at least one is high in the first kind data
In the subscriber association information of the first default Revenue, Second Type data include the use that at least one is lower than the second default Revenue
Family related information；Wherein the described first default Revenue is higher than the described second default Revenue.
5. according to the method in claim 2 or 3, which is characterized in that obtaining first prediction model or described second
After prediction model, further includes:
At least one set of test sample data are separately input into first prediction model and second prediction model, are obtained
Distribution curve corresponding with the test sample data；
If the distribution curve meets preset condition, using first prediction model and second preset model as mesh
Mark prediction model.
6. the method according to claim 1, wherein described according to the probability value, obtain and disposable income
It is corresponding to dominate probability value, comprising:
After default value is subtracted the second probability value, in addition the first probability value obtains median；
The median is obtained corresponding with institute's disposable income to dominate probability value divided by 2.
7. a kind of prediction meanss of disposable income characterized by comprising
Target user's related information obtains module, for obtaining destination-related information associated with target user；Wherein, described
Destination-related information includes the essential information of target user, the behavioral data of user and third number formulary relevant to the user
According to；
Probability value determining module, at least two target prediction models for being obtained based on preparatory training, determine respectively with it is described
The corresponding probability value of destination-related information；Wherein, at least two target predictions model includes the first prediction model and the
Two prediction models；
Disposable income determining module, for obtain it is corresponding with disposable income dominate probability value, and according to it is described can
Dominate the disposable income that probability value determines target user.
8. device according to claim 7, which is characterized in that the disposable income determining module is also used to:
After default value is subtracted the second probability value, in addition the first probability value obtains median；
The median is obtained corresponding with institute's disposable income to dominate probability value divided by 2.
9. a kind of server, which is characterized in that the server includes:
One or more processors；
Storage device, for storing one or more programs,
When one or more of programs are executed by one or more of processors, so that one or more of processors are real
Now such as the prediction technique of disposable income as claimed in any one of claims 1 to 6.
10. a kind of storage medium comprising computer executable instructions, the computer executable instructions are by computer disposal
For executing the prediction technique such as disposable income as claimed in any one of claims 1 to 6 when device executes.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910447918.3A CN110197301A (en) | 2019-05-27 | 2019-05-27 | A kind of prediction technique of disposable income, device, server and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910447918.3A CN110197301A (en) | 2019-05-27 | 2019-05-27 | A kind of prediction technique of disposable income, device, server and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110197301A true CN110197301A (en) | 2019-09-03 |
Family
ID=67753255
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910447918.3A Pending CN110197301A (en) | 2019-05-27 | 2019-05-27 | A kind of prediction technique of disposable income, device, server and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110197301A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111383028A (en) * | 2020-03-16 | 2020-07-07 | 支付宝(杭州)信息技术有限公司 | Prediction model training method and device, and prediction method and device |
| CN112668703A (en) * | 2021-01-19 | 2021-04-16 | 上海优扬新媒信息技术有限公司 | User income determination method and device |
| CN113537577A (en) * | 2021-06-25 | 2021-10-22 | 中国铁路广州局集团有限公司 | Revenue prediction method, system, electronic device, and computer-readable storage medium |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1795463A (en) * | 2003-05-22 | 2006-06-28 | 珀欣投资有限责任公司 | Customer revenue prediction method and system |
| US20140257924A1 (en) * | 2013-03-08 | 2014-09-11 | Corelogic Solutions, Llc | Automated rental amount modeling and prediction |
| CN108256681A (en) * | 2018-01-15 | 2018-07-06 | 吉浦斯信息咨询（深圳）有限公司 | A kind of income level Forecasting Methodology, device, storage medium and system |
| CN108629632A (en) * | 2018-05-09 | 2018-10-09 | 北京京东金融科技控股有限公司 | Predict the method, apparatus and computer readable storage medium of user's income |
| CN109034880A (en) * | 2018-07-09 | 2018-12-18 | 麒麟合盛网络技术股份有限公司 | revenue prediction method and device |
| CN109615129A (en) * | 2018-12-05 | 2019-04-12 | 重庆锐云科技有限公司 | Real estate client's conclusion of the business probability forecasting method, server and computer storage medium |
-
2019
- 2019-05-27 CN CN201910447918.3A patent/CN110197301A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1795463A (en) * | 2003-05-22 | 2006-06-28 | 珀欣投资有限责任公司 | Customer revenue prediction method and system |
| US20140257924A1 (en) * | 2013-03-08 | 2014-09-11 | Corelogic Solutions, Llc | Automated rental amount modeling and prediction |
| CN108256681A (en) * | 2018-01-15 | 2018-07-06 | 吉浦斯信息咨询（深圳）有限公司 | A kind of income level Forecasting Methodology, device, storage medium and system |
| CN108629632A (en) * | 2018-05-09 | 2018-10-09 | 北京京东金融科技控股有限公司 | Predict the method, apparatus and computer readable storage medium of user's income |
| CN109034880A (en) * | 2018-07-09 | 2018-12-18 | 麒麟合盛网络技术股份有限公司 | revenue prediction method and device |
| CN109615129A (en) * | 2018-12-05 | 2019-04-12 | 重庆锐云科技有限公司 | Real estate client's conclusion of the business probability forecasting method, server and computer storage medium |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111383028A (en) * | 2020-03-16 | 2020-07-07 | 支付宝(杭州)信息技术有限公司 | Prediction model training method and device, and prediction method and device |
| CN112668703A (en) * | 2021-01-19 | 2021-04-16 | 上海优扬新媒信息技术有限公司 | User income determination method and device |
| CN113537577A (en) * | 2021-06-25 | 2021-10-22 | 中国铁路广州局集团有限公司 | Revenue prediction method, system, electronic device, and computer-readable storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111340616B (en) | Method, device, equipment and medium for approving online loan | |
| CN110197301A (en) | A kind of prediction technique of disposable income, device, server and storage medium | |
| JP6756079B2 (en) | Artificial intelligence-based ternary check method, equipment and computer program | |
| KR102232866B1 (en) | Method for distributing functional element unit work of crowdsourcing based project for artificial intelligence training data generation | |
| CN107506300A (en) | A kind of ui testing method, apparatus, server and storage medium | |
| CN109599095A (en) | A kind of mask method of voice data, device, equipment and computer storage medium | |
| CN109783490A (en) | Data fusion method, device, computer equipment and storage medium | |
| CN112990294B (en) | Training method and device of behavior discrimination model, electronic equipment and storage medium | |
| CN110706096A (en) | Method and device for managing credit line based on salvage-back user and electronic equipment | |
| CN109933738B (en) | Method, device and computer storage medium for displaying reading content | |
| CN109960554B (en) | Method, device and computer storage medium for displaying reading content | |
| CN109784630A (en) | Automatic Pilot level evaluation method, device, computer equipment and storage medium | |
| CN111062770A (en) | Merchant identification method, equipment and computer readable medium | |
| CN107729944B (en) | Identification method and device of popular pictures, server and storage medium | |
| CN113762303B (en) | Image classification method, device, electronic equipment and storage medium | |
| CN108961065A (en) | Virtual currency value estimation method, device, electronic equipment and storage medium | |
| CN109471862A (en) | Data processing method and device, electronic equipment, storage medium | |
| CN111127223B (en) | Insurance product testing method, insurance product testing device and storage medium | |
| CN108399128A (en) | A kind of generation method of user data, device, server and storage medium | |
| CN111681050A (en) | Advertisement pushing method, device, equipment and storage medium | |
| CN116703626A (en) | Cross-border multi-platform-based fund accounting method, device, equipment and storage medium | |
| CN111047207A (en) | Capability level evaluation method, device, equipment and storage medium | |
| CN110333886A (en) | A kind of review procedure iteration update method, device, server and storage medium | |
| CN113516398A (en) | Risk equipment identification method and device based on hierarchical sampling and electronic equipment | |
| CN111859985B (en) | AI customer service model test method and device, electronic equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190903 |
|
| RJ01 | Rejection of invention patent application after publication |