CN107992978A - It is a kind of to net the method for prewarning risk and relevant apparatus for borrowing platform - Google Patents
It is a kind of to net the method for prewarning risk and relevant apparatus for borrowing platform Download PDFInfo
- Publication number
- CN107992978A CN107992978A CN201711396381.XA CN201711396381A CN107992978A CN 107992978 A CN107992978 A CN 107992978A CN 201711396381 A CN201711396381 A CN 201711396381A CN 107992978 A CN107992978 A CN 107992978A
- Authority
- CN
- China
- Prior art keywords
- data
- risk
- platform
- net
- temporal aspect
- 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
- G06Q10/0635—Risk analysis of enterprise or organisation activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/08—Insurance
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Development Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Physics & Mathematics (AREA)
- Marketing (AREA)
- General Physics & Mathematics (AREA)
- Finance (AREA)
- Accounting & Taxation (AREA)
- Game Theory and Decision Science (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Educational Administration (AREA)
- Technology Law (AREA)
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
Abstract
This application discloses a kind of method for prewarning risk for netting loan platform, including：From related web site crawl achievement data, operations risks situation data and social assessment data；Achievement data and social assessment data are handled to obtain temporal aspect, and feature selecting is carried out to temporal aspect and obtains available feature；Machine learning is carried out according to available feature and obtains risk evaluation model；Judge that net to be monitored borrows whether platform is excessive risk state using the risk evaluation model；If so, then carry out early warning operation.Characteristic processing is carried out by the data captured from website to train to obtain risk evaluation model, and the risk change of net loan platform can be learnt in the time scale of bigger by carrying out model training by temporal aspect, the accuracy rate of Risk-warning is improved, avoids loss caused by risk change.Netted disclosed herein as well is one kind and borrow platform Risk-warning device, server and computer-readable recording medium, there is above-mentioned beneficial effect.
Description
Technical field
It is more particularly to a kind of to net method for prewarning risk, the Risk-warning for borrowing platform this application involves field of computer technology
Device, server and computer-readable recording medium.
Background technology
With the fast development of internet, Internet technology and service trade constantly combine, and derive panoramic interconnection
Net service.Wherein, the development of net loan platform is especially prominent, and with the development of mobile Internet, net borrows platform to user
Life bring great convenience.
With the development of industry, rival and number of users constantly rise, and manager borrows the risk of platform for net
Control is just more and more important.In general, in the risk control that net borrows platform, the Risk-warning for the software is most important ring
Section, that is, the precondition of risk control is carried out to software.Generally, the Risk-warning for net loan platform is by artificial
What the data of processing carried out, when the software operation just carries out relevant early warning operation in excessive risk state.But due to resulting
Data for artificial treatment, it is often relatively low with the degree of correlation of risk factor, truth can not be accurately reflected, be unfavorable for into
The relevant analysis of row judges.And by then passing through artificial judgment, the accuracy rate of judgement is relatively low, be easy to cause the situation of erroneous judgement,
Cause to lose a large amount of economic resources, while also to pay great number monitoring cost.
Therefore, how to improve the accuracy rate of Risk-warning is those skilled in the art's Important Problems of interest.
The content of the invention
The purpose of the application be to provide it is a kind of net borrow the method for prewarning risk of platform, Risk-warning device, server and
Computer-readable recording medium, characteristic processing is carried out by the data grabbed from related web site, and training obtains risk assessment
Model, realize to net borrow platform carry out risk assessment, and by temporal aspect carry out model training can bigger when
Between on scale study net borrow the risk change of platform, improve the assessment accuracy rate of risk evaluation model, it is pre- to further improve risk
Alert accuracy rate, avoids loss caused by risk change.
In order to solve the above technical problems, the application provides a kind of method for prewarning risk netted and borrow platform, including：
From internet financial web site and portal website's crawl the relevant achievement data of platform, operations risks situation number are borrowed with net
According to this and social assessment data；
Data characteristics processing is carried out to the achievement data and the social assessment data, obtains temporal aspect data, and
Feature selecting processing is carried out according to the operations risks situation data to the temporal aspect data, obtains available feature data；
Machine learning training managing is carried out according to the available feature data, obtains risk evaluation model；
Judge that net to be monitored borrows whether platform is excessive risk state using the risk evaluation model；
If so, then carry out early warning operation.
Optionally, the relevant achievement data of platform, operation wind are borrowed with net from internet financial web site and portal website's crawl
Dangerous situation condition data and social assessment data, including：
The achievement data and the operations risks situation data are captured from the internet financial web site；Wherein, it is described
Achievement data include data collection time, average expectancy earning rate, investment number, loaning bill number, the average life of loan, treat it is also remaining
Volume, due-in investment number, treat also loaning bill number, borrow money mark number, per capita fund net inflow, borrowing balance, per capita investment amount；
The relevant negative report data of platform are borrowed from portal website crawl and the net, by the negative report data
As social assessment data.
Optionally, data characteristics processing is carried out to the achievement data and the social assessment data, obtains temporal aspect
Data, and feature selecting processing is carried out according to the operations risks situation data to the temporal aspect data, obtain available spy
Data are levied, including：
Frequency statistics is carried out to the negative vocabulary in the social assessment data, selecting frequency is more than the negative of predeterminated frequency
Vocabulary is added to negative antistop list as negative keyword；
Frequency statistics is carried out to the negative keyword of the social assessment data according to the negative antistop list, is born
The face information frequency of occurrences；
The achievement data and the negative report frequency of occurrences are subjected to temporal aspect calculating, it is special to obtain corresponding sequential
Levy data；Wherein, the temporal aspect, which calculates, includes mean value computation, variance calculates and 10 groups of Fourier correlation coefficients calculate；
Feature selecting is carried out according to the operations risks situation data to all temporal aspect data, obtain it is described can
Use characteristic.
Optionally, feature selecting is carried out to all temporal aspect data according to the operations risks situation data, obtained
To the available feature data, including：
Feature choosing is carried out to all temporal aspect data according to the operations risks situation data using mutual information method
Select, obtain the available feature data.
Optionally, feature selecting is carried out to all temporal aspect data according to the operations risks situation data, obtained
To the available feature data, including：
Feature is carried out to all temporal aspect data according to the operations risks situation data using variance back-and-forth method
Selection, obtains the available feature data.
Optionally, feature selecting is carried out to all temporal aspect data according to the operations risks situation data, obtained
To the available feature data, including：
Feature is carried out to all temporal aspect data according to the operations risks situation data using correlation coefficient process
Selection, obtains the available feature data.
Optionally, machine learning training managing is carried out according to the available feature data, obtains risk evaluation model, wrapped
Include：
Declined using gradient and return tree method according to available feature progress machine learning training managing, obtained risk and comment
Estimate model.
The application also provides a kind of Risk-warning device netted and borrow platform, including：
Data acquisition module, for borrowing the relevant index number of platform with net from internet financial web site and portal website's crawl
According to, operations risks situation data and social assessment data；
Characteristic extracting module, for carrying out data characteristics processing to the achievement data and the social assessment data, obtains
Feature selecting processing is carried out according to the operations risks situation data to temporal aspect data, and to the temporal aspect data,
Obtain available feature data；
Model training module, for carrying out machine learning training managing according to the available feature data, obtains risk and comments
Estimate model；
Judgment module, for borrowing social assessment data and the friendship of platform according to net to be monitored using the risk evaluation model
Easy data determine whether excessive risk state；
Warning module, for carrying out early warning operation.
The application also provides a kind of server, including：
Memory, for storing computer program；
The step of processor, for performing computer program when, realize method for prewarning risk as described above.
The application also provides a kind of computer-readable recording medium, and calculating is stored with the computer-readable recording medium
The step of machine program, the computer program realizes method for prewarning risk as described above when being executed by processor.
A kind of net provided herein borrows the method for prewarning risk of platform, including：From internet financial web site and door
Website is captured borrows platform relevant achievement data, operations risks situation data and social assessment data with net；To the index
Data and the social assessment data carry out data characteristics processing, obtain temporal aspect data, and to the temporal aspect data
Feature selecting processing is carried out according to the operations risks situation data, obtains available feature data；According to the available feature number
According to machine learning training managing is carried out, risk evaluation model is obtained；It is flat to judge that net to be monitored is borrowed using the risk evaluation model
Whether platform is excessive risk state；If so, then carry out early warning operation.
Characteristic processing is carried out by the data grabbed from related web site, and training obtains risk evaluation model, realizes
Platform is borrowed to net and carries out risk assessment, and model training is carried out by temporal aspect and can be learnt in the time scale of bigger
Net borrows the risk change of platform, improves the assessment accuracy rate of risk evaluation model, further improves the accuracy rate of Risk-warning, keep away
Loss caused by having exempted from risk change.
The application also provides a kind of net and borrows platform Risk-warning device, server and computer-readable recording medium, tool
There is above-mentioned beneficial effect, this will not be repeated here.
Brief description of the drawings
In order to illustrate the technical solutions in the embodiments of the present application or in the prior art more clearly, below will be to embodiment or existing
There is attached drawing needed in technology description to be briefly described, it should be apparent that, drawings in the following description are only this
The embodiment of application, for those of ordinary skill in the art, without creative efforts, can also basis
The attached drawing of offer obtains other attached drawings.
A kind of net that Fig. 1 is provided by the embodiment of the present application borrows the flow chart of the method for prewarning risk of platform；
The net that Fig. 2 is provided by the embodiment of the present application borrows the flow that the data characteristics in the method for prewarning risk of platform is handled
Figure；
A kind of net that Fig. 3 is provided by the embodiment of the present application borrows the structure diagram of the Risk-warning device of platform.
Embodiment
The core of the application be to provide it is a kind of net borrow the method for prewarning risk of platform, Risk-warning device, server and
Computer-readable recording medium, characteristic processing is carried out by the data grabbed from related web site, and training obtains risk assessment
Model, realize to net borrow platform carry out risk assessment, and by temporal aspect carry out model training can bigger when
Between on scale study net borrow the risk change of platform, improve the assessment accuracy rate of risk evaluation model, it is pre- to further improve risk
Alert accuracy rate, avoids loss caused by risk change.
To make the purpose, technical scheme and advantage of the embodiment of the present application clearer, below in conjunction with the embodiment of the present application
In attached drawing, the technical solution in the embodiment of the present application is clearly and completely described, it is clear that described embodiment is
Some embodiments of the present application, instead of all the embodiments.Based on the embodiment in the application, those of ordinary skill in the art
All other embodiments obtained without making creative work, shall fall in the protection scope of this application.
Please refer to Fig.1, a kind of net that Fig. 1 is provided by the embodiment of the present application borrows the flow of the method for prewarning risk of platform
Figure.
The present embodiment provides a kind of net to borrow the method for prewarning risk of platform, can improve the accuracy rate of Risk-warning, the party
Method can include：
S101, the relevant achievement data of platform, operations risks are borrowed from internet financial web site and portal website's crawl with net
Situation data and social assessment data；
This step is intended to borrow the relevant data of platform with net from the related web site crawl in internet, due to present techniques
Scheme is mainly that the carry out Risk-warning of platform is borrowed to net, but if carrying out data grabber in the four corner of internet,
Time and resource are expended, therefore the website of crawl is limited between internet financial web site and portal website by this step, can be with
Representative data can be quickly grabbed under limited time and resource, meet the data of subsequent treatment requirement.
Wherein, internet financial web site can include net loans home Web site, the data which is provided more it is accurate with
In time, and portal website can then include each network forum and news website, specifically from which website crawl data at this
Do not limited in embodiment, since the website data amount of internet is larger, multiple websites can be selected to carry out grabbing for related data
Take, do not limit in the present embodiment.
Wherein, achievement data refers to the data being quantized on transaction, number of users etc., for example, net borrows platform
Transaction data.Operations risks situation data refer to that the net borrows operation data when high risk condition occurs in platform, wherein, Gao Feng
Dangerous situation condition refers to that net borrows the situation that platform operation is on the verge of bankruptcy in bankruptcy, verge of bankruptcy etc., its main function is to reflect some
Net borrows whether platform is in excessive risk state.Social assessment data refer to the public opinion data of society, including popular public opinion data
With the public opinion data of media.
Optionally, this step can include：
Achievement data and operations risks situation data are captured from internet financial web site；Wherein, achievement data includes data
Acquisition time, average expectancy earning rate, investment number, loaning bill number, the average life of loan, remaining sum to be gone back, due-in investment number,
Treat also loaning bill number, borrow money mark number, per capita fund net inflow, borrowing balance, per capita investment amount；
Captured from portal website and borrow the relevant negative report data of platform with netting, using negative report data as social assessment
Data.
Achievement data and social assessment data are carried out data characteristics processing by S102, obtain temporal aspect data, and pair when
Sequence characteristics data carry out feature selecting processing according to operations risks situation data, obtain available feature data；
On the basis of step S101, the data that this step is intended to obtain previous step carry out data characteristics processing, then
Feature selecting processing is carried out, obtains the available feature data that machine learning can use.
Wherein, data characteristics processing is carried out to the data of acquisition, is principally obtaining corresponding temporal aspect data.Sequential is special
It is mainly to reflect the data of data variation within a period of time to levy data, therefore temporal aspect data can be in the time ruler of bigger
Observed data feature on degree, the further noise reduced in the data obtained and interference data, it is anti-to improve temporal aspect data
Reflect the accuracy rate that net borrows platform risk situation.
Wherein, the relevant treatment specifically carried out can be carried out according to different application environments and the requirement for expecting data
Selection, this will not be repeated here.
The item number of data that the temporal aspect data obtained at this time can be obtained more and it is different, it may be possible to multinomial sequential
Characteristic, it is therefore desirable to corresponding characteristic selection is carried out to this feature data and is handled, the selection processing of this feature data needs
To be handled according to the operations risks situation data of acquisition.Exist for example, operations risks situation data reflect net loan platform
Enter excessive risk state in certain time period, therefore the net is just borrowed into platform in the feature for changing the change in the period, as
Available feature data.
S103, carries out machine learning training managing according to available feature data, obtains risk evaluation model；
On the basis of step S102, this step is intended to carry out machine learning training managing according to available feature data, obtains
To corresponding risk evaluation model.
Optionally, it can be declined using gradient and return tree method according to available feature progress machine learning training managing, obtained
To risk evaluation model.
S104, judges that net to be monitored borrows whether platform is excessive risk state using the risk evaluation model；；
On the basis of step S103, this step be intended to using risk evaluation model judge it is to be monitored net borrow platform whether
Excessive risk state.
Wherein, specific determination methods can be that the social assessment number of platform is borrowed according to risk evaluation model and net to be monitored
According to and transaction data, assessed to obtain the risk status that the net borrows platform, judge risk status whether in excessive risk state.
S105, if so, then carrying out early warning operation.
On the basis of step S104, this step, which is intended to borrow when platform is in excessive risk state when net, carries out early warning operation.
The present embodiment provides a kind of method for prewarning risk for netting loan platform, passes through the data grabbed from related web site and carries out
Characteristic processing, and training obtains risk evaluation model, realizes and platform progress risk assessment is borrowed to net, and pass through temporal aspect
The risk change that net borrows platform can be learnt in the time scale of bigger by carrying out model training, improve commenting for risk evaluation model
Estimate accuracy rate, further improve the accuracy rate of Risk-warning, avoid loss caused by risk change.
Please refer to Fig.2, the net that Fig. 2 is provided by the embodiment of the present application borrows the data characteristics in the method for prewarning risk of platform
The flow chart of processing.
Based on a upper embodiment, the present embodiment handles what is done primarily directed to how to carry out data characteristics in a upper embodiment
One explanation, other parts are substantially the same with a upper embodiment, and same section may be referred to an embodiment, not do herein superfluous
State.
The present embodiment can include：
S201, carries out the negative vocabulary in social assessment data frequency statistics, and selecting frequency is more than the negative of predeterminated frequency
Face vocabulary is added to negative antistop list as negative keyword；
This step is intended to choose the negative keyword that the frequency of occurrences reaches pre- measured frequency, and forms and obtain negative keyword
Table.
Wherein, predeterminated frequency can be set according to the demand of specific applicable cases either user, in this reality
Apply predeterminated frequency in example and can be set as 10, other situations can set other data, and this will not be repeated here.
Wherein, the negative antistop list that this step obtains is the set containing multiple negative keywords.
S202, carries out frequency statistics to the negative keyword of social assessment data according to negative antistop list, obtains negative
The information frequency of occurrences；
On the basis of step S201, this step is intended to be commented according to negative all societies of antistop list statistics obtained above
There is the frequency of negative keyword in valence mumber, obtains the negative report frequency of occurrences.
Wherein, frequency statistics can be unit at a time interval, therefore when this step can also count each default
Between section the negative keyword of appearance frequency.Preset time period could be provided as daily, may be set to be weekly, can also set
Every month is set to, should specifically regard specific application environment and the demand reselection of user, this will not be repeated here.
S101 and S102 is the equal of that social assessment data are carried out data quantization, makes social assessment data and These parameters
Data show in the same manner, and then can be carried out at the same time processing.
S203, carries out temporal aspect calculating by achievement data and the negative report frequency of occurrences, obtains corresponding temporal aspect
Data；Wherein, temporal aspect, which calculates, includes mean value computation, variance calculates and 10 groups of Fourier correlation coefficients calculate；
On the basis of step S202, this step is intended to achievement data and the negative report frequency of occurrences carrying out sequential spy
Sign calculates, and obtains corresponding temporal aspect data.
Wherein, achievement data may include multinomial data, therefore time series data obtained by calculation also can be therewith
It is changed into multiple temporal aspect data.
Specifically, the temporal aspect conducted in it, which calculates, includes mean value computation, variance calculating and 10 groups of Fourier phases
Relation number calculates.Calculation formula in the present embodiment is as follows：
Wherein, t is number of days, and i is i-th of index, aitThen represent index i in the numerical value of number of days t, uiThen represent average.
Wherein, s then represents the variance of index i.
Wherein, XikThen represent the kth group Fourier correlation coefficient of index i.
The indices data of above-mentioned calculating are average, variance and the 10 groups of Fouriers of the achievement data of one month in the past
Related coefficient, therefore the data amount check of calculating is arranged to 30 in formula.
All temporal aspect data are carried out feature selecting according to operations risks situation data, obtain available feature by S204
Data.
On the basis of step S203, this step is intended to carry out feature selecting according to operations risks situation, obtains available spy
Levy data.
Wherein, feature selecting is link important in machine learning, after the algorithm of machine learning determines, determines identification
The upper limit of the accuracy rate of model is that the feature being trained is determined.Therefore, selected feature determines final result
Quality.And the selection of characteristic is carried out according to operations risks situation data in the present embodiment, it is possible to keep available feature
The quality of data, that is, the quality of the spy's feature improved.
Optionally, the present embodiment can use mutual information method according to the operations risks situation data to all sequential
Characteristic carries out feature selecting, obtains the available feature data.
Specifically, mutual information method is the information gain for calculating each feature in above-mentioned available feature data, further according to information
Gain selects maximally related feature.Therefore, the formula for information gain always being calculated in the present embodiment Huo can be as follows：
Wherein, IG (gk) represent k-th of feature gkInformation gain.p(c0), p (c1) represent to be in excessive risk in data set
The net of state borrows the ratio that platform borrows platform and account for all nets and borrow platform with being not at the net of excessive risk state.p(gk=j) represent the
The ratio that k feature value is j.p(ci|gk=j) represent to borrow the feature g of platform in all netskValue is place in the data of j
Platform is borrowed in the net of excessive risk state and is not at the ratio for netting loan platform and accounting for all nets and borrowing platform of excessive risk state.
Optionally, the present embodiment can also use variance back-and-forth method according to the operations risks situation data to all described
Temporal aspect data carry out feature selecting, obtain the available feature data.
Optionally, the present embodiment can also use correlation coefficient process according to the operations risks situation data to all described
Temporal aspect data carry out feature selecting, obtain the available feature data.
The embodiment of the present application provides a kind of method for prewarning risk netted and borrow platform, can be by being grabbed from related web site
Data carry out characteristic processing, and training obtains risk evaluation model, realizes and borrows platform to net and carry out risk assessment, and leads to
The risk change that net borrows platform can be learnt in the time scale of bigger by crossing temporal aspect progress model training, improved risk and commented
Estimate the assessment accuracy rate of model, further improve the accuracy rate of Risk-warning, avoid loss caused by risk change.
A kind of Risk-warning device for netting loan platform provided by the embodiments of the present application is introduced below, it is described below
A kind of Risk-warning device for netting loan platform can correspond ginseng with a kind of above-described method for prewarning risk for netting loan platform
According to.
Please refer to Fig.3, Fig. 3 is shown by the structure for the Risk-warning device that a kind of net that the embodiment of the present application provides borrows platform
It is intended to.
The present embodiment provides a kind of Risk-warning device for netting loan platform, which can include：
Data acquisition module 100, for borrowing the relevant finger of platform with net from internet financial web site and portal website's crawl
Mark data, operations risks situation data and social assessment data；
Characteristic extracting module 200, for carrying out data characteristics processing to achievement data and social assessment data, obtains sequential
Characteristic, and feature selecting processing is carried out according to operations risks situation data to temporal aspect data, obtain available feature number
According to；
Model training module 300, for carrying out machine learning training managing according to available feature data, obtains risk assessment
Model；
Judgment module 400, judges that net to be monitored borrows whether platform is excessive risk state using the risk evaluation model；
Warning module 500, for carrying out early warning operation.
The embodiment of the present application also provides a kind of server, can include：
Memory, for storing computer program；
Processor, the step of method for prewarning risk such as above-described embodiment is realized during for performing computer program.
The embodiment of the present application also provides a kind of computer-readable recording medium, and meter is stored with computer-readable recording medium
Calculation machine program, the step of method for prewarning risk such as above-described embodiment can be realized when computer program is executed by processor.
Each embodiment is described by the way of progressive in specification, and what each embodiment stressed is and other realities
Apply the difference of example, between each embodiment identical similar portion mutually referring to.For device disclosed in embodiment
Speech, since it is corresponded to the methods disclosed in the examples, so description is fairly simple, related part is referring to method part illustration
.
Professional further appreciates that, with reference to each exemplary unit of the embodiments described herein description
And algorithm steps, can be realized with electronic hardware, computer software or the combination of the two, in order to clearly demonstrate hardware and
The interchangeability of software, generally describes each exemplary composition and step according to function in the above description.These
Function is performed with hardware or software mode actually, application-specific and design constraint depending on technical solution.Specialty
Technical staff can realize described function to each specific application using distinct methods, but this realization should not
Think to exceed scope of the present application.
Can directly it be held with reference to the step of method or algorithm that the embodiments described herein describes with hardware, processor
Capable software module, or the two combination are implemented.Software module can be placed in random access memory (RAM), memory, read-only deposit
Reservoir (ROM), electrically programmable ROM, electrically erasable ROM, register, hard disk, moveable magnetic disc, CD-ROM or technology
In any other form of storage medium well known in field.
A kind of net provided herein is borrowed above the method for prewarning risk, Risk-warning device, server of platform with
And computer-readable recording medium is described in detail.Used herein principle and embodiment party of the specific case to the application
Formula is set forth, and the explanation of above example is only intended to help to understand the present processes and its core concept.It should refer to
Go out, for those skilled in the art, can also be to the application on the premise of the application principle is not departed from
Some improvement and modification are carried out, these are improved and modification is also fallen into the application scope of the claims.
Claims (10)
1. a kind of net the method for prewarning risk for borrowing platform, it is characterised in that including：
From internet financial web site and portal website's crawl and net borrow the relevant achievement data of platform, operations risks situation data with
And social assessment data；
Data characteristics processing is carried out to the achievement data and the social assessment data, obtains temporal aspect data, and to institute
State temporal aspect data and carry out feature selecting processing according to the operations risks situation data, obtain available feature data；
Machine learning training managing is carried out according to the available feature data, obtains risk evaluation model；
Judge that net to be monitored borrows whether platform is excessive risk state using the risk evaluation model；
If so, then carry out early warning operation.
2. method for prewarning risk according to claim 1, it is characterised in that grabbed from internet financial web site and portal website
Take and borrow platform relevant achievement data, operations risks situation data and social assessment data with net, including：
The achievement data and the operations risks situation data are captured from the internet financial web site；Wherein, the index
Data include data collection time, average expectancy earning rate, investment number, loaning bill number, the average life of loan, remaining sum to be gone back,
Due-in investment number, treat also loaning bill number, borrow money mark number, per capita fund net inflow, borrowing balance, per capita investment amount；
From the portal website crawl with it is described net borrow the relevant negative report data of platform, using the negative report data as
Social assessment data.
3. method for prewarning risk according to claim 2, it is characterised in that to the achievement data and the social assessment
Data carry out data characteristics processing, obtain temporal aspect data, and to the temporal aspect data according to the operations risks feelings
Condition data carry out feature selecting processing, obtain available feature data, including：
Frequency statistics is carried out to the negative vocabulary in the social assessment data, selecting frequency is more than the negative vocabulary of predeterminated frequency
As negative keyword, and it is added to negative antistop list；
Frequency statistics is carried out to the negative keyword of the social assessment data according to the negative antistop list, is negatively believed
Cease the frequency of occurrences；
The achievement data and the negative report frequency of occurrences are subjected to temporal aspect calculating, obtain corresponding temporal aspect number
According to；Wherein, the temporal aspect, which calculates, includes mean value computation, variance calculates and 10 groups of Fourier correlation coefficients calculate；
Feature selectings are carried out to all temporal aspect data according to the operations risks situation data, obtain described to use spy
Levy data.
4. method for prewarning risk according to claim 3, it is characterised in that according to the operations risks situation data to institute
There are the temporal aspect data to carry out feature selecting, obtain the available feature data, including：
Feature selecting is carried out to all temporal aspect data according to the operations risks situation data using mutual information method, is obtained
To the available feature data.
5. method for prewarning risk according to claim 3, it is characterised in that according to the operations risks situation data to institute
There are the temporal aspect data to carry out feature selecting, obtain the available feature data, including：
Feature selecting is carried out to all temporal aspect data according to the operations risks situation data using variance back-and-forth method,
Obtain the available feature data.
6. method for prewarning risk according to claim 3, it is characterised in that according to the operations risks situation data to institute
There are the temporal aspect data to carry out feature selecting, obtain the available feature data, including：
Feature selecting is carried out to all temporal aspect data according to the operations risks situation data using correlation coefficient process,
Obtain the available feature data.
7. method for prewarning risk according to claim 1, it is characterised in that carry out machine according to the available feature data
Learning training processing, obtains risk evaluation model, including：
Declined using gradient and return tree method according to available feature progress machine learning training managing, obtain risk assessment mould
Type.
8. a kind of net the Risk-warning device for borrowing platform, it is characterised in that including：
Data acquisition module, for from internet financial web site and portal website's crawl and net borrow the relevant achievement data of platform,
Operations risks situation data and social assessment data；
Characteristic extracting module, for carrying out data characteristics processing to the achievement data and the social assessment data, when obtaining
Sequence characteristics data, and feature selecting processing is carried out according to the operations risks situation data to the temporal aspect data, obtain
Available feature data；
Model training module, for carrying out machine learning training managing according to the available feature data, obtains risk assessment mould
Type；
Judgment module, judges that net to be monitored borrows whether platform is excessive risk state using the risk evaluation model；
Warning module, for carrying out early warning operation.
A kind of 9. server, it is characterised in that including：
Memory, for storing computer program；
Processor, such as claim 1 to 7 any one of them method for prewarning risk is realized during for performing the computer program
The step of.
10. a kind of computer-readable recording medium, it is characterised in that be stored with computer on the computer-readable recording medium
Program, is realized such as claim 1 to 7 any one of them method for prewarning risk when the computer program is executed by processor
Step.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711396381.XA CN107992978A (en) | 2017-12-21 | 2017-12-21 | It is a kind of to net the method for prewarning risk and relevant apparatus for borrowing platform |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711396381.XA CN107992978A (en) | 2017-12-21 | 2017-12-21 | It is a kind of to net the method for prewarning risk and relevant apparatus for borrowing platform |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107992978A true CN107992978A (en) | 2018-05-04 |
Family
ID=62038189
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711396381.XA Pending CN107992978A (en) | 2017-12-21 | 2017-12-21 | It is a kind of to net the method for prewarning risk and relevant apparatus for borrowing platform |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107992978A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109410543A (en) * | 2018-10-17 | 2019-03-01 | 深圳壹账通智能科技有限公司 | Early warning test control method, device, computer equipment and storage medium |
| CN111078978A (en) * | 2019-11-29 | 2020-04-28 | 上海观安信息技术股份有限公司 | Web credit website entity identification method and system based on website text content |
| CN111953794A (en) * | 2020-08-20 | 2020-11-17 | 深圳市富之富信息科技有限公司 | Group cheating and lending early warning method and device |
| CN112507269A (en) * | 2020-12-10 | 2021-03-16 | 中国农业科学院农业信息研究所 | Website background risk assessment system |
| CN113450116A (en) * | 2020-03-24 | 2021-09-28 | 国家计算机网络与信息安全管理中心 | Transaction risk analysis method and device, electronic equipment and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104636447A (en) * | 2015-01-21 | 2015-05-20 | 上海天呈医流科技股份有限公司 | Intelligent evaluation method and system for medical instrument B2B website users |
| CN105894369A (en) * | 2015-01-04 | 2016-08-24 | 必普(上海)金融信息服务有限公司 | Risk assessment model for internet financial network loan platform |
| CN106570631A (en) * | 2016-10-28 | 2017-04-19 | 南京邮电大学 | Method and system of facing P2P platform operation risk estimation |
-
2017
- 2017-12-21 CN CN201711396381.XA patent/CN107992978A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105894369A (en) * | 2015-01-04 | 2016-08-24 | 必普(上海)金融信息服务有限公司 | Risk assessment model for internet financial network loan platform |
| CN104636447A (en) * | 2015-01-21 | 2015-05-20 | 上海天呈医流科技股份有限公司 | Intelligent evaluation method and system for medical instrument B2B website users |
| CN106570631A (en) * | 2016-10-28 | 2017-04-19 | 南京邮电大学 | Method and system of facing P2P platform operation risk estimation |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109410543A (en) * | 2018-10-17 | 2019-03-01 | 深圳壹账通智能科技有限公司 | Early warning test control method, device, computer equipment and storage medium |
| CN109410543B (en) * | 2018-10-17 | 2022-09-09 | 深圳壹账通智能科技有限公司 | Early warning test control method and device, computer equipment and storage medium |
| CN111078978A (en) * | 2019-11-29 | 2020-04-28 | 上海观安信息技术股份有限公司 | Web credit website entity identification method and system based on website text content |
| CN111078978B (en) * | 2019-11-29 | 2024-02-27 | 上海观安信息技术股份有限公司 | Network credit website entity identification method and system based on website text content |
| CN113450116A (en) * | 2020-03-24 | 2021-09-28 | 国家计算机网络与信息安全管理中心 | Transaction risk analysis method and device, electronic equipment and storage medium |
| CN111953794A (en) * | 2020-08-20 | 2020-11-17 | 深圳市富之富信息科技有限公司 | Group cheating and lending early warning method and device |
| CN112507269A (en) * | 2020-12-10 | 2021-03-16 | 中国农业科学院农业信息研究所 | Website background risk assessment system |
| CN112507269B (en) * | 2020-12-10 | 2023-08-08 | 中国农业科学院农业信息研究所 | Website background risk assessment system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107992978A (en) | It is a kind of to net the method for prewarning risk and relevant apparatus for borrowing platform | |
| CN109670940A (en) | Credit Risk Assessment Model generation method and relevant device based on machine learning | |
| CN109784779B (en) | Financial risk prediction method, device and storage medium | |
| CN108763277A (en) | A kind of data analysing method, computer readable storage medium and terminal device | |
| CN107437198A (en) | Determine method, information recommendation method and the device of consumer's risk preference | |
| CN107203939A (en) | Determine method and device, the computer equipment of consumer's risk grade | |
| CN108898476A (en) | A kind of loan customer credit-graded approach and device | |
| CN106384197A (en) | Service quality evaluation method and device based on big data | |
| CN107784519A (en) | A kind of advertising results appraisal procedure and server | |
| CN104616198A (en) | P2P (peer-to-peer) network lending risk prediction system based on text analysis | |
| CN107563621A (en) | A kind of website user's wastage analysis method and device | |
| CN106570631B (en) | P2P platform-oriented operation risk assessment method and system | |
| CN113554350A (en) | Activity evaluation method and apparatus, electronic device and computer readable storage medium | |
| CN112581291A (en) | Risk assessment transaction detection method, device, equipment and storage medium | |
| CN115423499A (en) | Model training method, price prediction method, terminal device, and storage medium | |
| CN107464019A (en) | A kind of financial events method for early warning | |
| CN115203496A (en) | Project intelligent prediction and evaluation method and system based on big data and readable storage medium | |
| CN113822160A (en) | Evaluation method, system and equipment of deep forgery detection model | |
| CN106874286A (en) | A kind of method and device for screening user characteristics | |
| CN117764617A (en) | Property running cost prediction method and device based on big data analysis | |
| CN109814976A (en) | A kind of functional module arrangement method and device | |
| CN115760363A (en) | Interest rate measuring and calculating method and device based on pedestrian credit report | |
| CN108038085A (en) | The data calibration method and device of real-time task | |
| CN104050595B (en) | A kind of apparatus and method for selecting flow to supply according to the quality of flow supplier | |
| CN115018577A (en) | Information recommendation method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180504 |