CN107424070A - A kind of loan user credit ranking method and system based on machine learning - Google Patents
A kind of loan user credit ranking method and system based on machine learning Download PDFInfo
- Publication number
- CN107424070A CN107424070A CN201710197889.0A CN201710197889A CN107424070A CN 107424070 A CN107424070 A CN 107424070A CN 201710197889 A CN201710197889 A CN 201710197889A CN 107424070 A CN107424070 A CN 107424070A
- Authority
- CN
- China
- Prior art keywords
- credit
- user
- loan user
- machine learning
- new
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Software Systems (AREA)
- General Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Data Mining & Analysis (AREA)
- Strategic Management (AREA)
- Marketing (AREA)
- Economics (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Technology Law (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Development Economics (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
Abstract
The invention discloses a kind of loan user credit ranking method based on machine learning and system, method to include：The initial data of modeling is obtained, the initial data of modeling includes reference report and overdue trade company's list；Reference report is extracted and index is segmented, obtains predictive variable and its weight；It is modeled according to overdue trade company's list, obtained predictive variable and its weight using the method for machine learning, the forecast model of meet with a response variable and predictive variable；New loan user is predicted according to obtained forecast model, obtains the Default Probability of new loan user；The credit scoring of new loan user is calculated according to the Default Probability of new loan user.Present invention employs the method for machine learning to be modeled, and has adapted to the quick change request of loan user data under the new situation；Be additionally arranged to reference report extracted and index segment the step of, more comprehensively and conveniently.It the composite can be widely applied to computer application field.
Description
Technical field
The present invention relates to computer application field, especially a kind of loan user credit ranking method based on machine learning
And system.
Background technology
Credit rating is also known as " credit rating " or " credit assessment ", is the important content and base for establishing social credit system
Plinth.According to common definition, credit rating be credit rating service organization with third-party objective, just position, according to specification
Evaluation index system, with the appraisal procedure of science, strict appraisal procedure is fulfiled, to enterprise, financial institution, bond issue
The credit record of the market such as person and social organization participation main body, the quality of enterprise, managerial ability, management level, external environment condition, finance
Situation, development prospect etc. fully understanded, surveyed and studied, analyze and research after, it will be met commitment in following a period of time
The overall merit that ability, the various risks being likely to occur are done, and represent that it is good and bad with certain symbol and be published in social public affairs
A kind of many economic activities.Credit rating is evaluated by repaying risk to loan application obligatio personalis, in order to bank etc.
Financial institution carries out examination ＆ approval credit to loan application people.
Traditional credit rating method is mostly based on expert's rule or scorecard model, i.e., is formulated previously according to expertise
A set of code of points, further according to the real data of user, apply mechanically this set rule and carry out credit scoring.However, this credit rating
Mode is that had the scoring that experience carries out based on history, and its scoring has certain hysteresis quality, it is impossible to reacts under the new situation new
User situation, and the formulation and modification of its code of points are required for the cycle one for being expounded through peer review by strict, formulating and changing
As it is long, data change speed is slow.In addition, traditional credit rating method typically reads reference report by artificial mode
Accuse, or the docking entered using reference interface on line interface, reference report file can not be directly parsed, can not be by reference report
Index in announcement is finely divided, not comprehensive enough and conveniently.
The content of the invention
In order to solve the above technical problems, it is an object of the invention to：It is fast, comprehensive and conveniently to provide a kind of data change speed
, the loan user credit ranking method based on machine learning.
Another object of the present invention is to：It is fast, comprehensive and convenient to provide a kind of data change speed, based on machine learning
Loan user credit rating system.
The technical solution used in the present invention is：
A kind of loan user credit ranking method based on machine learning, comprises the following steps：
The initial data of modeling is obtained, the initial data of the modeling includes reference report and overdue trade company's list；
Reference report is extracted and index is segmented, obtains credit line, recent behavior, credit duration, account quantity
With the predictive variable and its weight of refund history this five dimensions；
It is modeled, is obtained using the method for machine learning according to overdue trade company's list, obtained predictive variable and its weight
To response variable and the forecast model of predictive variable, wherein, response variable is the whether overdue variable of reflection trade company；
New loan user is predicted according to obtained forecast model, obtains the Default Probability of new loan user；
The credit scoring of new loan user is calculated according to the Default Probability of new loan user.
Further, the reference report includes credit information, credit card information, quasi- credit card information and Query Information.
Further, the credit line, recent behavior, credit duration, account quantity and refund history this five dimensions
Predictive variable totally 143, the title and weight of this 143 predictive variables are as shown in table 1 below：
Table 1
Further, the method for the machine learning is that gradient lifts traditional decision-tree.
Further, the forecast model that the basis obtains is predicted to new loan user, obtains disobeying for new loan user
About probability the step for, it includes：
7 predictive variables, which are filtered out, from 143 predictive variables of new loan user reference report reports pass as reference
Key index, the reference reporting critical index are respectively that the average accrediting amount of credit card, the last loan refunded is used
Note time of the card away from the present, nearest 24 months inquiry times, the last credit card are away from the now time, earliest credit card away from present
Time, nearest 3 months inquiry times and nearest 6 months inquiry times；
The Default Probability of new loan user is predicted using obtained forecast model according to 7 predictive variables filtered out.
Further, the basis is newly provided a loan the Default Probability of user the step for calculating the credit scoring of new loan user,
It includes：
The preliminary credit scoring of new loan user is calculated according to the Default Probability of new loan user, wherein, newly provide a loan user
The preliminary credit scoring=100* Default Probability of user (1- newly provide a loan)；
Judge whether new loan user only has a reference report, if so, the then preliminary credit scoring directly to calculate
As final credit scoring, conversely, then take in the preliminary credit scoring that the report of all references calculates minimum is allocated as most
Whole credit scoring.
Another technical scheme for being taken of the present invention is：
A kind of loan user credit rating system based on machine learning, including：
Data acquisition module, for obtaining the initial data of modeling, the initial data of the modeling include reference report and
Overdue trade company's list；
Extraction with index subdivision module, for reference report extracted and index segment, obtain credit line, in the recent period
The predictive variable and its weight of this five dimensions of behavior, credit duration, account quantity and refund history；
Modeling module, for using the side of machine learning according to overdue trade company's list, obtained predictive variable and its weight
Method is modeled, the forecast model of meet with a response variable and predictive variable, wherein, whether response variable is overdue for reflection trade company
Variable；
Prediction module, for being predicted according to obtained forecast model to new loan user, obtain new loan user's
Default Probability；
Credit scoring module, for calculating the credit scoring of new loan user according to the Default Probability of new loan user.
Further, the credit line, recent behavior, credit duration, account quantity and refund history this five dimensions
Predictive variable totally 143, the title and weight of this 143 predictive variables are as shown in table 1 below：
Table 1
Further, the prediction module includes：
Screening unit, make for filtering out 7 predictive variables from 143 predictive variables of new loan user reference report
For reference reporting critical index, the reference reporting critical index is respectively that the average accrediting amount of credit card, recently is used
The credit card once refunded away from the present time, nearest 24 months inquiry times, the last credit card away from the now time, earliest
Credit card is away from the now time, nearest 3 months inquiry times and nearest 6 months inquiry times；
Default Probability predicting unit, for being predicted according to 7 predictive variables filtered out using obtained forecast model
The Default Probability of new loan user.
Further, the credit scoring module includes：
Preliminary credit scoring computing unit, for calculating the preliminary of new loan user according to the Default Probability of new loan user
Credit scoring, wherein, preliminary credit scoring=100* of the new user that the provides a loan Default Probability of user (1- newly provide a loan)；
Final credit scoring unit, for judging newly provide a loan, whether only a reference of user is reported, if so, then directly with
The preliminary credit scoring calculated is as final credit scoring, conversely, then taking the preliminary credit that all reference reports calculate
Minimum in scoring is allocated as final credit scoring.
The beneficial effects of the method for the present invention is：Include obtain modeling initial data, to reference report carry out extraction and
Index is segmented, and is modeled according to overdue trade company's list, obtained predictive variable and its weight using the method for machine learning, root
New loan user is predicted according to obtained forecast model and new loan is calculated according to the Default Probability of new loan user and is used
The step of credit scoring at family, the method for employing machine learning are modeled so that the training of forecast model and prediction mould
The scoring output of type rapidly iteration can update, and adapt to the quick change request of loan user data under the new situation；It is additionally arranged
To reference report extracted and index segment the step of, can directly parse reference report and by subdivision obtain credit line,
The predictive variable and its weight of this five dimensions of recent behavior, credit duration, account quantity and refund history, it is more comprehensively and square
Just.
The beneficial effect of system of the present invention is：Including data acquisition module, extraction and index subdivision module, modeling mould
Block, prediction module and credit scoring module, the method for employing machine learning are modeled so that the training of forecast model and
The scoring output of forecast model rapidly iteration can update, and adapt to the quick change request of loan user data under the new situation；
It is additionally arranged and segments module with index for the extraction extracted to reference report and index is segmented, can directly parses reference report
And the prediction change of this five dimensions of credit line, recent behavior, credit duration, account quantity and refund history is obtained by subdivision
Amount and its weight, more comprehensively and conveniently.
Brief description of the drawings
Fig. 1 is a kind of step flow chart of the loan user credit ranking method based on machine learning of the present invention.
Embodiment
The present invention is further explained and illustrated with reference to Figure of description and specific embodiment.
New user situation under the new situation can not be reacted for existing credit rating method, formulates and the cycle of modification is general
Long, data change speed is slow, not comprehensive enough and the problem of facilitate, and the present invention proposes a kind of brand-new loan user credit
Ranking method and system, credit scoring is carried out by way of machine autonomous learning, can iteratively faster, adapted to new shape rapidly
The growth requirement of gesture.
As shown in figure 1, the loan user credit ranking method mainly includes：
(1) initial data of modeling is obtained
The initial data that the present invention models includes two parts, and Part I is reference report (including the loan of the People's Bank
Information, credit card information, quasi- credit card information and Query Information), Part II is to converge with melting easy overdue trade company's list, such as table 2 below
It is shown：
Table 2
(2) data parsing and description are carried out by extraction and index subdivision.
The present invention establishes forecast model as response variable Y value using whether trade company is overdue, overdue to be designated as 1, is normally designated as
0.Meanwhile the present invention also extracts credit line, recent behavior, credit duration, account quantity and refund history from reference report
This five dimensions, predictive variable of totally 143 predictive variables as forecast model, is designated as X.
The present invention is as shown in table 1 below by each index name and weight extracted and index is segmented to obtain：
Table 1
The operational loan refunded in table 1 or other loans can have more, such as 7.
(3) forecast model is built
The present invention is modeled, the variable Y that meets with a response becomes with prediction according to the X and Y of (two) using the method for machine learning
Measure X forecast model.
Established in the forecast model of reality with training process, the gradient in machine learning can be used to lift decision tree
(gradient boosting decision tree) method, directly invoke python and increase income in the scikit-learn of storehouse
Sklearn.ensemble.GradientBoostingClassifier algorithms, become to build response variable Y and 143 predictions
X forecast model is measured, then calls python orders joblib.dump the forecast model of structure is saved as local file,
The credit scoring of newly-gained loan user is calculated for second load.The specific algorithm flow of gradient lifting decision tree can be continued to use existing
Some gradients lift decision Tree algorithms flow, and such as existing gradient based on residual error lifting lifts decision Tree algorithms flow.
(4) the newly Default Probability prediction of loan user
When newly arrive a loan customer when, the present invention can according to derived from step (3) training result, use step
Suddenly (three) forecast model predicts the Default Probability of new loan customer.
In order to further reduce operand and lifting predetermined speed, 7 can be screened from 143 predictive variables of table 1 in advance
Variable is surveyed as predictive variable, this 7 reference reporting critical indexs filtered out are as shown in table 3 below：
Table 3
| Predictive variable Chinese name | Predictive variable importance |
| The average accrediting amount of credit card is used | 0.1066 |
| Time of the credit card that the last time refunds away from the present | 0.0876 |
| Nearest 24 months inquiry times | 0.0666 |
| The last credit card is away from the now time | 0.0574 |
| Earliest credit card is away from the now time | 0.0528 |
| Nearest 3 months inquiry times | 0.0366 |
| Nearest 6 months inquiry times | 0.0332 |
(5) credit scoring calculates
The credit scoring of the present invention, for full marks, after the Default Probability of newly-gained loan user is predicted, can be passed through with 100 points
Conversion formula is converted into corresponding credit scoring, and specific conversion formula is：(1- is newly borrowed credit scoring=100* of new loan user
The Default Probability of money user).
When a loan application people has two parts or more than two parts of reference report, the present invention takes all references to believe in reporting
Minimum with scoring is allocated as final credit scoring.
Reference report is subdivided into 143 specific indexs by the present invention, and traditional expert's rule is when application
Can not possibly accomplish to be subdivided into so more indexs, thus the present invention in the credit scoring of loan customer than traditional expert's rule
It is more careful, more comprehensively.
In addition, credit-graded approach used in the present invention is built upon on the basis of machine learning so that prediction mould
The scoring output conversion of the re -training and forecast model of type rapidly iteration can update；And if using traditional expert
If rule, the foundation and modification of its code of points are required for by by being expounded through peer review repeatedly, iteration cycle length, it is difficult to suitable
Should be provided a loan the quick change request of user data under the new situation.
Above is the preferable implementation to the present invention is illustrated, but the present invention is not limited to the embodiment, ripe
A variety of equivalent variations or replacement can also be made on the premise of without prejudice to spirit of the invention by knowing those skilled in the art, this
Equivalent deformation or replacement are all contained in the application claim limited range a bit.
Claims (10)
- A kind of 1. loan user credit ranking method based on machine learning, it is characterised in that：Comprise the following steps：The initial data of modeling is obtained, the initial data of the modeling includes reference report and overdue trade company's list；Reference report is extracted and index is segmented, credit line, recent behavior, credit duration, account quantity is obtained and goes back The predictive variable and its weight of this five dimensions of money history；It is modeled, is rung using the method for machine learning according to overdue trade company's list, obtained predictive variable and its weight The forecast model of dependent variable and predictive variable, wherein, response variable is the whether overdue variable of reflection trade company；New loan user is predicted according to obtained forecast model, obtains the Default Probability of new loan user；The credit scoring of new loan user is calculated according to the Default Probability of new loan user.
- A kind of 2. loan user credit ranking method based on machine learning according to claim 1, it is characterised in that：Institute Stating reference report includes credit information, credit card information, quasi- credit card information and Query Information.
- A kind of 3. loan user credit ranking method based on machine learning according to claim 1, it is characterised in that：Institute The predictive variable totally 143 of credit line, recent behavior, credit duration, account quantity and refund history this five dimensions is stated, this The title and weight of 143 predictive variables are as shown in table 1 below：Table 1。
- A kind of 4. loan user credit ranking method based on machine learning according to claim 1, it is characterised in that：Institute The method for stating machine learning lifts traditional decision-tree for gradient.
- A kind of 5. loan user credit ranking method based on machine learning according to claim 3, it is characterised in that：Institute State and new loan user is predicted according to obtained forecast model, the step for obtaining the Default Probability of new loan user, its Including：7 predictive variables are filtered out from 143 predictive variables of new loan user reference report as reference reporting critical to refer to Mark, the reference reporting critical index are respectively that the average accrediting amount of credit card, the last credit card refunded is used Time, nearest 24 months inquiry times, the last credit card away from the present away from the now time, earliest credit card away from it is present when Between, nearest 3 months inquiry times and nearest 6 months inquiry times；The Default Probability of new loan user is predicted using obtained forecast model according to 7 predictive variables filtered out.
- 6. a kind of loan user credit ranking method based on machine learning according to claim any one of 1-6, it is special Sign is：The basis is newly provided a loan the Default Probability of user the step for calculating the credit scoring of new loan user, and it includes：According to the preliminary credit scoring of the new loan user of Default Probability calculating of new loan user, wherein, the new user's that provides a loan is first Step credit scoring=100* the Default Probability of user (1- newly provide a loan)；Judge whether new loan user only has a reference report, if so, then directly using the preliminary credit scoring that calculates as Final credit scoring, conversely, then take in the preliminary credit scoring that the report of all references calculates minimum is allocated as to be final Credit scoring.
- A kind of 7. loan user credit rating system based on machine learning, it is characterised in that：Including：Data acquisition module, for obtaining the initial data of modeling, the initial data of the modeling includes reference report and overdue Trade company's list；Extraction with index subdivision module, for reference report extracted and index segment, obtain credit line, recent row For the predictive variable and its weight of, credit duration, account quantity and refund history this five dimensions；Modeling module, for being entered according to overdue trade company's list, obtained predictive variable and its weight using the method for machine learning Row modeling, the forecast model of meet with a response variable and predictive variable, wherein, response variable is the whether overdue change of reflection trade company Amount；Prediction module, for being predicted according to obtained forecast model to new loan user, obtain the promise breaking of new loan user Probability；Credit scoring module, for calculating the credit scoring of new loan user according to the Default Probability of new loan user.
- A kind of 8. loan user credit rating system based on machine learning according to claim 7, it is characterised in that：Institute The predictive variable totally 143 of credit line, recent behavior, credit duration, account quantity and refund history this five dimensions is stated, this The title and weight of 143 predictive variables are as shown in table 1 below：Table 1。
- A kind of 9. loan user credit rating system based on machine learning according to claim 8, it is characterised in that：Institute Stating prediction module includes：Screening unit, for filtering out 7 predictive variables as sign from 143 predictive variables of new loan user reference report Believe reporting critical index, the reference reporting critical index is respectively that the average accrediting amount of credit card, the last time is used Time of the credit card of refund away from the present, nearest 24 months inquiry times, the last credit card are away from the now time, earliest credit Card is away from the now time, nearest 3 months inquiry times and nearest 6 months inquiry times；Default Probability predicting unit, for predicting new loan using obtained forecast model according to 7 predictive variables filtered out The Default Probability of money user.
- 10. a kind of loan user credit rating system based on machine learning according to claim 7,8 or 9, its feature It is：The credit scoring module includes：Preliminary credit scoring computing unit, for calculating the preliminary credit of new loan user according to the Default Probability of new loan user Scoring, wherein, preliminary credit scoring=100* of the new user that the provides a loan Default Probability of user (1- newly provide a loan)；Final credit scoring unit, for judging whether only a reference report of user of newly providing a loan, if so, then directly with calculating The preliminary credit scoring gone out is as final credit scoring, conversely, then taking the preliminary credit scoring that all reference reports calculate In minimum be allocated as final credit scoring.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710197889.0A CN107424070A (en) | 2017-03-29 | 2017-03-29 | A kind of loan user credit ranking method and system based on machine learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710197889.0A CN107424070A (en) | 2017-03-29 | 2017-03-29 | A kind of loan user credit ranking method and system based on machine learning |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107424070A true CN107424070A (en) | 2017-12-01 |
Family
ID=60423317
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710197889.0A Pending CN107424070A (en) | 2017-03-29 | 2017-03-29 | A kind of loan user credit ranking method and system based on machine learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107424070A (en) |
Cited By (32)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107945013A (en) * | 2017-12-22 | 2018-04-20 | 江苏满运软件科技有限公司 | A kind of system and control method of the control of truck man credit risk |
| CN108062709A (en) * | 2017-12-12 | 2018-05-22 | 北京奇虎科技有限公司 | Personal behavior model training method and device based on semi-supervised learning |
| CN108389120A (en) * | 2018-02-06 | 2018-08-10 | 广东弘贝融科信息科技有限公司 | Method, system and device for automatically rating internet credit assets |
| CN108615188A (en) * | 2018-03-22 | 2018-10-02 | 深圳光华普惠科技有限公司 | A kind of system and method that net borrows risk management |
| CN108629379A (en) * | 2018-05-10 | 2018-10-09 | 北京天元创新科技有限公司 | A kind of individual's reference appraisal procedure and system |
| CN109102396A (en) * | 2018-08-17 | 2018-12-28 | 北京玖富普惠信息技术有限公司 | A kind of user credit ranking method, computer equipment and readable medium |
| CN109117976A (en) * | 2018-06-22 | 2019-01-01 | 重庆小雨点小额贷款有限公司 | A kind of loan loss prediction technique, device, server and storage medium |
| CN109255506A (en) * | 2018-11-22 | 2019-01-22 | 重庆邮电大学 | A kind of internet finance user's overdue loan prediction technique based on big data |
| CN109272218A (en) * | 2018-09-04 | 2019-01-25 | 中国平安财产保险股份有限公司 | Bond batch ranking method, device, computer equipment and storage medium |
| CN109345371A (en) * | 2018-08-30 | 2019-02-15 | 成都数联铭品科技有限公司 | Personal reference report backtracking method and system |
| CN109447399A (en) * | 2018-09-17 | 2019-03-08 | 平安科技（深圳）有限公司 | Applicant's rating calculation method, apparatus, computer equipment and storage medium |
| CN109472586A (en) * | 2018-10-29 | 2019-03-15 | 北京网众共创科技有限公司 | Strategy determines method and device, storage medium, electronic device |
| CN109711981A (en) * | 2018-12-28 | 2019-05-03 | 上海点融信息科技有限责任公司 | The method, apparatus and storage medium of the accrediting amount are determined based on artificial intelligence |
| CN109727116A (en) * | 2018-08-17 | 2019-05-07 | 平安普惠企业管理有限公司 | Credit analysis method, device, equipment and computer readable storage medium |
| CN109934688A (en) * | 2019-03-22 | 2019-06-25 | 中国农业银行股份有限公司 | A kind of reference result determines method and device |
| WO2019141125A1 (en) * | 2018-01-18 | 2019-07-25 | 阿里巴巴集团控股有限公司 | Method and device for assessing financial default risk |
| CN110135626A (en) * | 2019-04-17 | 2019-08-16 | 平安科技（深圳）有限公司 | Credit management method and device, electronic equipment, storage medium |
| CN110288459A (en) * | 2019-04-24 | 2019-09-27 | 武汉众邦银行股份有限公司 | Loan prediction technique, device, equipment and storage medium |
| CN110322334A (en) * | 2018-03-29 | 2019-10-11 | 上海麦子资产管理集团有限公司 | Credit rating method and device, computer readable storage medium, terminal |
| CN110659979A (en) * | 2019-09-03 | 2020-01-07 | 深圳中兴飞贷金融科技有限公司 | Method and apparatus for predicting loss rate of default, storage medium, and electronic device |
| CN110689427A (en) * | 2019-10-12 | 2020-01-14 | 杭州绿度信息技术有限公司 | Consumption stage default probability model based on survival analysis |
| CN110910002A (en) * | 2019-11-15 | 2020-03-24 | 安徽海汇金融投资集团有限公司 | Account receivable default risk identification method and system |
| CN111079941A (en) * | 2019-12-03 | 2020-04-28 | 武汉纺织大学 | Credit information system combining expert experience model and supervised machine learning algorithm |
| CN111353882A (en) * | 2020-04-17 | 2020-06-30 | 新分享科技服务(深圳)有限公司 | Privatized deployment retail asset wind control method and device and electronic equipment |
| CN111695989A (en) * | 2020-06-18 | 2020-09-22 | 新分享科技服务(深圳)有限公司 | Modeling method and platform of wind-control credit model |
| CN111833179A (en) * | 2020-07-17 | 2020-10-27 | 浙江网商银行股份有限公司 | Resource allocation platform, resource allocation method and device |
| CN112017060A (en) * | 2020-07-15 | 2020-12-01 | 北京淇瑀信息科技有限公司 | Method and device for resource allocation for target user and electronic equipment |
| CN112037013A (en) * | 2020-08-25 | 2020-12-04 | 成都榕慧科技有限公司 | Pedestrian credit variable derivation method and device |
| CN113298510A (en) * | 2018-07-10 | 2021-08-24 | 马上消费金融股份有限公司 | Deduction instruction initiating method and device |
| CN113344438A (en) * | 2021-06-29 | 2021-09-03 | 百维金科(上海)信息科技有限公司 | Loan system, loan monitoring method, loan monitoring apparatus, and loan medium for monitoring loan behavior |
| CN116433363A (en) * | 2023-04-18 | 2023-07-14 | 上海德易车信息科技有限公司 | Automobile finance intelligent interest settlement system based on user demand analysis |
| US20230334506A1 (en) * | 2018-09-25 | 2023-10-19 | Capital One Services, Llc | Machine learning-driven servicing interface |
-
2017
- 2017-03-29 CN CN201710197889.0A patent/CN107424070A/en active Pending
Cited By (37)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108062709A (en) * | 2017-12-12 | 2018-05-22 | 北京奇虎科技有限公司 | Personal behavior model training method and device based on semi-supervised learning |
| CN107945013A (en) * | 2017-12-22 | 2018-04-20 | 江苏满运软件科技有限公司 | A kind of system and control method of the control of truck man credit risk |
| WO2019141125A1 (en) * | 2018-01-18 | 2019-07-25 | 阿里巴巴集团控股有限公司 | Method and device for assessing financial default risk |
| CN108389120B (en) * | 2018-02-06 | 2022-03-15 | 广东弘贝融科信息科技有限公司 | Method, system and device for automatically rating internet credit assets |
| CN108389120A (en) * | 2018-02-06 | 2018-08-10 | 广东弘贝融科信息科技有限公司 | Method, system and device for automatically rating internet credit assets |
| CN108615188A (en) * | 2018-03-22 | 2018-10-02 | 深圳光华普惠科技有限公司 | A kind of system and method that net borrows risk management |
| CN110322334A (en) * | 2018-03-29 | 2019-10-11 | 上海麦子资产管理集团有限公司 | Credit rating method and device, computer readable storage medium, terminal |
| CN108629379A (en) * | 2018-05-10 | 2018-10-09 | 北京天元创新科技有限公司 | A kind of individual's reference appraisal procedure and system |
| CN109117976A (en) * | 2018-06-22 | 2019-01-01 | 重庆小雨点小额贷款有限公司 | A kind of loan loss prediction technique, device, server and storage medium |
| CN109117976B (en) * | 2018-06-22 | 2021-05-11 | 重庆小雨点小额贷款有限公司 | Loan loss prediction method, loan loss prediction device, loan loss server and storage medium |
| CN113298510A (en) * | 2018-07-10 | 2021-08-24 | 马上消费金融股份有限公司 | Deduction instruction initiating method and device |
| CN109102396A (en) * | 2018-08-17 | 2018-12-28 | 北京玖富普惠信息技术有限公司 | A kind of user credit ranking method, computer equipment and readable medium |
| CN109727116A (en) * | 2018-08-17 | 2019-05-07 | 平安普惠企业管理有限公司 | Credit analysis method, device, equipment and computer readable storage medium |
| CN109345371A (en) * | 2018-08-30 | 2019-02-15 | 成都数联铭品科技有限公司 | Personal reference report backtracking method and system |
| CN109272218B (en) * | 2018-09-04 | 2023-12-01 | 中国平安财产保险股份有限公司 | Method, device, computer equipment and storage medium for batch rating bonds |
| CN109272218A (en) * | 2018-09-04 | 2019-01-25 | 中国平安财产保险股份有限公司 | Bond batch ranking method, device, computer equipment and storage medium |
| CN109447399A (en) * | 2018-09-17 | 2019-03-08 | 平安科技（深圳）有限公司 | Applicant's rating calculation method, apparatus, computer equipment and storage medium |
| US20230334506A1 (en) * | 2018-09-25 | 2023-10-19 | Capital One Services, Llc | Machine learning-driven servicing interface |
| CN109472586A (en) * | 2018-10-29 | 2019-03-15 | 北京网众共创科技有限公司 | Strategy determines method and device, storage medium, electronic device |
| CN109255506A (en) * | 2018-11-22 | 2019-01-22 | 重庆邮电大学 | A kind of internet finance user's overdue loan prediction technique based on big data |
| CN109711981A (en) * | 2018-12-28 | 2019-05-03 | 上海点融信息科技有限责任公司 | The method, apparatus and storage medium of the accrediting amount are determined based on artificial intelligence |
| CN109934688A (en) * | 2019-03-22 | 2019-06-25 | 中国农业银行股份有限公司 | A kind of reference result determines method and device |
| CN110135626A (en) * | 2019-04-17 | 2019-08-16 | 平安科技（深圳）有限公司 | Credit management method and device, electronic equipment, storage medium |
| CN110288459A (en) * | 2019-04-24 | 2019-09-27 | 武汉众邦银行股份有限公司 | Loan prediction technique, device, equipment and storage medium |
| CN110659979A (en) * | 2019-09-03 | 2020-01-07 | 深圳中兴飞贷金融科技有限公司 | Method and apparatus for predicting loss rate of default, storage medium, and electronic device |
| CN110689427A (en) * | 2019-10-12 | 2020-01-14 | 杭州绿度信息技术有限公司 | Consumption stage default probability model based on survival analysis |
| CN110910002A (en) * | 2019-11-15 | 2020-03-24 | 安徽海汇金融投资集团有限公司 | Account receivable default risk identification method and system |
| CN111079941A (en) * | 2019-12-03 | 2020-04-28 | 武汉纺织大学 | Credit information system combining expert experience model and supervised machine learning algorithm |
| CN111079941B (en) * | 2019-12-03 | 2024-02-20 | 武汉纺织大学 | Credit information processing method, credit information processing system, terminal and storage medium |
| CN111353882A (en) * | 2020-04-17 | 2020-06-30 | 新分享科技服务(深圳)有限公司 | Privatized deployment retail asset wind control method and device and electronic equipment |
| CN111695989A (en) * | 2020-06-18 | 2020-09-22 | 新分享科技服务(深圳)有限公司 | Modeling method and platform of wind-control credit model |
| CN112017060A (en) * | 2020-07-15 | 2020-12-01 | 北京淇瑀信息科技有限公司 | Method and device for resource allocation for target user and electronic equipment |
| CN112017060B (en) * | 2020-07-15 | 2024-04-26 | 北京淇瑀信息科技有限公司 | Method and device for allocating resources for target user and electronic equipment |
| CN111833179A (en) * | 2020-07-17 | 2020-10-27 | 浙江网商银行股份有限公司 | Resource allocation platform, resource allocation method and device |
| CN112037013A (en) * | 2020-08-25 | 2020-12-04 | 成都榕慧科技有限公司 | Pedestrian credit variable derivation method and device |
| CN113344438A (en) * | 2021-06-29 | 2021-09-03 | 百维金科(上海)信息科技有限公司 | Loan system, loan monitoring method, loan monitoring apparatus, and loan medium for monitoring loan behavior |
| CN116433363A (en) * | 2023-04-18 | 2023-07-14 | 上海德易车信息科技有限公司 | Automobile finance intelligent interest settlement system based on user demand analysis |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107424070A (en) | A kind of loan user credit ranking method and system based on machine learning | |
| Kara et al. | A data mining-based framework for supply chain risk management | |
| CN104321794B (en) | A kind of system and method that the following commercial viability of an entity is determined using multidimensional grading | |
| CN111652657A (en) | Commodity sales prediction method and device, electronic equipment and readable storage medium | |
| CN107818344A (en) | The method and system that user behavior is classified and predicted | |
| Sohrabinejad et al. | Risk determination, prioritization, and classifying in construction project case study: gharb tehran commercial-administrative complex | |
| WO2022155740A1 (en) | Systems and methods for outlier detection of transactions | |
| CN114048436A (en) | Construction method and construction device for forecasting enterprise financial data model | |
| CN109214915A (en) | Borrow risk methods of marking, system and computer readable storage medium | |
| CN106407305A (en) | Data mining system and method | |
| CN112116103A (en) | Method, device and system for evaluating personal qualification based on federal learning and storage medium | |
| CN115545886A (en) | Overdue risk identification method, overdue risk identification device, overdue risk identification equipment and storage medium | |
| CN107256254A (en) | A kind of Industrial Cycle index acquisition methods, storage device and terminal | |
| US20210117828A1 (en) | Information processing apparatus, information processing method, and program | |
| CN115689713A (en) | Abnormal risk data processing method and device, computer equipment and storage medium | |
| US20220404778A1 (en) | Intellectual quality management method, electronic device and computer readable storage medium | |
| Xia et al. | Analysis and prediction of telecom customer churn based on machine learning | |
| CN114626898A (en) | Sales forecasting method, tool, system, equipment and storage medium | |
| CN113361911A (en) | New media content delivery method and equipment based on asset wind control | |
| CN113256404A (en) | Data processing method and device | |
| CN112184035A (en) | Customer characteristic element statistical system and method | |
| CN112686553A (en) | Enterprise automatic rating management system, method, server and readable storage device | |
| CN113743695A (en) | International engineering project bid quotation risk management method based on big data | |
| Ibrahem et al. | THE IMPACT OF BUSINESS INTELLIGENCE TECHNOLOGIES ON ACTIVATING ACCOUNTING INFORMATION SYSTEMS: A SURVEY STUDY ON A SAMPLE OF ACADEMICIANS IN SOME IRAQI UNIVERSITIES | |
| Yan et al. | Predictive Analysis of Customer Churn in Community-Supported Agriculture Based on RFM Modeling |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| 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: 20171201 |