CN105335886A - Method and device for processing financial data - Google Patents
Method and device for processing financial data Download PDFInfo
- Publication number
- CN105335886A CN105335886A CN201410228281.6A CN201410228281A CN105335886A CN 105335886 A CN105335886 A CN 105335886A CN 201410228281 A CN201410228281 A CN 201410228281A CN 105335886 A CN105335886 A CN 105335886A
- Authority
- CN
- China
- Prior art keywords
- user
- moment
- random forest
- forest model
- 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
Landscapes
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
Abstract
The embodiments of the invention provide a method and device for processing financial data. The method comprises the following steps: determining a target user set from a first user set according to financial behavior data of users in the first user set; determining financial behavior data of each user in the target user set and maximum net asset data of first time; and training a first random forest model by use of the financial behavior data of all the users in the target user set and the maximum net asset data of the first time. The first random forest model determined by use of the technical scheme provided by the invention can be applied to prediction of financial assets of the users in a time period in future rather than mere estimation of current financial assets of the users. The first random forest model is determined based on different training samples, thereby being applied to the users with different characteristics for estimation of financial assets. In such a way, the financial asset estimation efficiency can be improved, and the accuracy is quite high.
Description
Technical field
The embodiment of the present invention relates to technical field of data processing, and more specifically, relates to the method and apparatus of processing financial data.
Background technology
The assets of the frequent Water demand of bank or other financial institutions to user are assessed.Under normal circumstances, the information of user can comprise personal attribute information, such as age, sex, occupation etc.The information of user can also comprise data of financial transaction, such as, transfer accounts, take in, pay, loan etc.
The method of existing appraise assets uses artificial rule how much to assess the assets of user.Such as, for a certain user, the maximum deposit in this user 3 months in the past can be added up, in 3 months past maximum single proceed to the amount of money and pass by 3 months on average wait pay out wages, go out the current possible financial asset of this user according to above-mentioned data estimation.
The factor using artificial rule estimation financial asset to consider is too simple, and ratio of precision is poor, can not be suitable for the feature of different user.The not easily Adjustable calculation rule when the self-condition of outside environmental elements and user changes.In addition, artificial rule only can assess the current possible total assets of user, cannot predict the total assets in this user following a period of time.
Summary of the invention
The embodiment of the present invention provides the method and apparatus of processing financial data, can improve accuracy rate and the efficiency of assessment user financial asset, and can predict the financial asset of this user in future.
First aspect, the embodiment of the present invention provides a kind of method of processing financial data, and the method comprises: according to the banking operation data of the user in first user set, from this first user set, determine that targeted customer gathers; Determine the banking operation data of each user in this targeted customer set and the maximum net asset data in the first moment; The banking operation data of all users in using this targeted customer to gather and the maximum net asset data in this first moment train the first Random Forest model, to use this first Random Forest model to determine the maximum net asset data of any instant of any one user after this first moment in this first user set.
In conjunction with first aspect, in the implementation that the first is possible, these banking operation data comprise following one or more: bank card business dealing record, credit card trade record, finance product transaction record, real estate valuation, loan data.
In conjunction with first aspect or the first possible implementation, in the implementation that the second is possible, this is according to the banking operation data of the user in this first user set, determine the user in this targeted customer set, comprise: according to the banking operation data of the user in this first user set, from this first user set, select user form this targeted customer set, at least two kinds of users that user wherein in this targeted customer set is any one user following or belongs in following multiple user simultaneously: any active ues, stablize user and monthly mortgage user, wherein this any active ues is in first time period, on average proceed to the user of fund higher than the first thresholding, the fund that produces within the second time period is this stable user to the number of times of account of the same name not belonging to this mechanism belonging to first user set lower than the user of the second thresholding, this monthly mortgage user is the user having real estate monthly mortgage to provide a loan.
In conjunction with the implementation that the second is possible, in the implementation that the third is possible, this maximum net asset data is by the difference that maximum total assets and the monthly mortgage of very first time grain size statistics are provided a loan in this moment in the first moment to the second, wherein these maximum total assets are by the maximum fund numerical value of this very first time grain size statistics and real estate valuation in this first moment to this second moment, this second time be engraved in this first moment before.
In conjunction with first aspect or any one possible implementation above-mentioned, in the 4th kind of possible implementation, the method also comprises: use this first Random Forest model to determine the financial asset of any instant of any one user after this first moment in this first user set.
In conjunction with first aspect or any one possible implementation above-mentioned, in the 5th kind of possible implementation, the method also comprises: determine to upgrade user from this targeted customer set, wherein this renewal user is different at the maximum net asset data in this first moment from this renewal user at the maximum net asset data in the 3rd moment, this first time be engraved in for the 3rd moment before; The decision tree corresponding to this renewal user is determined and the decision tree of deleting from this first Random Forest model corresponding to this renewal user from this first Random Forest model; The maximum net asset data in the banking operation data of this renewal user and the 3rd moment is used to make training second Random Forest model; According to this first Random Forest model and this second Random Forest model, determine to upgrade Random Forest model, to use this renewal Random Forest model to determine the maximum net asset data of any instant of any one user after the 3rd moment in this first user set.
Second aspect, the embodiment of the present invention provides a kind of equipment, and this equipment comprises: the first determining unit, for the banking operation data according to the user in first user set, from this first user set, determines that targeted customer gathers; Second determining unit, for determine this targeted customer set in the banking operation data of each user and the maximum net asset data in the first moment; 3rd determining unit, in this targeted customer set determined for using this second determining unit, the banking operation data of all users and the maximum net asset data in this first moment train the first Random Forest model, to use this first Random Forest model to determine the maximum net asset data of any instant of any one user after this first moment in this first user set.
In conjunction with second aspect, in the implementation that the first is possible, this first determining unit, specifically for the banking operation data according to the user in this first user set, from this first user set, select user form this targeted customer set, user wherein in this targeted customer set is any one or at least two kinds of users simultaneously belonging in following multiple user: any active ues, stablize user and monthly mortgage user, wherein this any active ues is in first time period, on average proceed to the user of fund higher than the first thresholding, the fund that produces within the second time period is this stable user to the number of times of account of the same name not belonging to this mechanism belonging to first user set lower than the user of the second thresholding, this monthly mortgage user is the user having real estate monthly mortgage to provide a loan.
In conjunction with second aspect or the first possible implementation, in the implementation that the second is possible, this equipment also comprises: the 4th determining unit, for the financial asset using this first Random Forest model to determine any instant of any one user after this first moment in this first user set.
In conjunction with second aspect or any one possible implementation above-mentioned, in the implementation that the third is possible, this equipment also comprises: the 5th determining unit, for determining to upgrade user from this targeted customer set, wherein this renewal user is different at the maximum net asset data in this first moment from this renewal user at the maximum net asset data in the 3rd moment, this first time be engraved in for the 3rd moment before; 3rd determining unit, also for determining to correspond to the decision tree of this renewal user and delete from this first Random Forest model and correspond to the decision tree of this renewal user from this first Random Forest model; 3rd determining unit, also for using the maximum net asset data of the banking operation data of this renewal user and the 3rd moment to train the second Random Forest model; 3rd determining unit, also for according to this first Random Forest model and this second Random Forest model, determine to upgrade Random Forest model, to use this renewal Random Forest model to determine the maximum net asset data of any instant of any one user after the 3rd moment in this first user set.
The first Random Forest model that technique scheme is determined may be used for predicting the financial asset of user in following a period of time, instead of only estimates the financial asset that user is current.This first Random Forest model determines based on different training samples.Therefore, this first Random Forest model goes for having the user of different characteristics to estimate financial asset.The efficiency of estimation financial asset can be improved like this and there is higher accuracy rate.
Accompanying drawing explanation
In order to be illustrated more clearly in the technical scheme of the embodiment of the present invention, be briefly described to the accompanying drawing used required in the embodiment of the present invention below, apparently, accompanying drawing described is below only some embodiments of the present invention, for those of ordinary skill in the art, under the prerequisite not paying creative work, other accompanying drawing can also be obtained according to these accompanying drawings.
Fig. 1 is the indicative flowchart of the method for the processing financial data provided according to the embodiment of the present invention.
Fig. 2 is the indicative flowchart of the method according to processing financial data provided by the invention.
Fig. 3 is the structured flowchart of the equipment provided according to the embodiment of the present invention.
Fig. 4 is the structured flowchart of the equipment provided according to the embodiment of the present invention.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, be clearly and completely described the technical scheme in the embodiment of the present invention, obviously, described embodiment is a part of embodiment of the present invention, instead of whole embodiment.Based on the embodiment in the present invention, the every other embodiment that those of ordinary skill in the art obtain under the prerequisite not making creative work, all should belong to the scope of protection of the invention.
Fig. 1 is the indicative flowchart of the method for the processing financial data provided according to the embodiment of the present invention.
101, according to the banking operation data of the user in first user set, from this first user set, determine that targeted customer gathers.
102, determine the banking operation data of each user in this targeted customer set and the maximum net asset data in the first moment.
103, the banking operation data of all users and the maximum net asset data in this first moment in this target data set is used to train the first Random Forest model, to use this first Random Forest model to determine the maximum net asset data of any instant of any one user after this first moment in this first user set.
Method shown in Fig. 1 can by first Random Forest model of selected suitable training sample training for determining user's assets.Like this, this first Random Forest model can be used to predict the financial asset of user, instead of only estimate the financial asset that user is current.This first Random Forest model determines based on different training samples.Therefore, this first Random Forest model goes for having the user of different characteristics to estimate financial asset.The efficiency of estimation financial asset can be improved like this and there is higher accuracy rate.
Concrete, this first user set can be all or part of user of bank or other financial institutions.First user set can only be made up of the identifier of user, then searches the specifying information of this user according to the identifier of this user, and this specifying information comprises the banking operation data of user and the personal information of this user.This identifier can be the passport NO. of user or other can distinguish the identifier etc. of different user.This personal information can comprise the sex, work unit, telephone number etc. of user.These banking operation data can be following any one or more: transaction record, real estate valuation, loan data etc.This transaction record can comprise following any one or more: bank card business dealing record, credit card trade record, finance product transaction record etc.
Because the user in first user set may exist the user of some inactive users or instability, therefore need from this first user set, screen the stable user of a part of revenue and expenditure and form targeted customer's set, the user in using this targeted customer to gather is as sample training first Random Forest model.The principle of the user in screening targeted customer set can be consider the banking operation data of user.
Particularly, this is according to the banking operation data of the user in this first user set, determine the user in this goal set, comprise: according to the banking operation data of the user in this first user set, from this first user set, select user form this goal set, at least two kinds of users that user wherein in this goal set is any one user in following multiple user or belongs in following multiple user simultaneously: any active ues, stablize user and monthly mortgage user, wherein this any active ues is in first time period, on average proceed to the user of fund higher than the first thresholding, the fund that produces within the second time period is this stable user to the number of times of account of the same name not belonging to this mechanism belonging to first user set lower than the user of the second thresholding, this monthly mortgage user is the user having real estate monthly mortgage to provide a loan.
Specifically, the user in this first user set is not applicable to train the first Random Forest model for estimating user's financial asset.Such as, the fund in certain user may be remain unchanged for a long period of time or only have few change.The banking operation data of these users are not just suitable for trains the first Random Forest model as sample.Therefore these users can be excluded from targeted customer's set.The fund that on average proceeds in this first time period can to refer within 1 year average monthly proceed to fund, also can refer within half a year average monthly proceed to fund, or within multiple moon average monthly proceed to fund.First time period can adjust as required.In addition, on average proceed to fund and on average can proceed to fund based on the second time granularity, this second granularity can be one month, one season, several weeks etc.Can be such as on average monthly proceed to fund, also can be average each season proceed to fund, or average several days or a few week proceed to fund.The cycle that calculating on average proceeds to fund can be set as required.For another example, the account of certain user may often change, such as, in oneself account under one's name that the fund in account is often produced other financial institutions (such as bank) by this user.The fluctuation of the banking operation data of these users is too frequent, is not therefore also suitable for as sample to train this first Random Forest model.So also these users can be excluded from targeted customer's set.The number of times producing fund in second time period can refer to and also can refer to the number of times of the fund produced within half a year by the number of times of the fund produced within a year, or the number of times of the fund produced within multiple moon.Second time period can adjust as required.And for example, may there are the factors of instability in the certain user in first user set, and typical case causes unstable reason to be exactly that these users may apply for that large loan buys real estate or other assets.This can cause the fund of these users and average trading volume to have greatly changed.Therefore, also these users can be got rid of outside this targeted customer set.In the case, can select those apply for loan buy real estate and the user carrying out monthly mortgage as the user in this targeted customer set.Because the revenue and expenditure of these users is all comparatively steady, be convenient to train reliable and stable first Random Forest model.In sum, any active ues in this first user set can form this targeted customer set, or, stable user during this first user combines can form this targeted customer set, or, monthly mortgage user in this first user set can form this targeted customer set, or, the user simultaneously belonging to this any active ues and this stable user in this first user set can form this targeted customer set, or, the user simultaneously belonging to this any active ues and this monthly mortgage user in this first user set can form this targeted customer set, or, the user simultaneously belonging to this stable user and this monthly mortgage user in this first user set can form this targeted customer set, or, belong to this any active ues in this first user set simultaneously, the user of this stable user and this monthly mortgage user forms this targeted customer set.Certainly, those skilled in the art can also design other filtering rules, to select suitable user as the user in this targeted customer set from this first user set, so that the banking operation data of the user in using this targeted customer to gather and maximum net asset data train this first Random Forest model.
This maximum net asset data can be by the difference that maximum total assets and the monthly mortgage of very first time grain size statistics are provided a loan in this moment in the first moment to the second, wherein these maximum total assets are maximum fund numerical value and the real estate valuation of this very first time grain size statistics of case in this first moment to this second moment, this second time be engraved in this first moment before.This second moment and this very first time granularity can be arranged as required.Such as, maximum fund numerical value in two years and real estate valuation can monthly be added up.In the case, the period in this first moment to this second moment is 2 years, and this very first time granularity is one month.Or, the maximum fund numerical value in 3 years and real estate valuation can be added up quarterly.In the case, the period in this first moment to this second moment is 3 years, and this very first time granularity is a season.This maximum fund numerical value is added up and is obtained from these banking operation data.Such as, if these banking operation data only comprise bank card business dealing record, so this maximum fund numerical value is exactly the numerical value that in very first time granularity, bank card remaining sum is the highest.And for example, if these banking operation data are bank card business dealing record and finance product transaction record, then this maximum fund numerical value is exactly bank card remaining sum and the maximum numerical value of finance product market value sum in very first time granularity.
Further, after determining this first Random Forest model, this first Random Forest model can be used to determine the financial asset of any instant of any one user after this first moment in this first user set.This financial asset is the maximum net asset data in this moment.
Particularly, the first Random Forest model may be used for the maximum net asset data determining any instant of this user after this first moment according to the banking operation data of user.That is, this first Random Forest model may be used for the financial asset predicting user, this first Random Forest model is for predicting that the prerequisite of the financial asset of user is the banking operation data obtaining this user, and the target of prediction of this first Random Forest model is exactly this user maximum net asset data at a time.This first Random Forest model may be used for predicting that the reason of maximum net asset data of user obtains because this first Random Forest model is the banking operation data in the past of user according to targeted customer's set and maximum net asset data training in the past.
Further, can also the method shown in Fig. 1 can also comprise: determine to upgrade user from this targeted customer set, wherein this renewal be used for different at the maximum net asset data in this first moment from this renewal user at the maximum net asset data in the 3rd moment, this first time be engraved in for the 3rd moment before; The decision tree corresponding to this renewal user is determined and the decision tree of deleting from this first probabilistic model user should be upgraded from this first Random Forest model; The maximum net asset data in the banking operation data of this renewal user and the 3rd moment is used to obtain the second Random Forest model as sample; According to this first Random Forest model and this second Random Forest model, determine to upgrade Random Forest model, to use the financial asset of this renewal Random Forest model to the user after the 3rd moment to estimate.
Below in conjunction with specific embodiment, the present invention is described further, it is to be appreciated that the following examples are only better understand the present invention to help, and is not limitation of the present invention.
Fig. 2 is the indicative flowchart of the method according to processing financial data provided by the invention.
201, obtain the data of the user in first user set.
Particularly, first user set always total M user and can this M user K from January, 1970 banking operation data of individual month be known is supposed.Obtain the banking operation data in this M user K month.The jth of i-th user in this M user from January, 1970 the banking operation data of individual month can form the proper vector of a N dimension, can be designated as
Wherein R
nrepresent the set of the free N dimensional feature vector (i.e. the banking operation data in all months) of M user's statistics.
202, from this first user set, filter out targeted customer's set.
Such as, this targeted customer set belongs to any active ues simultaneously, stablizes user and monthly mortgage user.Total T user in this targeted customer set.In this T user, each user has a corresponding maximum net asset data, and i-th user in T user can be designated as V at this maximum net asset data of jth month
i j, i ∈ T, j={0,1,2 ..., K}.
203, train the first Random Forest model.
This T user is known from the jth banking operation data of individual month to jth+K month.This T user is also known from the jth maximum net asset data of individual month to jth+K month.Therefore, can train this first Random Forest model according to the banking operation data of the j+n of all users in this T user month and the maximum net asset data of jth+a month, wherein n is less than the positive integer of a and j+a≤K.
Each decision tree in this first random forest corresponds to banking operation data and the maximum net asset data of a user in this T user.Therefore, for each decision tree in this first Random Forest model, the training sample set of this decision tree of structure can be recorded.
204, use this first Random Forest model, determine any one user maximum net asset data at a time belonged in this M user, wherein this moment is any one moment after K month of in January, 1970.
Particularly, when determining the maximum net asset data of this user in this moment, need these banking operation data to be before that moment input in this first Random Forest model.The result exported is exactly the maximum net asset data in this moment of prediction.
Further, if the data of a user in this T user upgrade, then find the decision tree corresponding to this user in this first Random Forest model, from this first forest model, delete the decision tree corresponding to this user.The data using this user to upgrade train separately a new Random Forest model, and this Random Forest model can be called the second Random Forest model.This second Random Forest model and this first Random Forest model is used to determine to upgrade Random Forest model.The financial asset of the Random Forest model after upgrading to the user after renewal is used to predict.
Fig. 3 is the structured flowchart of the equipment provided according to the embodiment of the present invention.Equipment shown in Fig. 3 can perform each function of Fig. 1 or Fig. 2.As shown in Figure 3, equipment 300 comprises the first determining unit 301, second determining unit 302 and the 3rd determining unit 303.
First determining unit 301, for the banking operation data according to the user in first user set, determines that targeted customer gathers from this first user set.
Second determining unit 302, for determine this targeted customer set in the banking operation data of each user and the maximum net asset data in the first moment.
3rd determining unit 303, in this targeted customer set determined for using the second determining unit 302, the banking operation data of all users and the maximum net asset data in this first moment train the first Random Forest model, to use this first Random Forest model to determine the maximum net asset data of any instant of any one user after this first moment in this first user set.
Equipment 300 shown in Fig. 3 can by first Random Forest model of selected suitable training sample training for determining user's assets.Like this, this first Random Forest model can be used to predict the financial asset of user, instead of only estimate the financial asset that user is current.This first Random Forest model determines based on different training samples.Therefore, this first Random Forest model goes for having the user of different characteristics to estimate financial asset.The efficiency of estimation financial asset can be improved like this and there is higher accuracy rate.
These banking operation data can be following any one or more: transaction record, real estate valuation, loan data etc.This transaction record can comprise following any one or more: bank card business dealing record, credit card trade record, finance product transaction record etc.
This maximum net asset data can be by the difference that maximum total assets and the monthly mortgage of very first time grain size statistics are provided a loan in this moment in the first moment to the second, wherein these maximum total assets are maximum fund numerical value and the real estate valuation of this very first time grain size statistics of case in this first moment to this second moment, this second time be engraved in this first moment before.This second moment and this very first time granularity can be arranged as required.
First determining unit 301, specifically for the banking operation data according to the user in this first user set, from this first user set, select user form this targeted customer set, user wherein in this targeted customer set is any one or at least two kinds of users simultaneously belonging in following multiple user: any active ues, stablize user and monthly mortgage user, wherein this any active ues is in first time period, on average proceed to the user of fund higher than the first thresholding, the fund that produces within the second time period is this stable user to the number of times of account of the same name not belonging to this mechanism belonging to first user set lower than the user of the second thresholding, this monthly mortgage user is the user having real estate monthly mortgage to provide a loan.
Further, equipment 300 also comprises: the 4th determining unit 304, for the financial asset using this first Random Forest model to determine any instant of any one user after this first moment in this first user set.
Further, equipment 300 also comprises: the 5th determining unit 305, for determining to upgrade user from this targeted customer set, wherein this renewal user is different at the maximum net asset data in this first moment from this renewal user at the maximum net asset data in the 3rd moment, this first time be engraved in for the 3rd moment before.In the case, the 3rd determining unit 303, also for determining to correspond to the decision tree of this renewal user and delete from this first Random Forest model and correspond to the decision tree of this renewal user from this first Random Forest model.3rd determining unit 303, also for using the maximum net asset data in the banking operation data of this renewal user and the 3rd moment to train the second Random Forest model.3rd determining unit 303, also for according to the first Random Forest model and the second Random Forest model, determine to upgrade Random Forest model, to use this renewal Random Forest model to determine the maximum net asset data of any instant of any one user after the 3rd moment in this first user set.
Fig. 4 is the structured flowchart of the equipment provided according to the embodiment of the present invention.Equipment shown in Fig. 4 can perform each step of Fig. 1 or Fig. 2.Equipment 400 shown in Fig. 4 comprises: processor 401 and storer 402.
Each assembly in equipment 400 is coupled by bus system 403, and wherein bus system 403 is except comprising data bus, also comprises power bus, control bus and status signal bus in addition.But for the purpose of clearly demonstrating, in the diagram various bus is all designated as bus system 403.
The method that the invention described above embodiment discloses can be applied in processor 401, or is realized by processor 401.Processor 401 may be a kind of integrated circuit (IC) chip, has the processing power of signal.In implementation procedure, each step of said method can be completed by the instruction of the integrated logic circuit of the hardware in processor 401 or software form.Above-mentioned processor 401 can be general processor, digital signal processor (DigitalSignalProcessor, DSP), special IC (ApplicationSpecificIntegratedCircuit, ASIC), ready-made programmable gate array (FieldProgrammableGateArray, FPGA) or other programmable logic device (PLD), discrete gate or transistor logic, discrete hardware components.Can realize or perform disclosed each method, step and the logic diagram in the embodiment of the present invention.The processor etc. of general processor can be microprocessor or this processor also can be any routine.Step in conjunction with the method disclosed in the embodiment of the present invention directly can be presented as that hardware decoding processor is complete, or combines complete by the hardware in decoding processor and software module.Software module can be positioned at random access memory (RandomAccessMemory, in the storage medium of RAM), this area maturation such as flash memory, ROM (read-only memory) (Read-OnlyMemory, ROM), programmable read only memory or electrically erasable programmable storer, register.This storage medium is positioned at storer 402, and processor 401 reads the instruction in storer 402, completes the step of said method in conjunction with its hardware.
Processor 401, for the banking operation data according to the user in first user set, determines that targeted customer gathers from this first user set.
Processor 401, for determine this targeted customer set in the banking operation data of each user and the maximum net asset data in the first moment.
Processor 401, in this targeted customer set determined for making purpose processor 401, the banking operation data of all users and the maximum net asset data in this first moment train the first Random Forest model, to use this first Random Forest model to determine the maximum net asset data of any instant of any one user after this first moment in this first user set.
Equipment 400 shown in Fig. 4 can by first Random Forest model of selected suitable training sample training for determining user's assets.Like this, this first Random Forest model can be used to predict the financial asset of user, instead of only estimate the financial asset that user is current.This first Random Forest model determines based on different training samples.Therefore, this first Random Forest model goes for having the user of different characteristics to estimate financial asset.The efficiency of estimation financial asset can be improved like this and there is higher accuracy rate.
These banking operation data can be following any one or more: transaction record, real estate valuation, loan data etc.This transaction record can comprise following any one or more: bank card business dealing record, credit card trade record, finance product transaction record etc.
This maximum net asset data can be by the difference that maximum total assets and the monthly mortgage of very first time grain size statistics are provided a loan in this moment in the first moment to the second, wherein these maximum total assets are maximum fund numerical value and the real estate valuation of this very first time grain size statistics of case in this first moment to this second moment, this second time be engraved in this first moment before.This second moment and this very first time granularity can be arranged as required.
Processor 401, specifically for the banking operation data according to the user in this first user set, from this first user set, select user form this targeted customer set, user wherein in this targeted customer set is any one or at least two kinds of users simultaneously belonging in following multiple user: any active ues, stablize user and monthly mortgage user, wherein this any active ues is in first time period, on average proceed to the user of fund higher than the first thresholding, the fund that produces within the second time period is this stable user to the number of times of account of the same name not belonging to this mechanism belonging to first user set lower than the user of the second thresholding, this monthly mortgage user is the user having real estate monthly mortgage to provide a loan.
Further, processor 401, can also be used for using this first Random Forest model to determine the financial asset of any instant of any one user after this first moment in this first user set.
Further, processor 401, can also be used for from this targeted customer set determine upgrade user, wherein this renewal user is different at the maximum net asset data in this first moment from this renewal user at the maximum net asset data in the 3rd moment, this first time be engraved in for the 3rd moment before.In the case, processor 401, also for determining to correspond to the decision tree of this renewal user and delete from this first Random Forest model and correspond to the decision tree of this renewal user from this first Random Forest model.Processor 401, also for using the maximum net asset data in the banking operation data of this renewal user and the 3rd moment to train the second Random Forest model.Processor 401, also for according to the first Random Forest model and the second Random Forest model, determine to upgrade Random Forest model, to use this renewal Random Forest model to determine the maximum net asset data of any instant of any one user after the 3rd moment in this first user set.
Those of ordinary skill in the art can recognize, in conjunction with unit and the algorithm steps of each example of embodiment disclosed herein description, can realize with the combination of electronic hardware or computer software and electronic hardware.These functions perform with hardware or software mode actually, depend on application-specific and the design constraint of technical scheme.Professional and technical personnel can use distinct methods to realize described function to each specifically should being used for, but this realization should not thought and exceeds scope of the present invention.
Those skilled in the art can be well understood to, and for convenience and simplicity of description, the specific works process of the system of foregoing description, device and unit, with reference to the corresponding process in preceding method embodiment, can not repeat them here.
In several embodiments that the application provides, should be understood that disclosed system, apparatus and method can realize by another way.Such as, device embodiment described above is only schematic, such as, the division of described unit, be only a kind of logic function to divide, actual can have other dividing mode when realizing, such as multiple unit or assembly can in conjunction with or another system can be integrated into, or some features can be ignored, or do not perform.Another point, shown or discussed coupling each other or direct-coupling or communication connection can be by some interfaces, and the indirect coupling of device or unit or communication connection can be electrical, machinery or other form.
The described unit illustrated as separating component or can may not be and physically separates, and the parts as unit display can be or may not be physical location, namely can be positioned at a place, or also can be distributed in multiple network element.Some or all of unit wherein can be selected according to the actual needs to realize the object of the present embodiment scheme.
In addition, each functional unit in each embodiment of the present invention can be integrated in a processing unit, also can be that the independent physics of unit exists, also can two or more unit in a unit integrated.
If described function using the form of SFU software functional unit realize and as independently production marketing or use time, can be stored in a computer read/write memory medium.Based on such understanding, the part of the part that technical scheme of the present invention contributes to prior art in essence in other words or this technical scheme can embody with the form of software product, this computer software product is stored in a storage medium, comprising some instructions in order to make a computer equipment (can be personal computer, server, or the network equipment etc.) or processor (processor) perform all or part of step of method described in each embodiment of the present invention.And aforesaid storage medium comprises: USB flash disk, portable hard drive, ROM (read-only memory) (ROM, Read-OnlyMemory), random access memory (RAM, RandomAccessMemory), magnetic disc or CD etc. various can be program code stored medium.
The above; be only the specific embodiment of the present invention; but protection scope of the present invention is not limited thereto; anyly be familiar with those skilled in the art in the technical scope that the present invention discloses; the change that can expect easily or replacement; all should be encompassed within protection scope of the present invention, therefore protection scope of the present invention should be as the criterion with the protection domain of claim.
Claims (10)
1. a method for processing financial data, is characterized in that, described method comprises:
According to the banking operation data of the user in first user set, from described first user set, determine that targeted customer gathers;
Determine the banking operation data of each user in described targeted customer set and the maximum net asset data in the first moment;
Banking operation data and the maximum net asset data in described first moment of all users in using described targeted customer to gather train the first Random Forest model, to use described first Random Forest model to determine the maximum net asset data of any instant of any one user after described first moment in described first user set.
2. the method for claim 1, is characterized in that, described banking operation data comprise following one or more: bank card business dealing record, credit card trade record, finance product transaction record, real estate valuation, loan data.
3. method as claimed in claim 1 or 2, is characterized in that, the described banking operation data according to the user in described first user set, determines the user in described targeted customer set, comprising:
According to the banking operation data of the user in described first user set, from described first user set, select user form described targeted customer set, at least two kinds of users that user in wherein said targeted customer's set is any one user following or belongs in following multiple user simultaneously: any active ues, stablize user and monthly mortgage user, wherein said any active ues is in first time period, on average proceed to the user of fund higher than the first thresholding, the fund that produces within the second time period is described stable user to the number of times of account of the same name not belonging to the mechanism belonging to described first user set lower than the user of the second thresholding, described monthly mortgage user is the user having real estate monthly mortgage to provide a loan.
4. method as claimed in claim 3, it is characterized in that, described maximum net asset data is by the difference that maximum total assets and the monthly mortgage of very first time grain size statistics are provided a loan in described first moment in moment to the second, wherein said maximum total assets are by the maximum fund numerical value of described very first time grain size statistics and real estate valuation in described first moment to described second moment, before being engraved in described first moment when described second.
5. the method according to any one of Claims 1-4, is characterized in that, described method also comprises:
Described first Random Forest model is used to determine the financial asset of any instant of any one user after described first moment in described first user set.
6. the method according to any one of claim 1 to 5, is characterized in that, described method also comprises:
Determine to upgrade user from described targeted customer set, wherein said renewal user at the maximum net asset data in the 3rd moment from described to upgrade user different at the maximum net asset data in described first moment, before being engraved in described 3rd moment when described first;
The decision tree corresponding to described renewal user is determined and the decision tree of deleting from described first Random Forest model corresponding to described renewal user from described first Random Forest model;
The banking operation data of described renewal user and the maximum net asset data in described 3rd moment is used to make training second Random Forest model;
According to described first Random Forest model and described second Random Forest model, determine to upgrade Random Forest model, to use described renewal Random Forest model to determine the maximum net asset data of any instant of any one user after described 3rd moment in described first user set.
7. an equipment, is characterized in that, described equipment comprises:
First determining unit, for the banking operation data according to the user in first user set, determines that targeted customer gathers from described first user set;
Second determining unit, for determine described targeted customer set in the banking operation data of each user and the maximum net asset data in the first moment;
3rd determining unit, in the described targeted customer set determined for using described second determining unit, the banking operation data of all users and the maximum net asset data in described first moment train the first Random Forest model, to use described first Random Forest model to determine the maximum net asset data of any instant of any one user after described first moment in described first user set.
8. equipment as claimed in claim 7, it is characterized in that, described first determining unit, specifically for the banking operation data according to the user in described first user set, from described first user set, select user form described targeted customer set, user in wherein said targeted customer set is any one or at least two kinds of users simultaneously belonging in following multiple user: any active ues, stablize user and monthly mortgage user, wherein said any active ues is in first time period, on average proceed to the user of fund higher than the first thresholding, the fund that produces within the second time period is described stable user to the number of times of account of the same name not belonging to the mechanism belonging to described first user set lower than the user of the second thresholding, described monthly mortgage user is the user having real estate monthly mortgage to provide a loan.
9. equipment as claimed in claim 7 or 8, it is characterized in that, described equipment also comprises:
4th determining unit, for the financial asset using described first Random Forest model to determine any instant of any one user after described first moment in described first user set.
10. the equipment according to any one of claim 7 to 9, is characterized in that, described equipment also comprises:
5th determining unit, for determining to upgrade user from described targeted customer's set, wherein said renewal user at the maximum net asset data in the 3rd moment from described to upgrade user different at the maximum net asset data in described first moment, before being engraved in described 3rd moment when described first;
Described 3rd determining unit, also for determine from described first Random Forest model correspond to described renewal user decision tree and from described first Random Forest model delete correspond to described renewal user decision tree;
Described 3rd determining unit, banking operation data and the maximum net asset data in described 3rd moment also for using described renewal user train the second Random Forest model;
Described 3rd determining unit, also for according to described first Random Forest model and described second Random Forest model, determine to upgrade Random Forest model, to use described renewal Random Forest model to determine the maximum net asset data of any instant of any one user after described 3rd moment in described first user set.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410228281.6A CN105335886A (en) | 2014-05-28 | 2014-05-28 | Method and device for processing financial data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410228281.6A CN105335886A (en) | 2014-05-28 | 2014-05-28 | Method and device for processing financial data |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105335886A true CN105335886A (en) | 2016-02-17 |
Family
ID=55286397
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410228281.6A Pending CN105335886A (en) | 2014-05-28 | 2014-05-28 | Method and device for processing financial data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105335886A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106227776A (en) * | 2016-07-18 | 2016-12-14 | 四川君逸数码科技股份有限公司 | A kind of data preprocessing method supporting wisdom finance and device |
| CN107451920A (en) * | 2017-08-16 | 2017-12-08 | 深圳平安讯科技术有限公司 | Calculation server, dynamic income processing method and storage medium |
| CN107622326A (en) * | 2017-09-13 | 2018-01-23 | 阿里巴巴集团控股有限公司 | User's classification, available resources Forecasting Methodology, device and equipment |
| CN109816507A (en) * | 2018-11-08 | 2019-05-28 | 深圳壹账通智能科技有限公司 | Statistical method and device, storage medium, the computer equipment of financial asset data |
| WO2020181798A1 (en) * | 2019-03-12 | 2020-09-17 | 平安普惠企业管理有限公司 | Data processing method and device, server and computer-readable storage medium |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101116099A (en) * | 2004-10-13 | 2008-01-30 | 阿雷斯资产管理有限公司 | Data processing system supporting decisions to accept or reject applications for financial accommodation |
| CN101739648A (en) * | 2008-11-04 | 2010-06-16 | 上海迪哈大计算机科技有限公司 | Control method and control system for monitoring financial assets |
-
2014
- 2014-05-28 CN CN201410228281.6A patent/CN105335886A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101116099A (en) * | 2004-10-13 | 2008-01-30 | 阿雷斯资产管理有限公司 | Data processing system supporting decisions to accept or reject applications for financial accommodation |
| CN101739648A (en) * | 2008-11-04 | 2010-06-16 | 上海迪哈大计算机科技有限公司 | Control method and control system for monitoring financial assets |
Non-Patent Citations (1)
| Title |
|---|
| 曹正凤 等: "使用随机森林算法实现优质股票的选择", 《首都经济贸易大学学报》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106227776A (en) * | 2016-07-18 | 2016-12-14 | 四川君逸数码科技股份有限公司 | A kind of data preprocessing method supporting wisdom finance and device |
| CN107451920A (en) * | 2017-08-16 | 2017-12-08 | 深圳平安讯科技术有限公司 | Calculation server, dynamic income processing method and storage medium |
| CN107622326A (en) * | 2017-09-13 | 2018-01-23 | 阿里巴巴集团控股有限公司 | User's classification, available resources Forecasting Methodology, device and equipment |
| CN107622326B (en) * | 2017-09-13 | 2021-02-09 | 创新先进技术有限公司 | User classification and available resource prediction method, device and equipment |
| CN109816507A (en) * | 2018-11-08 | 2019-05-28 | 深圳壹账通智能科技有限公司 | Statistical method and device, storage medium, the computer equipment of financial asset data |
| WO2020181798A1 (en) * | 2019-03-12 | 2020-09-17 | 平安普惠企业管理有限公司 | Data processing method and device, server and computer-readable storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10360575B2 (en) | Consumer household spend capacity | |
| Asker et al. | Dynamic inputs and resource (mis) allocation | |
| Khan et al. | Credit shocks and aggregate fluctuations in an economy with production heterogeneity | |
| Christoffersen et al. | Backtesting value-at-risk: A duration-based approach | |
| Staum et al. | Systemic risk components in a network model of contagion | |
| US20080221990A1 (en) | Estimating the spend capacity of consumer households | |
| CN105335886A (en) | Method and device for processing financial data | |
| Hainaut | A model for interest rates with clustering effects | |
| Cherubini | Credit valuation adjustment and wrong way risk | |
| CN109740792A (en) | Data predication method, system, terminal and computer storage medium | |
| Carmichael | Modeling default for peer-to-peer loans | |
| D’Innocenzo et al. | Modeling extreme events: time-varying extreme tail shape | |
| CN104751234B (en) | A kind of prediction technique and device of user's assets | |
| CN109767333A (en) | Select based method, device, electronic equipment and computer readable storage medium | |
| Malyscheff et al. | Natural gas storage valuation via least squares Monte Carlo and support vector regression | |
| Raddant et al. | Transitions in the stock markets of the US, UK and Germany | |
| CN112116401A (en) | Pressure testing method, device, equipment and storage medium | |
| Cheng et al. | Fundamentalists in the cryptocurrency markets | |
| CN116800831A (en) | Service data pushing method, device, storage medium and processor | |
| CN108256667A (en) | Asset data processing method, device, storage medium and computer equipment | |
| Rapih | How international capital inflows and domestic financial institutional development affect domestic credit: Evidence from developing countries | |
| Xie et al. | Simulation solution to a two-dimensional mortgage refinancing problem | |
| CN112200389A (en) | Data prediction method, device, equipment and storage medium | |
| Tabash et al. | Role of 2008 financial contagion in effecting the mediating role of stock market indices between the exchange rates and oil prices: Application of the unrestricted VAR | |
| Seijas-Giménez et al. | Financing entrepreneurial activity in Uruguay: time to default in a public microcredit institution |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20160217 |
|
| RJ01 | Rejection of invention patent application after publication |