CN108805689A - A kind of loan risk evaluation control method and device - Google Patents
A kind of loan risk evaluation control method and device Download PDFInfo
- Publication number
- CN108805689A CN108805689A CN201710282369.XA CN201710282369A CN108805689A CN 108805689 A CN108805689 A CN 108805689A CN 201710282369 A CN201710282369 A CN 201710282369A CN 108805689 A CN108805689 A CN 108805689A
- Authority
- CN
- China
- Prior art keywords
- user
- loan application
- behavioral data
- loan
- data sequence
- 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
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/03—Credit; Loans; Processing thereof
Landscapes
- Business, Economics & Management (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Engineering & Computer Science (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- Technology Law (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
Abstract
The present invention relates to field of computer technology more particularly to a kind of loan risk evaluation control method and device, this method to be, receives user on the subscriber terminal, is asked by the loan application that social networking application is sent；Obtain the user in the previous time period for sending the loan application request, be at least partially based on the behavioral data sequence of the social networking application generation；The behavioral data sequence is analyzed, determines the current affective state of the user, according to the current affective state of the user, obtains the reasonable score value of the loan application；And according to the reasonable score value, risk assessment control is carried out to the loan application, in this way, according to behavioral data sequence of the user in social networking application, it can judge the whether reasonable rationality of loan application, and then can effectively control credit risk, credit risk control is more accurate.
Description
Technical field
The present invention relates to field of computer technology more particularly to a kind of loan risk evaluation control method and device.
Background technology
Currently, there is the platform that can much provide a loan with internet, user can submit application to obtain on computer or mobile phone
Loan.For example, internet social networking application, user can not only be exchanged with good friend in social networking application or interactive, Hen Duoshe
It hands over using function of itself also offering a loan, for example, wechat, which provides particle, borrows function, user can be straight using particle loan in wechat
Connect application loan.
While internet lends user and brings convenience, also more risk challenge is brought to lending side.User gets excited
Irrational debt-credit often brings bad credit, causes the loss of lending side.In the prior art, for the wind of the loan in social networking application
Danger assessment, is typically assessed according to the third-party reference of user point, and then decide whether to agree to the loan application of the user.
Loan risk evaluation control is carried out according to the reference of user point, what it is due to reference point reflection is that user is relatively long-term
, the financial status stablized, and cannot effectively judge that user gets excited irrational behavior of lending, can cause, in this way although may
The reference of the user point is relatively high, but its irrational behavior of lending of getting excited still results in the loss of lending side, improves loan
Money risk.
Invention content
The embodiment of the present invention provides a kind of loan risk evaluation control method and device, to solve to be directed to society in the prior art
The problem of handing over the risk assessment of the upper loan of application inaccurate, cannot effectively controlling credit risk.
Specific technical solution provided in an embodiment of the present invention is as follows：
A kind of loan risk evaluation control method, including：
It receives user on the subscriber terminal, is asked by the loan application that social networking application is sent；
Obtain the user in the previous time period for sending the loan application request, be at least partially based on the social activity
Using the behavioral data sequence of generation；
The behavioral data sequence is analyzed, the current affective state of the user is determined, is worked as according to the user
Preceding affective state obtains the reasonable score value of the loan application；And
According to the reasonable score value, risk assessment control is carried out to the loan application.
Preferably, obtain the user in the previous time period for sending the loan application request, be at least partially based on
The behavioral data sequence that the social networking application generates, specifically includes：
At set time intervals, it obtains respectively in the previous time period for sending the loan application request, at least
It is based partially on the behavioral data that the social networking application generates；
It by the behavioral data in each time interval got, arranges sequentially in time, obtains the row of the user
For data sequence.
Preferably, the behavioral data, including following a kind of or arbitrary combination：User-pay information, user's Receiving information,
Accept covert payment information, the rubescent package informatin of user, user's single chat information, user's group chat information, user of user adds group number, user to move back group
Number, user's active plusing good friend number, user receive good friend and invite number, user La Hei good friends number, user that black number, user is drawn to send out friend
The number of turns and user friend's circle and good friend's interaction number.
Preferably, analyzing the behavioral data sequence, specifically include：
Using machine learning model trained in advance, the behavioral data sequence is analyzed；
The training method of the machine learning model is：
According to the history loan application data of the user of acquisition, the sample set of loan application is determined, wherein the history is borrowed
Money request for data includes at least the corresponding returned money data of loan application；
Behavioral data sequence of the corresponding user of each sample in the preset period is obtained respectively；
According to the sample set and the corresponding behavioral data sequence of the sample set, the machine learning model is trained, really
Determine the current affective state of user, and according to the current affective state of user, determines rationally dividing for the loan application prediction of user
Value.
Preferably, further comprising：
Using predetermined period, the sample set of the loan application is updated, and according to newer sample set, update pair
The behavioral data sequence answered；
According to newer sample set and corresponding behavioral data sequence, machine learning model described in re -training.
Preferably, further comprising：
Obtain the reference point of the user；
According to the reasonable score value, risk assessment is carried out to the loan application, is specifically included：
According to the reference of the reasonable score value of the loan application and the user point, the risk etc. of the loan application is determined
Grade；
According to the risk class of the loan application, using preset mode, risk assessment is carried out to the loan application
Control.
A kind of loan risk evaluation control device, including：
Receiving module is asked for receiving user on the subscriber terminal by the loan application that social networking application is sent；
First acquisition module, for obtain the user in the previous time period for sending the loan application request, extremely
It is at least partly based on the behavioral data sequence that the social networking application generates；
Analysis module determines the current affective state of the user, root for analyzing the behavioral data sequence
According to the current affective state of the user, the reasonable score value of the loan application is obtained；And
Control module, for according to the reasonable score value, risk assessment control to be carried out to the loan application.
Preferably, obtain the user in the previous time period for sending the loan application request, be at least partially based on
The behavioral data sequence that the social networking application generates, the first acquisition module are specifically used for：
At set time intervals, it obtains respectively in the previous time period for sending the loan application request, at least
It is based partially on the behavioral data that the social networking application generates；
It by the behavioral data in each time interval got, arranges sequentially in time, obtains the row of the user
For data sequence.
Preferably, the behavioral data, including following a kind of or arbitrary combination：User-pay information, user's Receiving information,
Accept covert payment information, the rubescent package informatin of user, user's single chat information, user's group chat information, user of user adds group number, user to move back group
Number, user's active plusing good friend number, user receive good friend and invite number, user La Hei good friends number, user that black number, user is drawn to send out friend
The number of turns and user friend's circle and good friend's interaction number.
Preferably, analyzing the behavioral data sequence, analysis module is specifically used for：
Using machine learning model trained in advance, the behavioral data sequence is analyzed；
The training method of the machine learning model is：
Determining module is used for the history loan application data of the user according to acquisition, determines the sample set of loan application,
In, the history loan application data include at least the corresponding returned money data of loan application；
Second acquisition module, for obtaining behavior number of the corresponding user of each sample in the preset period respectively
According to sequence；
Training module, for according to the sample set and the corresponding behavioral data sequence of the sample set, the training machine
Device learning model determines the current affective state of user, and according to the current affective state of user, determines the loan application of user
The reasonable score value of prediction.
Preferably, further comprising, update module is used for：
Using predetermined period, the sample set of the loan application is updated, and according to newer sample set, update pair
The behavioral data sequence answered；
According to newer sample set and corresponding behavioral data sequence, machine learning model described in re -training.
Preferably, further comprising：
Third acquisition module, the reference point for obtaining the user；
According to the reasonable score value, risk assessment is carried out to the loan application, control unit is specifically used for：
According to the reference of the reasonable score value of the loan application and the user point, the risk etc. of the loan application is determined
Grade；
According to the risk class of the loan application, using preset mode, risk assessment is carried out to the loan application
Control.
A kind of server, including：
At least one processor, for storing program instruction；
At least one processor, for calling the program instruction stored in the memory, according to the program instruction of acquisition
Execute the loan risk evaluation control method in the embodiment of the present invention.
In the embodiment of the present invention, user is received on the subscriber terminal, asked by the loan application that social networking application is sent；It obtains
It takes the user in the previous time period for sending the loan application request, be at least partially based on what the social networking application generated
Behavioral data sequence；The behavioral data sequence is analyzed, the current affective state of the user is determined, according to the use
The current affective state in family obtains the reasonable score value of the loan application；And according to the reasonable score value, to the loan Shen
Row risk assessment that come in controls, in this way, since the behavioral data sequence of user can reflect the affective state of user, and then reflect
Whether the loan application of user is rationality, therefore, when receiving loan application request, obtains user and sends loan application request
Previous time period in, be at least partially based on social networking application generation behavioral data sequence, analyzed, obtain loan application
Reasonable score value so as to judge the reasonability of loan application, and then can more accurately and efficiently control credit risk.
Description of the drawings
Fig. 1 is the flow chart for the loan risk evaluation control method that the embodiment of the present invention one provides；
Fig. 2 is the implementation procedure flow chart of loan risk evaluation control method provided by Embodiment 2 of the present invention；
Fig. 3 is the server architecture environment schematic that the embodiment of the present invention three provides；
Fig. 4 is the loan risk evaluation controling device structure diagram that the embodiment of the present invention five provides；
Fig. 5 is the server architecture schematic diagram that the embodiment of the present invention six provides；
Fig. 6 is the interface schematic diagram for the loan application process that the embodiment of the present invention seven provides；
Fig. 7 is the user terminal structural schematic diagram that the embodiment of the present invention eight provides.
Specific implementation mode
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation describes, it is clear that described embodiments are only a part of the embodiments of the present invention, is not whole embodiment.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
Embodiment one：
As shown in Figure 1, the loan risk evaluation control method that the embodiment of the present invention one provides, specifically includes following steps：
Step 100：It receives user on the subscriber terminal, is asked by the loan application that social networking application is sent.
Wherein, social networking application is essentially a kind of social platform, for example, various chat applications (Application,
APP)。
In practice, for the loan application in social networking application, due to can be related to more Social behaviors in real time and
Have the characteristics that more convenient, simple and quick, therefore, with traditional loan application there are a great difference, brings more
Risk challenge.And user applies for before loan that behavior in social networking application can usually reflect whether is user this application behavior of lending
Rationality is reasonable, this also directly determines the returned money success rate of the quality and later stage of this loan.For example, if user applies for loan
Preceding frequently active in multiple groups, and have many transmitting-receiving red packets or behavior of transferring accounts again, then this loan application of the user is very
May be irrational, unreasonable.
In the embodiment of the present invention, real-time behavioral data sequence of the user in social networking application is considered, more reasonably and accurately
The risk of loan is assessed, and then is taken appropriate measures, to efficiently control credit risk.
For example, user logs in QQ in mobile phone, when user needs to apply loan, QQ wallets are opened, particle is found and borrows function,
Loaning bill button is clicked, corresponding information is filled in, then submits, user just has sent loan application request by the QQ on mobile phone.
Step 110：Obtain the user in the previous time period for sending the loan application request, be at least partially based on
The behavioral data sequence that the social networking application generates.
When executing step 110, specifically include：
First, at set time intervals, user's time before sending the loan application request is obtained respectively
Duan Zhong, it is at least partially based on the behavioral data that the social networking application generates.
Wherein, in the previous time period for sending loan application request, for example, loan application asks the corresponding application time
It before 24 hours, in the embodiment of the present invention, and is not limited, can be configured according to actual demand.
Wherein, the time interval of setting, for example, every a hour, similarly, for time interval setting not yet
It is limited, can be configured according to actual demand.
Wherein, the behavioral data is the various behaviors that can characterize user feeling of the user in social networking application, example
Such as, Social behaviors, user-pay behavior (for example, going back payment etc. under credit card, line), user's reading behavior etc., the present invention is implemented
In example and it is not limited.
Specifically include following a kind of or arbitrary combination：
A) user-pay information, it may for example comprise payment number, transfer amounts；
B) user's Receiving information, it may for example comprise gathering number, collection amount；
C) user accepts covert payment information, it may for example comprise number of accepting covert payment, the amount of money of accepting covert payment；
D) the rubescent package informatin of user, it may for example comprise give bonus number, and give bonus the amount of money；
E) user's single chat information, it may for example comprise single chat number, messaging number；
F) user's group chat information, it may for example comprise participate in group's number of chat, messaging number；
G) user adds group number；
H) user moves back group number；
I) user's active plusing good friend number；
J) user receives good friend's invitation number；
K) user La Hei good friends number；
L) user is drawn black number；
M) user sends out friend's number of turns；
N) user friend's circle and good friend's interaction number.
Certainly, in present example, the behavioral data of user is not limited in it is above-mentioned several, can be according to actually asking
Situation obtains corresponding behavioral data.
Finally, it by the behavioral data in each time interval got, arranges sequentially in time, obtains the user
Behavioral data sequence.
For example, the loan application corresponding application time is on 2 5th, 2,017 8:00, then as unit of hour, obtain respectively
In passing by 24 hours before the time, i.e., on 2 4th, 2,017 8:00-2017 5 days 2 months 8:Between 00 period, each
The behavioral data of the user, the behavioral data in each hour can form a vector in hour, and sequentially in time
It is arranged, so that it may the sequence vector that the vector to obtain 24 having time sequencings forms.Wherein, the dimension of each vector
Equal to the type of the behavioral data of acquisition.
It is worth noting that in present example, behavioral data is sorted and obtains behavioral data sequence, this behavioral data
Sequence is having time sequencing, rather than only obtains behavioral data, each behavioral data is organized into feature vector,
This is because this can lose all time order and function relation informations according only to behavioral data, and these information are also most important
, for example, a user transfers accounts, number or the amount of money are increasing and smaller and smaller, it may be possible to which antipodal two tendencies are (more
Illustrate that risk is very high come bigger, even if smaller and smaller illustrate that user has impulsion and gradually returns rationality), if only using
Number or the amount of money, this trend information are just lost completely.
In this way, in the embodiment of the present invention, after receiving a loan application request, need first to pull from background data base
All behavioral data sequences of the user in one period of past obtain user's real-time having time sequence in social networking application
The behavioral data of row is more accurate and reliable for the risk assessment control of loan.
Step 120：The behavioral data sequence is analyzed, the current affective state of the user is determined, according to institute
The current affective state of user is stated, the reasonable score value of the loan application is obtained.
When executing step 120, specifically include：
First, based on machine learning model trained in advance, the behavioral data sequence is analyzed.
Wherein, machine learning model trained in advance, for example, supports the deep neural network of time series forecasting analysis
Model, such as long-term memory network (Long Short Term Memory Network, LSTM) model.
In the embodiment of the present invention, for the machine learning model used, and it is not limited, may be used in the prior art
Support the arbitrary model of time series forecasting analysis.
Then, it is determined that the affective state that the user is current obtains the loan according to the current affective state of the user
The reasonable score value of money application.
Specifically, after getting the behavioral data sequence of user, the behavioral data sequence of user is organized into feature vector,
It is input in advance trained machine learning model, machine learning model is according to the pattern information learnt in advance, to this
Behavioral data sequence analyzed, judge the current affective state of user, it is determined whether be rationality state, and then determine this
Whether secondary loan application is that rationality is rational, exports the reasonable score value of this loan application.
Wherein, according to machine learning model trained in advance, process that the behavioral data sequence of user is analyzed
Algorithm is realized, just without describing in detail in the embodiment of the present invention.
Step 130：According to the reasonable score value, risk assessment control is carried out to the loan application.
When executing step 130, following methods can be divided into：
First way：If it is determined that rationally score value is less than predetermined threshold value, then refuse the loan application.
The second way：If it is determined that rationally score value is less than predetermined threshold value, then warning information is sent out, prompts lending side to described
Loan application carries out manual examination and verification.
Certainly, in the embodiment of the present invention, and without being limited to above two mode, other risk assessment can also be used
Control strategy.
In this way, smaller for reasonable score value, i.e. the higher loan application of risk controls plan using corresponding risk assessment
Slightly, it such as is strictly audited or is refused, reduced the loss of lending side, effectively control credit risk.
Preferably, the reference point of user can also be obtained further.
It is specially then：According to the reference of the reasonable score value of the loan application and the user point, the loan Shen is determined
Risk class please；And according to the risk class of the loan application, using preset mode, to the loan application into sector-style
Danger assessment control.
Wherein, according to the reference of the reasonable score value of the loan application and the user point, the loan application is determined
The mode of risk class is：
If reasonable score value is lower, it is determined that the risk class of loan application is higher.
If the reference of user point is higher, it is determined that the risk class of loan application is higher.
For example, the reasonable score value of loan application is less than predetermined threshold value, judges that the risk class of the loan application is higher, then may be used
It directly to refuse the loan application, or is alerted, lending side is prompted to carry out stringent manual examination and verification.
In another example pre-setting the correspondence of reasonable score value and risk class, reasonable score value can be divided into three sections, point
High-risk grade, risk grade and low risk level are not corresponded to, and to different risk class settings accordingly to Shen of providing a loan
Processing mode please, and then risk control is carried out to loan application.
In this way, in the embodiment of the present invention, machine learning model trained according to the behavioral data sequence of user and in advance obtains
The reasonable score value of loan application is taken, reflection is the current state of the corresponding user of this loan application, and the reference of user point
What is reflected is the financial status of the relatively long-term stabilization of user, and the two combines, and reasonability and risk assessment for loan application are more
It is accurate to add, and can effectively control credit risk.
The training method of machine learning model is briefly described below, the training method of machine learning model is：
First, according to the history loan application data of the user of acquisition, the sample set of loan application is determined, wherein described
History loan application data include at least the corresponding returned money data of loan application.
In the embodiment of the present invention, multiple corresponding historical datas of user's loan application can be obtained from database, obtained
The sample of multiple loan applications, what user's application loan was refunded on time after obtaining fund is considered as a good loan application sample
This, whereas if occurring violation of agreement after obtaining fund with application loan is considered as primary bad loan application sample, certain sample
It is more, machine learning model is trained more acurrate.Wherein, good loan application sample indicates that the loan application is reasonable rationality
, risk is relatively low, and bad loan application sample indicates that the loan application is unreasonable, and risk is higher.
Then, behavioral data sequence of the corresponding user of each sample in the preset period is obtained respectively.
Wherein, the mode for specifically obtaining behavioral data sequence is identical with the implementation procedure of above-mentioned steps 110, here
It no longer carries out repeating.
Finally, according to the sample set and the corresponding behavioral data sequence of the sample set, the training machine learning mould
Type determines the current affective state of user, and according to the current affective state of user, determines the conjunction of the loan application prediction of user
Manage score value.
Specifically, sample set and behavioral data sequence are quantified, is then introduced into preset machine learning model, for example,
It in LSTM models, is trained, behavioral data sequence is as list entries, according to the corresponding returned money data of loan application, to row
It is analyzed for data sequence, determines the current affective state of user, and determine that the reasonable of corresponding prediction of loan application is divided
Value, portraying user, this time whether loan application is that rationality is rational.
For example, using LSTM models, input is the vectorial (embodiment of the present invention of the relevant series of features of time series
In, corresponding is exactly the behavioral data sequence vector of user), output is that (present invention is implemented these corresponding target predictions of vector
In example, whether corresponding be exactly to loan application be judgement that rationality is reasonably normally provided a loan), machine learning model can learn
Pattern information in these behavioral data sequences, to make more accurate judgement.
Further, it is also necessary to machine learning model is updated, specially：
First, using predetermined period, the sample set of the loan application is updated, and according to newer sample set,
Update corresponding behavioral data sequence.
In practice, since the period of loan is all relatively long, above-mentioned predetermined period, for example, every half a year or 1 year,
Specifically, it in the embodiment of the present invention and is not limited.When updating sample set, the newly-increased loan phase of the predetermined period can be obtained
Data are closed, oldest loan related data is replaced.
Then, according to newer sample set and corresponding behavioral data sequence, machine learning model described in re -training.
Specifically, sample set and corresponding behavioral data sequence are updated, can obtain new training sample, and then can be with
Using the training method of above-mentioned machine learning model, machine learning model is trained again, the good engineering of re -training
Practise model, so that it may directly to be used in actual scene.
In this way, regularly updating machine learning model, it can improve and whether loan application is managed based on the machine learning model
Property it is rational judge it is more accurate.
Embodiment two：
Further description is made to above-described embodiment using a specific application scenarios below.Referring particularly to Fig. 2
Shown, in the embodiment of the present invention two, the implementation procedure of loan risk evaluation control method is specific as follows：
Step 200：Obtain history loan application data.
Wherein, history loan application data include at least the corresponding returned money data of loan application.
Step 201：Determine sample set.
Specifically, according to the corresponding returned money data of loan application, if judging loan application without promise breaking, it is determined that the loan Shen
Please loan application sample preferably；If judging, loan application has promise breaking, it is determined that the loan application is bad loan application sample.
Step 202：Obtain behavioral data of the corresponding user of loan application in the previous time period of loan application time
Sequence.
For example, as unit of hour, the row before the acquisition loan application corresponding application time in 24 hours this periods
For data sequence, the behavioral data of user is a vector in each hour, then can get 24 having time sequences
Vector.
Step 203：The deep neural network model of time series forecasting analysis is supported in training.
Wherein, in the embodiment of the present invention, be with machine learning model be support time series forecasting analysis depth nerve
For network model, illustrate.
Pair specifically, it is input with the vector of this 24 having time sequences, exports these corresponding target predictions of vector, i.e.,
The rational judgement of loan application.
Step 204：Actual scene application, when receiving loan application request, according to trained support time series forecasting
The deep neural network model of analysis carries out risk assessment control to the loan application.
In this way, can accurately and effectively judge loan application, whether rationality is reasonable, and auxiliary lending side makes decisions, effectively
Control credit risk.
Step 205：New loan application data are added in history loan application data, and regularly update support time series
The deep neural network model of forecast analysis.
Wherein, the loan application in step 204 asks corresponding loan data, subsequently can serve as new loan application number
According to the deep neural network model for updating support time series forecasting analysis.
In the embodiment of the present invention, the real-time behavioral data sequence before acquisition user's loan application in social networking application, root
Carry out training machine learning model according to behavioral data sequence, improves accuracy and effect that machine learning model judges, and then connecing
When receiving loan application request, using machine learning model trained in advance, the loan application received is judged, is determined
Whether rationality is reasonable for the loan application, assesses the risk of the loan application, can help the side's of lending effectively control loan
Risk.
Embodiment three：
In the embodiment of the present invention, as shown in fig.3, in the embodiment of the present invention three, server architecture environment schematic.
In practical application, server is connect with user terminal, and user on the subscriber terminal, is sent by social networking application and provided a loan
Application request.Server receives loan application request, and based on the loan risk evaluation control method in the embodiment of the present invention, right
The loan application carries out risk assessment control, and returns to feedback result to user terminal.
Wherein, user terminal can be any smart machines such as mobile phone, computer, ipad.
Social networking application is also not limited, and can be the social networking application of any function of supporting to provide a loan, such as wechat, QQ, micro-
It wins.
Example IV：
Based on above-described embodiment, in the embodiment of the present invention four, using a concrete application scene, above-described embodiment is carried out
It is simple to introduce.By taking user is by the wechat application loan on mobile phone as an example.
First, user logs in wechat by mobile phone, clicks " wallet ", finds " particle loan is borrowed money ", user clicks " particle loan
Borrow money ", the amount of oneself is obtained, according to the amount of oneself, the amount of the loan is inputted, clicks " borrowing money ".
Then, waiting for server side audit feedback.
Specifically, after server receives loan application, the user is obtained in wechat, is sending the previous of loan application
Behavioral data sequence in period, and according to machine learning model trained in advance, the reasonable score value of the loan application is obtained,
Risk assessment is carried out to the loan application, for example, one threshold value of setting judges the loan when determining that the reasonable score value is less than threshold value
Money application risk class is high, then directly refuses the loan application, and feed back to the user.
Finally, handset Wechat termination receives the feedback result for the loan application.
For example, receiving refusal loan or loan failure news, or the successful message of loan.
Embodiment five：
Based on above-described embodiment, as shown in fig.4, in the embodiment of the present invention five, loan risk evaluation control device, specifically
Including：
Receiving module 40 is asked for receiving user on the subscriber terminal by the loan application that social networking application is sent；
First acquisition module 41, for obtain the user in the previous time period for sending the loan application request,
It is at least partially based on the behavioral data sequence that the social networking application generates；
Analysis module 42 determines the current affective state of the user for analyzing the behavioral data sequence,
According to the current affective state of the user, the reasonable score value of the loan application is obtained；And
Control module 43, for according to the reasonable score value, risk assessment control to be carried out to the loan application.
Preferably, obtain the user in the previous time period for sending the loan application request, be at least partially based on
The behavioral data sequence that the social networking application generates, the first acquisition module 41 are specifically used for：
At set time intervals, it obtains respectively in the previous time period for sending the loan application request, at least
It is based partially on the behavioral data that the social networking application generates；
It by the behavioral data in each time interval got, arranges sequentially in time, obtains the row of the user
For data sequence.
Preferably, the behavioral data, including following a kind of or arbitrary combination：User-pay information, user's Receiving information,
Accept covert payment information, the rubescent package informatin of user, user's single chat information, user's group chat information, user of user adds group number, user to move back group
Number, user's active plusing good friend number, user receive good friend and invite number, user La Hei good friends number, user that black number, user is drawn to send out friend
The number of turns and user friend's circle and good friend's interaction number.
Preferably, analyzing the behavioral data sequence, analysis module 42 is specifically used for：
Using machine learning model trained in advance, the behavioral data sequence is analyzed；
The training method of the machine learning model is：
Determining module 44 is used for the history loan application data of the user according to acquisition, determines the sample set of loan application,
Wherein, the history loan application data include at least the corresponding returned money data of loan application；
Second acquisition module 45, for obtaining behavior of the corresponding user of each sample in the preset period respectively
Data sequence；
Training module 46, described according to the sample set and the corresponding behavioral data sequence of the sample set, training
Machine learning model determines the current affective state of user, and according to the current affective state of user, determines the loan Shen of user
The reasonable score value that please be predict.
Preferably, update module 47 is used for：
Using predetermined period, the sample set of the loan application is updated, and according to newer sample set, update pair
The behavioral data sequence answered；
According to newer sample set and corresponding behavioral data sequence, machine learning model described in re -training.
Preferably, further comprising：
Third acquisition module 48, the reference point for obtaining the user；
According to the reasonable score value, risk assessment is carried out to the loan application, control module 43 is specifically used for：
According to the reference of the reasonable score value of the loan application and the user point, the risk etc. of the loan application is determined
Grade；
According to the risk class of the loan application, using preset mode, risk assessment is carried out to the loan application
Control.
In the embodiment of the present invention, according to the behavioral data sequence of user, training machine learning model so that the machine learning
Model can accurate judgement loan application whether be reasonable with rationality, and then receive loan application request when, according to
The behavioral data sequence of the machine learning model and user determines the reasonable score value of loan application, can effectively control loan wind
Danger.
Embodiment six：
Based on above-described embodiment, as shown in fig.5, in the embodiment of the present invention six, a kind of structural schematic diagram of server.
The embodiment of the present invention six provides a kind of server, which may include 510 (Center of processor
Processing Unit, CPU), memory 520, input equipment 530 and output equipment 540 etc., input equipment 530 may include
Keyboard, mouse, touch screen etc., output equipment 540 may include display equipment, such as liquid crystal display (Liquid Crystal
Display, LCD), cathode-ray tube (Cathode Ray Tube, CRT) etc..
Memory 520 may include read-only memory (ROM) and random access memory (RAM), and be carried to processor 510
For the program instruction and data stored in memory 520.In embodiments of the present invention, memory 520 can be used for storing loan
The program of risk assessment control method.
Processor 510 is by the program instruction for calling memory 520 to store, and processor 510 is for the program according to acquisition
Instruction execution：
It receives user on the subscriber terminal, is asked by the loan application that social networking application is sent；
Obtain the user in the previous time period for sending the loan application request, be at least partially based on the social activity
Using the behavioral data sequence of generation；
The behavioral data sequence is analyzed, the current affective state of the user is determined, is worked as according to the user
Preceding affective state obtains the reasonable score value of the loan application；And
According to the reasonable score value, risk assessment control is carried out to the loan application.
Preferably, obtain the user in the previous time period for sending the loan application request, be at least partially based on
The behavioral data sequence that the social networking application generates, processor 510 are specifically used for：
At set time intervals, it obtains respectively in the previous time period for sending the loan application request, at least
It is based partially on the behavioral data that the social networking application generates；
It by the behavioral data in each time interval got, arranges sequentially in time, obtains the row of the user
For data sequence.
Preferably, the behavioral data, including following a kind of or arbitrary combination：User-pay information, user's Receiving information,
Accept covert payment information, the rubescent package informatin of user, user's single chat information, user's group chat information, user of user adds group number, user to move back group
Number, user's active plusing good friend number, user receive good friend and invite number, user La Hei good friends number, user that black number, user is drawn to send out friend
The number of turns and user friend's circle and good friend's interaction number.
Preferably, analyzing the behavioral data sequence, processor 510 is specifically used for：
Using machine learning model trained in advance, the behavioral data sequence is analyzed；
The training method of the machine learning model is that processor 510 is used for：
According to the history loan application data of the user of acquisition, the sample set of loan application is determined, wherein the history is borrowed
Money request for data includes at least the corresponding returned money data of loan application；
Behavioral data sequence of the corresponding user of each sample in the preset period is obtained respectively；
According to the sample set and the corresponding behavioral data sequence of the sample set, the machine learning model is trained, really
Determine the current affective state of user, and according to the current affective state of user, determines rationally dividing for the loan application prediction of user
Value.
Preferably, processor 510 is further used for：
Using predetermined period, the sample set of the loan application is updated, and according to newer sample set, update pair
The behavioral data sequence answered；
According to newer sample set and corresponding behavioral data sequence, machine learning model described in re -training.
Preferably, processor 510 is further used for：Obtain the reference point of the user；
According to the reasonable score value, risk assessment is carried out to the loan application, processor 510 is specifically used for：
According to the reference of the reasonable score value of the loan application and the user point, the risk etc. of the loan application is determined
Grade；
According to the risk class of the loan application, using preset mode, risk assessment is carried out to the loan application
Control.
In the embodiment of the present invention, user is received on the subscriber terminal, asked by the loan application that social networking application is sent,
Obtain user send loan application request previous time period in, be at least partially based on the social networking application generate behavioral data
Sequence analyzes the behavioral data sequence of user, so as to judge whether the loan application is reasonable rationality, more
Add accurately and reliably, can effectively control credit risk.
Embodiment seven：
Based on above-described embodiment, in the embodiment of the present invention seven, using a concrete application scene, above-described embodiment is carried out
It is simple to introduce.By taking user is by the microblogging application loan on mobile phone as an example, as shown in fig.6, passing through for user micro- on mobile phone
The process schematic of rich application loan.
First, it after user's opening microblogging is applied and logged in, clicks " I ", finds " microblogging wallet ", and click " microblogging money
After packet ", " borrowing money " is found, refering to (1) figure in Fig. 6.
Then, " borrowing money " is clicked, after completing phone number registration, inputs borrowing balance, application is submitted, refering to (2) in Fig. 6
Figure, for example, borrowing balance input by user is 2000.
Then, waiting system server side is audited, after server receives loan audit, based in the embodiments of the present invention
Loan risk evaluation control method, however, it is determined that the reasonable score value of the loan application is smaller, that is, judges the risk of the loan application
It is higher ranked, then it can directly refuse the loan application, in this way, mobile phone microblogging end can receive loan failure news, example
Such as, refering to shown in (3) in Fig. 6, user receives the message of " your loan application is irrational loan, loan failure ".Certainly, such as
Fruit user is bundled with phone number, can also send loan application result to user by short message.
It is worth noting that Fig. 6 is only a kind of example that may be implemented, simultaneously for setting of function button or interface etc.
It is not limited.
Embodiment eight：
As shown in fig.7, in the embodiment of the present invention eight, a kind of structural schematic diagram of user terminal.
The embodiment of the present invention eight provides a kind of user terminal, and user terminal can be but be not limited to mobile phone, tablet computer
Deng.The user terminal may include：Memory 710, input module 720, sending module 730, receiving module 740, output module
750, wireless communication module 760 and processor 770 etc..Specially：
Memory 710 may include read-only memory (ROM) and random access memory (RAM), and be carried to processor 770
For the program instruction and data stored in memory 710, operating system, the application program of user terminal can also be stored
Various data etc. used in (Application, APP) (for example, APP of social networking application), module and user terminal.
Input module 720 may include keyboard, mouse, touch screen etc., for receiving digital, character information input by user
Or touch operation, and the input etc. of key signals related with the user setting of user terminal and function control is generated, for example,
In the embodiment of the present invention, input module 720 can receive the clicking operation that user executes in the social networking application of user terminal, defeated
The amount of the loan etc. entered.
Sending module 730 can provide the interface between user terminal and server, for example, in the embodiment of the present invention, use
In the loan application request for sending user to server.
Receiving module 740 equally provides the interface between user terminal and server, for example, in the embodiment of the present invention, uses
In the auditing result and loan application progress etc. that receive the loan application request that server returns.
Output module 750 may include display module, as liquid crystal display (Liquid Crystal Display, LCD),
Cathode-ray tube (Cathode Ray Tube, CRT) etc., wherein display module is displayed for information input by user
Or it is supplied to the information or various user terminals or the menu of social networking application, user interface etc. of user.For example, the present invention is implemented
In example, it can be used for showing the feedback result of loan application to user.
Wireless communication module 760 includes but not limited to Wireless Fidelity (wireless fidelity, WiFi) module, bluetooth
Module, infrared communication module etc..For example, in the embodiment of the present invention, receiving module 740 and sending module 730 in user terminal, to
Server sends the loan application auditing result that loan application asks and receives server return, is realized by wifi module
Communication between server.
Processor 770 is the control centre of user terminal, utilizes each of various interfaces and the entire user terminal of connection
A part by running or execute the software program and/or module that are stored in memory 710, and calls and is stored in storage
Data in device 710 execute the various functions and processing data of user terminal, to carry out integral monitoring to user terminal.
Certainly, the structure of user terminal shown in fig. 7, only one of which example, may include more than illustrating
Or less component, either combine certain components or different components arrangement.
It should be understood by those skilled in the art that, the embodiment of the present invention can be provided as method, system or computer program
Product.Therefore, complete hardware embodiment, complete software embodiment or reality combining software and hardware aspects can be used in the present invention
Apply the form of example.Moreover, the present invention can be used in one or more wherein include computer usable program code computer
The computer program production implemented in usable storage medium (including but not limited to magnetic disk storage, CD-ROM, optical memory etc.)
The form of product.
The present invention be with reference to according to the method for the embodiment of the present invention, the flow of equipment (system) and computer program product
Figure and/or block diagram describe.It should be understood that can be realized by computer program instructions every first-class in flowchart and/or the block diagram
The combination of flow and/or box in journey and/or box and flowchart and/or the block diagram.These computer programs can be provided
Instruct the processor of all-purpose computer, special purpose computer, Embedded Processor or other programmable data processing devices to produce
A raw machine so that the instruction executed by computer or the processor of other programmable data processing devices is generated for real
The device for the function of being specified in present one flow of flow chart or one box of multiple flows and/or block diagram or multiple boxes.
These computer program instructions, which may also be stored in, can guide computer or other programmable data processing devices with spy
Determine in the computer-readable memory that mode works so that instruction generation stored in the computer readable memory includes referring to
Enable the manufacture of device, the command device realize in one flow of flow chart or multiple flows and/or one box of block diagram or
The function of being specified in multiple boxes.
These computer program instructions also can be loaded onto a computer or other programmable data processing device so that count
Series of operation steps are executed on calculation machine or other programmable devices to generate computer implemented processing, in computer or
The instruction executed on other programmable devices is provided for realizing in one flow of flow chart or multiple flows and/or block diagram one
The step of function of being specified in a box or multiple boxes.
Although preferred embodiments of the present invention have been described, it is created once a person skilled in the art knows basic
Property concept, then additional changes and modifications may be made to these embodiments.So it includes excellent that the following claims are intended to be interpreted as
It selects embodiment and falls into all change and modification of the scope of the invention.
Obviously, those skilled in the art can carry out the embodiment of the present invention various modification and variations without departing from this hair
The spirit and scope of bright embodiment.In this way, if these modifications and variations of the embodiment of the present invention belong to the claims in the present invention
And its within the scope of equivalent technologies, then the present invention is also intended to include these modifications and variations.
Claims (11)
1. a kind of loan risk evaluation control method, which is characterized in that including：
It receives user on the subscriber terminal, is asked by the loan application that social networking application is sent；
Obtain the user in the previous time period for sending the loan application request, be at least partially based on the social networking application
The behavioral data sequence of generation；
The behavioral data sequence is analyzed, determines the current affective state of the user, current according to the user
Affective state obtains the reasonable score value of the loan application；And
According to the reasonable score value, risk assessment control is carried out to the loan application.
2. the method as described in claim 1, which is characterized in that obtain the user before sending the loan application request
In one period, it is at least partially based on the behavioral data sequence that the social networking application generates, is specifically included：
At set time intervals, it obtains respectively in the previous time period for sending the loan application request, at least partly
The behavioral data generated based on the social networking application；
It by the behavioral data in each time interval got, arranges sequentially in time, obtains the behavior number of the user
According to sequence.
3. method as claimed in claim 1 or 2, which is characterized in that the behavioral data sequence is analyzed, it is specific to wrap
It includes：
Using machine learning model trained in advance, the behavioral data sequence is analyzed；
The training method of the machine learning model is：
According to the history loan application data of the user of acquisition, the sample set of loan application is determined, wherein history loan Shen
Please data include at least the corresponding returned money data of loan application；
Behavioral data sequence of the corresponding user of each sample in the preset period is obtained respectively；
According to the sample set and the corresponding behavioral data sequence of the sample set, the training machine learning model is determined and is used
The current affective state in family, and according to the current affective state of user, determine the reasonable score value of the loan application prediction of user.
4. method as claimed in claim 3, which is characterized in that further comprise：
Using predetermined period, the sample set of the loan application is updated, and according to newer sample set, update is corresponding
Behavioral data sequence；
According to newer sample set and corresponding behavioral data sequence, machine learning model described in re -training.
5. the method as described in claim 1, which is characterized in that further comprise：
Obtain the reference point of the user；
According to the reasonable score value, risk assessment is carried out to the loan application, is specifically included：
According to the reference of the reasonable score value of the loan application and the user point, the risk class of the loan application is determined；
According to the risk class of the loan application, using preset mode, risk assessment control is carried out to the loan application.
6. a kind of loan risk evaluation control device, which is characterized in that including：
Receiving module is asked for receiving user on the subscriber terminal by the loan application that social networking application is sent；
First acquisition module, for obtaining the user in the previous time period for sending the loan application request, at least portion
Divide the behavioral data sequence generated based on the social networking application；
Analysis module determines the current affective state of the user, according to institute for analyzing the behavioral data sequence
The current affective state of user is stated, the reasonable score value of the loan application is obtained；And
Control module, for according to the reasonable score value, risk assessment control to be carried out to the loan application.
7. device as claimed in claim 6, which is characterized in that obtain the user before sending the loan application request
In one period, it is at least partially based on the behavioral data sequence that the social networking application generates, the first acquisition module is specifically used for：
At set time intervals, it obtains respectively in the previous time period for sending the loan application request, at least partly
The behavioral data generated based on the social networking application；
It by the behavioral data in each time interval got, arranges sequentially in time, obtains the behavior number of the user
According to sequence.
8. device as claimed in claims 6 or 7, which is characterized in that analyze the behavioral data sequence, analysis module
It is specifically used for：
Using machine learning model trained in advance, the behavioral data sequence is analyzed；
The training method of the machine learning model is：
Determining module is used for the history loan application data of the user according to acquisition, determines the sample set of loan application, wherein
The history loan application data include at least the corresponding returned money data of loan application；
Second acquisition module, for obtaining behavioral data sequence of the corresponding user of each sample in the preset period respectively
Row；
Training module, for according to the sample set and the corresponding behavioral data sequence of the sample set, the training engineering
Model is practised, determines the current affective state of user, and according to the current affective state of user, determine the loan application prediction of user
Reasonable score value.
9. device as claimed in claim 8, which is characterized in that further comprise, update module is used for：
Using predetermined period, the sample set of the loan application is updated, and according to newer sample set, update is corresponding
Behavioral data sequence；
According to newer sample set and corresponding behavioral data sequence, machine learning model described in re -training.
10. device as claimed in claim 6, which is characterized in that further comprise：
Third acquisition module, the reference point for obtaining the user；
According to the reasonable score value, risk assessment is carried out to the loan application, control unit is specifically used for：
According to the reference of the reasonable score value of the loan application and the user point, the risk class of the loan application is determined；
According to the risk class of the loan application, using preset mode, risk assessment control is carried out to the loan application.
11. a kind of server, which is characterized in that including：
At least one processor, for storing program instruction；
At least one processor is executed for calling the program instruction stored in the memory according to the program instruction of acquisition
The claims 1-5 any one of them methods.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710282369.XA CN108805689A (en) | 2017-04-26 | 2017-04-26 | A kind of loan risk evaluation control method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710282369.XA CN108805689A (en) | 2017-04-26 | 2017-04-26 | A kind of loan risk evaluation control method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108805689A true CN108805689A (en) | 2018-11-13 |
Family
ID=64068912
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710282369.XA Pending CN108805689A (en) | 2017-04-26 | 2017-04-26 | A kind of loan risk evaluation control method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108805689A (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109685643A (en) * | 2018-12-13 | 2019-04-26 | 平安科技（深圳）有限公司 | Loan audit risk grade determines method, apparatus, equipment and storage medium |
| CN110348208A (en) * | 2019-06-29 | 2019-10-18 | 上海淇馥信息技术有限公司 | A kind of risk control method based on user behavior and neural network, device and electronic equipment |
| CN110766541A (en) * | 2019-09-25 | 2020-02-07 | 平安科技（深圳）有限公司 | Loan risk assessment method, loan risk assessment device, loan risk assessment equipment and computer-readable storage medium |
| CN111681118A (en) * | 2020-05-29 | 2020-09-18 | 泰康保险集团股份有限公司 | Data processing method and device |
| CN112037039A (en) * | 2020-09-02 | 2020-12-04 | 中国银行股份有限公司 | Loan assessment method and device |
| CN112634025A (en) * | 2020-12-29 | 2021-04-09 | 平安消费金融有限公司 | Wind control rule generation method, device, equipment and computer readable storage medium |
| WO2021068635A1 (en) * | 2019-10-11 | 2021-04-15 | 支付宝(杭州)信息技术有限公司 | Information processing method and apparatus, and electronic device |
| CN112862472A (en) * | 2021-03-15 | 2021-05-28 | 深圳市心版图科技有限公司 | Loan payment system based on business display industry chain and loan payment judgment method |
| CN113129021A (en) * | 2021-05-18 | 2021-07-16 | 中国银行股份有限公司 | Block chain-based method and device for preventing malicious overdraft of credit card |
| CN113240509A (en) * | 2021-05-18 | 2021-08-10 | 重庆邮电大学 | Loan risk assessment method based on multi-source data federal learning |
| CN117172908A (en) * | 2023-09-05 | 2023-12-05 | 中铁商业保理有限公司 | Business loan stage decision-making auxiliary system based on consumption bill analysis |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103345688A (en) * | 2013-07-05 | 2013-10-09 | 倪慎瑜 | Dual-feedback credit assessment system and method based on emotion and credit |
| US20150142446A1 (en) * | 2013-11-21 | 2015-05-21 | Global Analytics, Inc. | Credit Risk Decision Management System And Method Using Voice Analytics |
| CN106056444A (en) * | 2016-05-25 | 2016-10-26 | 腾讯科技（深圳）有限公司 | Data processing method and device |
-
2017
- 2017-04-26 CN CN201710282369.XA patent/CN108805689A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103345688A (en) * | 2013-07-05 | 2013-10-09 | 倪慎瑜 | Dual-feedback credit assessment system and method based on emotion and credit |
| US20150142446A1 (en) * | 2013-11-21 | 2015-05-21 | Global Analytics, Inc. | Credit Risk Decision Management System And Method Using Voice Analytics |
| CN106056444A (en) * | 2016-05-25 | 2016-10-26 | 腾讯科技（深圳）有限公司 | Data processing method and device |
Non-Patent Citations (1)
| Title |
|---|
| 袁晶莹 等: "起底微粒贷", 《二十一世纪商业评论》 * |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109685643A (en) * | 2018-12-13 | 2019-04-26 | 平安科技（深圳）有限公司 | Loan audit risk grade determines method, apparatus, equipment and storage medium |
| CN110348208A (en) * | 2019-06-29 | 2019-10-18 | 上海淇馥信息技术有限公司 | A kind of risk control method based on user behavior and neural network, device and electronic equipment |
| CN110766541A (en) * | 2019-09-25 | 2020-02-07 | 平安科技（深圳）有限公司 | Loan risk assessment method, loan risk assessment device, loan risk assessment equipment and computer-readable storage medium |
| WO2021068635A1 (en) * | 2019-10-11 | 2021-04-15 | 支付宝(杭州)信息技术有限公司 | Information processing method and apparatus, and electronic device |
| CN111681118A (en) * | 2020-05-29 | 2020-09-18 | 泰康保险集团股份有限公司 | Data processing method and device |
| CN111681118B (en) * | 2020-05-29 | 2024-03-15 | 泰康保险集团股份有限公司 | Data processing method and device |
| CN112037039A (en) * | 2020-09-02 | 2020-12-04 | 中国银行股份有限公司 | Loan assessment method and device |
| CN112634025A (en) * | 2020-12-29 | 2021-04-09 | 平安消费金融有限公司 | Wind control rule generation method, device, equipment and computer readable storage medium |
| CN112862472A (en) * | 2021-03-15 | 2021-05-28 | 深圳市心版图科技有限公司 | Loan payment system based on business display industry chain and loan payment judgment method |
| CN113129021A (en) * | 2021-05-18 | 2021-07-16 | 中国银行股份有限公司 | Block chain-based method and device for preventing malicious overdraft of credit card |
| CN113240509A (en) * | 2021-05-18 | 2021-08-10 | 重庆邮电大学 | Loan risk assessment method based on multi-source data federal learning |
| CN113240509B (en) * | 2021-05-18 | 2022-04-22 | 重庆邮电大学 | Loan risk assessment method based on multi-source data federal learning |
| CN117172908A (en) * | 2023-09-05 | 2023-12-05 | 中铁商业保理有限公司 | Business loan stage decision-making auxiliary system based on consumption bill analysis |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108805689A (en) | A kind of loan risk evaluation control method and device | |
| US10496935B2 (en) | System modernization using machine learning | |
| CN102387093B (en) | The method and system that a kind of instant messaging good friend shares with group | |
| Ferreira et al. | A multiple criteria framework to evaluate bank branch potential attractiveness | |
| JP5497263B2 (en) | System and method for improving investment | |
| US20110307304A1 (en) | Crowd-sourced competition platform | |
| CN109472700A (en) | Prediction technique, server and the storage medium of stock price | |
| AU2016361460A1 (en) | Computer-implemented frameworks and methodologies configured to enable delivery of content and/or user interface functionality based on monitoring of activity in a user interface environment and/or control access to services delivered in an online environment responsive to operation of a risk assessment protocol | |
| TWI756688B (en) | Credit-based interactive credit assessment method and device, computing device and computer-readable storage medium therefor | |
| EP3576023A1 (en) | Trade platform with reinforcement learning network and matching engine | |
| Aagaard | A theoretical model of supporting open source front end innovation through idea management | |
| CN110322093A (en) | Information processing method, information display method, device and calculating equipment | |
| CN110070452A (en) | Model training method, calculates equipment and computer readable storage medium at device | |
| Ameen | Arab users’ acceptance and use of mobile phones: a case of young users in Iraq, Jordan and UAE | |
| US20180315341A1 (en) | Virtual stock market environment | |
| CN117522256A (en) | Purchasing screening method, system, storage medium and terminal based on supplier portrait | |
| CN107918584A (en) | Information generating system, device, method and computer-readable recording medium | |
| CN111046156A (en) | Method and device for determining reward data and server | |
| US20190164453A1 (en) | Virtual stock market environment | |
| CN113205391B (en) | Historical order matching degree based order dispatching method, electronic equipment and computer readable medium | |
| CN114140033B (en) | Service personnel allocation method and device, electronic equipment and storage medium | |
| KR20200068064A (en) | Differential fee payment system through professional experts | |
| CN116226116A (en) | Joint learning task viewing method and device, electronic equipment and storage medium | |
| Wang et al. | Optimal dynamic regulation in retail electricity market with consumer feedback and social learning | |
| CN113362185A (en) | Investment portfolio data processing 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 |
Application publication date: 20181113 |
|
| RJ01 | Rejection of invention patent application after publication |