CN107633265A - For optimizing the data processing method and device of credit evaluation model - Google Patents
For optimizing the data processing method and device of credit evaluation model Download PDFInfo
- Publication number
- CN107633265A CN107633265A CN201710785991.2A CN201710785991A CN107633265A CN 107633265 A CN107633265 A CN 107633265A CN 201710785991 A CN201710785991 A CN 201710785991A CN 107633265 A CN107633265 A CN 107633265A
- Authority
- CN
- China
- Prior art keywords
- data
- variable
- model
- borrower
- feature value
- 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.)
- Granted
Links
Landscapes
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
Abstract
The present invention relates to the data processing method and device for optimizing credit evaluation model, methods described includes：The relevant information of borrower is obtained as sample data；The sample data is divided into training set and test set；Data modeling is carried out using the training set, obtains entry evaluation model；The entry evaluation model is tested using the test set；If test result is unsatisfactory for evaluation criteria, training set and test set are repartitioned, data modeling and test are carried out using training set and the test set training repartitioned；If test result meets evaluation criteria, terminate to train, it is determined that final assessment models.Data processing method and device provided by the present invention for optimizing credit evaluation model, can optimize credit evaluation model, improve Evaluation accuracy.
Description
Technical field
The present invention relates to finance data processing technology field, and in particular to a kind of data for being used to optimize credit evaluation model
Processing method and processing device.
Background technology
At present, on the market personal debt-credit software is more, different software towards target group it is different.In order to reduce wind
Danger is, it is necessary to assess the loan repayment capacity of user, for accurate lock onto target client, it is necessary to which the debt-credit tendency to user is carried out
Assess.
But in actual application, loan platform big data is adapted to the application of Data Analyst.If commented in credit
There occurs some missing or invalid values, the model in sub-model possibly can not successfully detect, and then borrower be produced inclined
Partial estimation.Also, in startup stage, finance company may be unaware that the feature of which type of borrower in credit scoring pattern
In be important.Credit scoring pattern from large-scale finance company may be too advanced, it is impossible to uses.Therefore, for initial stage
Sample is few, if user data information is not complete, shortage of data, can not build suitable assessment models and be assessed.For example, refund
One of variable of the assessment models of ability is wage income, if the wage income of user can not be obtained, can not accurately be commented
Estimate its loan repayment capacity.
After credit evaluation model has been built, how Optimized model, improve Evaluation accuracy, be that those skilled in the art need badly
Solve the problems, such as.
The content of the invention
For in the prior art the defects of, provided by the present invention for optimize credit evaluation model data processing method and
Device, credit evaluation model can be optimized, improve Evaluation accuracy.
In a first aspect, the invention provides a kind of data processing method for being used to optimize credit evaluation model, including：
The relevant information of borrower is obtained as sample data；
The sample data is divided into training set and test set；
Data modeling is carried out using the training set, obtains entry evaluation model；
The entry evaluation model is tested using the test set；
If test result is unsatisfactory for evaluation criteria, training set and test set are repartitioned, utilizes the training repartitioned
Collection and test set training carry out data modeling and test；
If test result meets evaluation criteria, terminate to train, it is determined that final assessment models.
Data processing method provided by the present invention for optimizing credit evaluation model, divides sample data to training set and survey
Examination collection, assessment models are built by training set, the predictive ability of assessment models is tested by test set, not conformed in inspection
During lattice, by repartitioning training set and test set, variable is reclassified, new model feature value is obtained, by upper
Cross validation method is stated, realizes the optimization of assessment models, improves Evaluation accuracy.In addition, cross validation method can effective land productivity
With all information in sample data, depth excavates the feature of borrower, to improve the Evaluation accuracy of model, and solved plan
Conjunction problem.
Preferably, it is described to carry out data modeling using the training set, entry evaluation model is obtained, including：
Segment processing is carried out to the continuous variable in the training set using decision Tree algorithms, by the continuous variable
Be converted to discrete variable；
Classification processing is carried out to the discrete variable in the training set using clustering algorithm；
Variable is merged according to classification results, determines rudimentary model characteristic value；
Logistic regression is carried out to the sample data of the model feature value, establishes entry evaluation model.
Preferably, before logistic regression is carried out, in addition to：
If the model feature value of borrower lacks data, the data of the completion model feature value.
Preferably, if the model feature value of the borrower lacks data, the data of the completion model feature value, bag
Include：
If the model feature value of borrower lacks data, the replacement variable of the model feature value is found；
According to the data for replacing the Supplementing Data of the variable model feature value found.
Preferably, determining the method for the replacement variable includes：
Calculate the Euclidean distance between variable；
Two variables that Euclidean distance is less than threshold value replace variable each other.
Preferably, if the model feature value of the borrower lacks data, the data of the completion model feature value, bag
Include：
If the model feature value of borrower lacks data, calculate all borrower's model feature values average or in
Value；
According to the model feature value for lacking data for the average or intermediate value completion borrower being calculated.
Preferably, in addition to：Obtain outside statistics；
If the model feature value of the borrower lacks data, the data of the completion model feature value, including：
If the model feature value of borrower lacks data, according to lacking for the outside statistics completion borrower
The model feature value of data.
Preferably, before logistic regression is carried out, in addition to：
Calculate the information value of each variable；
Tested according to predetermined value threshold value, whether judgment variable is effective；
Logistic regression is not involved in for invalid variable.
Second aspect, the invention provides a kind of data processing equipment for being used to optimize credit evaluation model, including：
Data acquisition module, for obtaining the relevant information of borrower as sample data；
Sample division module, for the sample data to be divided into training set and test set；
Model training module, for carrying out data modeling using the training set, obtain entry evaluation model；
Model measurement module, for being tested using the test set the entry evaluation model；If test result
Evaluation criteria is unsatisfactory for, then repartitions training set and test set, is trained and carried out using the training set and test set repartitioned
Data modeling and test；If test result meets evaluation criteria, terminate to train, it is determined that final assessment models.
The third aspect, the invention provides a kind of computer-readable recording medium, computer program is stored thereon with, the journey
The either method described in above-mentioned first aspect is realized when sequence is executed by processor.
Brief description of the drawings
The flow chart for being used to optimize the data processing method of credit evaluation model that Fig. 1 is provided by the embodiment of the present invention；
The structural frames for being used to optimize the data processing equipment of credit evaluation model that Fig. 2 is provided by the embodiment of the present invention
Figure；
The structured flowchart for the model training module that Fig. 3 is provided by the embodiment of the present invention.
Embodiment
The embodiment of technical solution of the present invention is described in detail below in conjunction with accompanying drawing.Following examples are only used for
Clearly illustrate technical scheme, therefore be intended only as example, and the protection of the present invention can not be limited with this
Scope.
It should be noted that unless otherwise indicated, technical term or scientific terminology used in this application should be this hair
The ordinary meaning that bright one of ordinary skill in the art are understood.
As shown in figure 1, a kind of data processing method for being used to optimize credit evaluation model is present embodiments provided, including：
Step S1, the relevant information of borrower is obtained as sample data.
Wherein, the sample data includes continuous variable and discrete variable.The relevant information of borrower be it is all can
To disclose the information of the specific behavioural characteristic of borrower, herein below can be included but is not limited to：Age, wage income, wedding
Relation by marriage situation, house-purchase situation, employment status, insurance purchase situation, situation etc. of receiving an education, information above may all influence to borrow money
The ability of the repaying of people, these informational influences are borrowed or lent money to the variable assessed., can be by sample data according to the type of sample data
It is divided into continuous variable and discrete variable, such as：Age, wage income etc. have concrete numerical value and in continuously distributed states
Data are continuous variable, and it is that discrete type becomes that situation of receiving an education etc., which is not concrete numerical value or the data that discretization is distributed are presented,
Amount.
Wherein, the sample data of each borrower also includes the violation of agreement of the borrower, that is, the borrower of promise breaking be present
It is artificial " hospitable family " in the absence of the loaning bill of promise breaking for " bad client ".
Step S2, the sample data is divided into training set and test set.
Preferably, can be by sample data with 7:3 ratio is divided into training set and test set.
Step S3, data modeling is carried out using the training set, obtains entry evaluation model.
Step S4, the entry evaluation model is tested using the test set.
Wherein, entry evaluation model output valve is the credit predicted value of borrower in test set, i.e., according in test set
Sample data inputs entry evaluation model, and it is " hospitable family " or " bad client " to obtain borrower.
Step S5, if test result is unsatisfactory for evaluation criteria, training set and test set are repartitioned, using repartitioning
Training set and test set training carry out data modeling and test.
Step S6, if test result meets evaluation criteria, terminate to train, it is determined that final assessment models.
Wherein, the method for assessing test result is：By the credit predicted value and sample data of the step S4 borrowers exported
In the violation of agreement of the borrower be compared, see predicting whether correct, the accuracy rate of statistical test collection sees whether accuracy rate reaches
To evaluation criteria.
What the present embodiment provided is used to optimize the data processing method of credit evaluation model, by sample data divide training set and
Test set, assessment models are built by training set, the predictive ability of assessment models tested by test set, examined not
When qualified, by repartitioning training set and test set, variable is reclassified, new model feature value is obtained, passes through
Above-mentioned cross validation method, the optimization of assessment models is realized, improve Evaluation accuracy.In addition, cross validation method can be effectively
Using all information in sample data, depth is excavated the feature of borrower, to improve the Evaluation accuracy of model, and solved
Fitting problems.
Wherein, the preferred embodiment of the step S4 includes：
Step S401, segment processing is carried out to the continuous variable in the training set using decision Tree algorithms, by described in
Continuous variable is converted to discrete variable.
Wherein, when the subdivision that borrower breaks a contract between possibility prediction and borrower's feature is widely different, by that will become
Amount is divided into multiple segments, and analytic statistics is carried out respectively to each segment, and the spy for analyzing borrower is more suitable for than single variable
Sign, to optimize the classification of borrower's feature.Segment processing is carried out to continuous variable by decision Tree algorithms, by continuous variable
Discretization, borrower can be divided into different homogeneity subgroups, to improve the performance of logistic regression.Wherein, decision Tree algorithms
Existing decision Tree algorithms can be used to realize, will not be repeated here.The present embodiment is preferably using the interaction inspection automatically of card side
Survey (CHAID), CHAID is a kind of nonparametric decision tree method, and it is efficiently applied to the visitor in various research fields, such as marketing
The family propensity to consume, human behavior and landslide in psychology, can be segmented to continuous variable very well, be borrowed with optimization
The classification of money people's feature, when being applied in logistic regression, it will overcome nonlinear shortcoming.
Step S402, classification processing is carried out to the discrete variable in the training set using clustering algorithm.
Wherein, the discrete variable in step S3 includes discrete variable original in sample data, and passes through step
The discrete variable that S2 is converted to.
Wherein, cluster is by the unsupervised learning grader of the data group synthesis set of clusters with similar characteristics, can be incited somebody to action
Homogeneous feature is associated in sample data, to reduce the mistake classification effect between variable.Cluster in the present embodiment refers to become
Amount cluster (also known as R types cluster), the sample data by each debtor is that variable is classified, and finds out the generation in every class
Table element (i.e. model feature value).By separating isomery borrower, the variable after cluster can improve forecasting efficiency.Therefore, exist
In the present embodiment, variable is subjected to classification merging using clustering technique, the characteristic sub-area of variable can be improved, returned with adaptation logic
Return, to improve credit violation correction performance.Wherein, clustering algorithm can use existing clustering algorithm to realize, no longer superfluous herein
State.In this implementation, clustered using Ward minimum variance layered approach, the phase between small sample variable is found according to minimum variance
Guan Xing, one kind is classified as, solves the problems, such as that variable small sample can hardly participate in statistics calculating in recurrence.It is for example, right
In the classification of some small samples, such as " majoring in " education background, " scholar " is combined as the new category of " this is above section level ".
Step S403, variable is merged according to classification results, determines rudimentary model characteristic value.
Wherein, variable is merged according to classification results to be accomplished by the following way：To the variable in same class,
The correlation between each variable is calculated, finds out a variable maximum with other correlation of variables, it is special as such model
Sign amount, to substitute its dependent variable in same class, simplify the input variable of assessment models.
Wherein, model feature value is the key character that the possibility found out causes the borrower of loan defaults.
Step S404, logistic regression is carried out to the sample data of the model feature value, establishes entry evaluation model.
Wherein, the predictive ability of logistic regression is strong and operability is simple, can more conveniently realize prediction target.Logic is returned
The independent variable returned is model feature value, and the binary dependent variable of logistic regression is the violation of agreement of borrower, i.e., " hospitable family " and
" bad client ".The relation between independent variable and dependent variable is found using logistic regression, you can obtain assessment models, the process is to patrol
The general training process returned is collected to will not be repeated here.
The above method, continuous variable can be segmented very well by decision tree classification, to optimize borrower's feature
Classification, when being applied in logistic regression, it will overcome nonlinear shortcoming；Solves the sample in logistic regression by cluster
Notebook data can hardly participate in the problem of statistics calculates, and take full advantage of Small Sample Database, improve the estimated accuracy of model；Knot
Close and state various algorithms, suitable model feature value can be excavated, improve the Evaluation accuracy of credit evaluation model.
Because the source in sample data is complex, it is difficult to ensure the complete of sample data, in order to be deposited in sample data
Remain to effectively be analyzed using the sample data in missing, the method for the present embodiment, before logistic regression is carried out, also wrap
Step S405 is included, if the model feature value of borrower lacks data, the data of the completion model feature value.
Wherein, the preferred embodiment of the step S405 specifically includes：
Step S511, if the model feature value of borrower lacks data, find the replacement variable of the model feature value.
Wherein, there is certain correlation between replacement variable, situation about be able to can not be used in the data of a variable
The lower data with replacement variable are substituted, completion sample data, improve the utilization rate of sample data.
Step S512, according to the data for replacing the Supplementing Data of the variable model feature value found.
Wherein it is determined that the method for replacing variable comprises the following steps：
Calculate the Euclidean distance between variable；
Two variables that Euclidean distance is less than threshold value replace variable each other.
Wherein, threshold value can determine according to actual conditions, be not easy it is excessive or too small, it is too small to can not find substitute variable, mistake
Cause substitute variable improper greatly.Alternatively, it is also possible to the substitute variable using two minimum variables of Euclidean distance as other side.
During the shortage of data of one variable, it can be used to replace the data pair of variable
Wherein, step S405 another preferred embodiment specifically includes：
Step S521, if the model feature value of borrower lacks data, calculate all borrower's model feature values
Average or intermediate value.
Step S522, according to the model feature value for lacking data for the average or intermediate value completion borrower being calculated.
Step S405 another preferred embodiment specifically includes：If the model feature value of borrower lacks data, root
According to the model feature value for lacking data of the outside statistics completion borrower.
Wherein, the sample data stage is being obtained also including obtaining outside statistics.Outside statistics refers to count class
Data, such as Shenzhen's employment rate, Shenzhen's average salary.
Not all variable can all have an impact to final assessment result, in order to lower data processing amount, it is necessary to enter
The variable invalid to assessment result is filtered out before row logistic regression, is specifically included：
Calculate the information value of each variable；
Tested according to predetermined value threshold value, whether judgment variable is effective；
Logistic regression is not involved in for invalid variable.
Whether above-mentioned judgment variable is effectively step, can be assessed before variable classification to reduce the change for participating in clustering
Amount；Or Effective judgement only can carried out to the variable for being defined as model feature value, further reduce and participate in model foundation
Irrelevant variable.
In actual application, evidence weight is that the ratio of " good " borrower's feature corresponds to " bad " to borrower
The Logarithmic calculation of the ratio of feature, for assessment and the relative risk of more different classes of variable.The specific calculating of evidence weight
Formula is as follows：
Wherein, WOE represents the evidence weight of a certain characteristic variable, and DistrGoods represents " good " in sample data and borrowed money
The distribution proportion in this feature variable of people, DistrBads represent sample data in " bad " borrower in this feature variable
Distribution proportion.WOE on the occasion of higher, the credit default risk of customer action is lower, and WOE negative value is bigger, customer action
Credit default risk it is higher.Variable can be converted into the form of rule and information by WOE, and this causes different types of variable
Can be in identical method.Variable can be transferred in WOE, can more effectively protect the free degree of small sample problem.Therefore,
The different variables for using WOE to be concentrated with smaller sample data.Information value can assess the predictive ability of characteristic variable, specifically
Calculation formula is as follows：
IV=(DistrGoods-DistrBads) * WOE,
Wherein, IV represents the information value of a certain characteristic variable, and DistrGoods represents " good " in sample data and borrowed money
The distribution proportion in this feature variable of people, DistrBads represent sample data in " bad " borrower in this feature variable
Distribution proportion, WOE represents the evidence weight of this feature variable.
As shown in Fig. 2 based on the above-mentioned data digging method identical inventive concept for credit evaluation, the present embodiment
A kind of data processing equipment for being used to optimize credit evaluation model is provided, including：
Data acquisition module, for obtaining the relevant information of borrower as sample data；
Sample division module, for the sample data to be divided into training set and test set；
Model training module, for carrying out data modeling using the training set, obtain entry evaluation model；
Model measurement module, for being tested using the test set the entry evaluation model；If test result
Evaluation criteria is unsatisfactory for, then repartitions training set and test set, is trained and carried out using the training set and test set repartitioned
Data modeling and test；If test result meets evaluation criteria, terminate to train, it is determined that final assessment models.
Preferably, as shown in figure 3, the model training module specifically includes：
First sort module, for being carried out using decision Tree algorithms to the continuous variable in the training set at segmentation
Reason, discrete variable is converted to by the continuous variable；
Second sort module, for carrying out classification processing to the discrete variable in the training set using clustering algorithm；
Variable merging module, for being merged according to classification results to variable, determine rudimentary model characteristic value；
Logistic Regression module, for carrying out logistic regression to the sample data of the model feature value, establish entry evaluation
Model.
Preferably, in addition to variable module is replaced, be used for：
Calculate the Euclidean distance between variable；
Two variables that Euclidean distance is less than threshold value replace variable each other.
Preferably, in addition to Supplementing Data module is used for：Before logistic regression is carried out, if the model feature value of borrower
Lack data, then the data of the completion model feature value.
Preferably, the Supplementing Data module is specifically used for：
If the model feature value of borrower lacks data, the replacement variable of the model feature value is found；
According to the data for replacing the Supplementing Data of the variable model feature value found.
Preferably, the Supplementing Data module is used for：
If the model feature value of borrower lacks data, calculate all borrower's model feature values average or in
Value；
According to the model feature value for lacking data for the average or intermediate value completion borrower being calculated.
Preferably, the data acquisition module can be also used for obtaining outside statistics；Correspondingly, the Supplementing Data
Module is specifically used for：If the model feature value of borrower lacks data, according to described outside statistics completion borrower
The model feature value for lacking data.
Preferably, in addition to variable cleaning module, it is used for：Before logistic regression is carried out, the information of each variable is calculated
Value；Tested according to predetermined value threshold value, whether judgment variable is effective；Logic is not involved in for invalid characteristic variable
Return.
A kind of data mining device for credit evaluation that the present embodiment provides and the above-mentioned data for credit evaluation
Method for digging has identical beneficial effect, here is omitted for identical inventive concept.
Based on providing a kind of meter with the above-mentioned data digging method identical inventive concept for credit evaluation, this implementation
Calculation machine readable storage medium storing program for executing, is stored thereon with computer program, it is characterised in that the side of stating is realized when the program is executed by processor
Any described method in method embodiment.
Finally it should be noted that：Various embodiments above is merely illustrative of the technical solution of the present invention, rather than its limitations；To the greatest extent
The present invention is described in detail with reference to foregoing embodiments for pipe, it will be understood by those within the art that：Its according to
The technical scheme described in foregoing embodiments can so be modified, either which part or all technical characteristic are entered
Row equivalent substitution；And these modifications or replacement, the essence of appropriate technical solution is departed from various embodiments of the present invention technology
The scope of scheme, it all should cover among the claim of the present invention and the scope of specification.
Claims (10)
- A kind of 1. data processing method for being used to optimize credit evaluation model, it is characterised in that including：The relevant information of borrower is obtained as sample data；The sample data is divided into training set and test set；Data modeling is carried out using the training set, obtains entry evaluation model；The entry evaluation model is tested using the test set；If test result is unsatisfactory for evaluation criteria, repartition training set and test set, using the training set repartitioned and Test set training carries out data modeling and test；If test result meets evaluation criteria, terminate to train, it is determined that final assessment models.
- 2. according to the method for claim 1, it is characterised in that it is described to carry out data modeling using the training set, obtain Entry evaluation model, including：Segment processing is carried out to the continuous variable in the training set using decision Tree algorithms, the continuous variable is changed For discrete variable；Classification processing is carried out to the discrete variable in the training set using clustering algorithm；Variable is merged according to classification results, determines rudimentary model characteristic value；Logistic regression is carried out to the sample data of the model feature value, establishes entry evaluation model.
- 3. according to the method for claim 2, it is characterised in that before logistic regression is carried out, in addition to：If the model feature value of borrower lacks data, the data of the completion model feature value.
- 4. according to the method for claim 3, it is characterised in that if the model feature value of the borrower lacks data, The data of the completion model feature value, including：If the model feature value of borrower lacks data, the replacement variable of the model feature value is found；According to the data for replacing the Supplementing Data of the variable model feature value found.
- 5. according to the method for claim 4, it is characterised in that determining the method for the replacement variable includes：Calculate the Euclidean distance between variable；Two variables that Euclidean distance is less than threshold value replace variable each other.
- 6. according to the method for claim 3, it is characterised in that if the model feature value of the borrower lacks data, The data of the completion model feature value, including：If the model feature value of borrower lacks data, the average or intermediate value of all borrower's model feature values are calculated；According to the model feature value for lacking data for the average or intermediate value completion borrower being calculated.
- 7. according to the method for claim 3, it is characterised in that also include：Obtain outside statistics；If the model feature value of the borrower lacks data, the data of the completion model feature value, including：If the model feature value of borrower lacks data, data are lacked according to the outside statistics completion borrower Model feature value.
- 8. according to the method for claim 2, it is characterised in that before logistic regression is carried out, in addition to：Calculate the information value of each variable；Tested according to predetermined value threshold value, whether judgment variable is effective；Logistic regression is not involved in for invalid variable.
- A kind of 9. data processing equipment for being used to optimize credit evaluation model, it is characterised in that including：Data acquisition module, for obtaining the relevant information of borrower as sample data；Sample division module, for the sample data to be divided into training set and test set；Model training module, for carrying out data modeling using the training set, obtain entry evaluation model；Model measurement module, for being tested using the test set the entry evaluation model；If test result is discontented with Sufficient evaluation criteria, then repartition training set and test set, and data are carried out using training set and the test set training repartitioned Modeling and test；If test result meets evaluation criteria, terminate to train, it is determined that final assessment models.
- 10. a kind of computer-readable recording medium, is stored thereon with computer program, it is characterised in that the program is by processor The method described in one of claim 1-8 is realized during execution.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710785991.2A CN107633265B (en) | 2017-09-04 | 2017-09-04 | Data processing method and device for optimizing credit evaluation model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710785991.2A CN107633265B (en) | 2017-09-04 | 2017-09-04 | Data processing method and device for optimizing credit evaluation model |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107633265A true CN107633265A (en) | 2018-01-26 |
| CN107633265B CN107633265B (en) | 2021-03-30 |
Family
ID=61101009
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710785991.2A Active CN107633265B (en) | 2017-09-04 | 2017-09-04 | Data processing method and device for optimizing credit evaluation model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107633265B (en) |
Cited By (30)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108711103A (en) * | 2018-06-04 | 2018-10-26 | 中国平安人寿保险股份有限公司 | Personal loan repays Risk Forecast Method, device, computer equipment and medium |
| CN108898479A (en) * | 2018-06-28 | 2018-11-27 | 中国农业银行股份有限公司 | The construction method and device of Credit Evaluation Model |
| CN108960505A (en) * | 2018-05-31 | 2018-12-07 | 试金石信用服务有限公司 | Quantitative estimation method, device, system and the storage medium of personal finance credit |
| CN108985341A (en) * | 2018-06-26 | 2018-12-11 | 四川斐讯信息技术有限公司 | A kind of the training set appraisal procedure and system of neural network model |
| CN109325020A (en) * | 2018-08-20 | 2019-02-12 | 中国平安人寿保险股份有限公司 | Small sample application method, device, computer equipment and storage medium |
| CN109389030A (en) * | 2018-08-23 | 2019-02-26 | 平安科技（深圳）有限公司 | Facial feature points detection method, apparatus, computer equipment and storage medium |
| CN109408583A (en) * | 2018-09-25 | 2019-03-01 | 平安科技（深圳）有限公司 | Data processing method and device, computer readable storage medium, electronic equipment |
| CN109409672A (en) * | 2018-09-25 | 2019-03-01 | 深圳市元征科技股份有限公司 | A kind of auto repair technician classifies grading modeling method and device |
| CN109480864A (en) * | 2018-10-26 | 2019-03-19 | 首都医科大学附属北京安定医院 | A kind of schizophrenia automatic evaluation system based on nervous functional defects and machine learning |
| CN109583590A (en) * | 2018-11-29 | 2019-04-05 | 深圳和而泰数据资源与云技术有限公司 | Data processing method and data processing equipment |
| CN110162995A (en) * | 2019-04-22 | 2019-08-23 | 阿里巴巴集团控股有限公司 | Assess the method and device thereof of contribution data degree |
| 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 |
| CN110348722A (en) * | 2019-07-01 | 2019-10-18 | 百维金科(上海)信息科技有限公司 | A kind of internet finance air control model based on XGBoost |
| CN110363077A (en) * | 2019-06-05 | 2019-10-22 | 平安科技（深圳）有限公司 | Sign Language Recognition Method, device, computer installation and storage medium |
| CN110458383A (en) * | 2019-06-24 | 2019-11-15 | 平安国际智慧城市科技股份有限公司 | Demand handles implementation method, device and the computer equipment of serviceization, storage medium |
| CN110472802A (en) * | 2018-05-09 | 2019-11-19 | 阿里巴巴集团控股有限公司 | A kind of data characteristics appraisal procedure, device and equipment |
| 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 |
| CN110909970A (en) * | 2018-09-17 | 2020-03-24 | 北京京东金融科技控股有限公司 | Credit scoring method and device |
| CN111047542A (en) * | 2019-12-31 | 2020-04-21 | 成都奥伦达科技有限公司 | Power line point supplementing method |
| CN111164633A (en) * | 2018-05-31 | 2020-05-15 | 重庆小雨点小额贷款有限公司 | Method and device for adjusting grading card model, server and storage medium |
| CN111724374A (en) * | 2020-06-22 | 2020-09-29 | 林晨 | Evaluation method of analysis result and terminal |
| CN111949640A (en) * | 2020-08-04 | 2020-11-17 | 上海微亿智造科技有限公司 | Intelligent parameter adjusting method and system based on industrial big data |
| CN112085595A (en) * | 2020-09-27 | 2020-12-15 | 中国建设银行股份有限公司 | Credit scoring model monitoring method and device |
| CN112115955A (en) * | 2019-06-20 | 2020-12-22 | 中科聚信信息技术（北京）有限公司 | Data processing method, data processing device and electronic equipment |
| CN112258312A (en) * | 2020-10-16 | 2021-01-22 | 银联商务股份有限公司 | Personal credit scoring method and system, electronic device and storage medium |
| CN112365186A (en) * | 2020-11-27 | 2021-02-12 | 中国电建集团海外投资有限公司 | Health degree evaluation method and system for electric power information system |
| CN112365104A (en) * | 2020-12-07 | 2021-02-12 | 杭州师范大学 | Marital matching method for predicting maximum marital satisfaction |
| CN112580252A (en) * | 2020-11-26 | 2021-03-30 | 北京航天智造科技发展有限公司 | Rotor drop-out fault diagnosis and analysis method and device |
| CN112613157A (en) * | 2020-11-26 | 2021-04-06 | 北京航天智造科技发展有限公司 | Rotor fault analysis method and device |
| CN112700280A (en) * | 2020-12-31 | 2021-04-23 | 上海竞动科技有限公司 | Short-term discontinuous user behavior evaluation method and device |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070198401A1 (en) * | 2006-01-18 | 2007-08-23 | Reto Kunz | System and method for automatic evaluation of credit requests |
| CN101145171A (en) * | 2007-09-15 | 2008-03-19 | 中国科学院合肥物质科学研究院 | Gene microarray data predication method based on independent component integrated study |
| CN101944122A (en) * | 2010-09-17 | 2011-01-12 | 浙江工商大学 | Incremental learning-fused support vector machine multi-class classification method |
| CN104574220A (en) * | 2015-01-30 | 2015-04-29 | 国家电网公司 | Power customer credit assessment method based on least square support vector machine |
| CN104820716A (en) * | 2015-05-21 | 2015-08-05 | 中国人民解放军海军工程大学 | Equipment reliability evaluation method based on data mining |
| CN105354210A (en) * | 2015-09-23 | 2016-02-24 | 深圳市爱贝信息技术有限公司 | Mobile game payment account behavior data processing method and apparatus |
| CN106204246A (en) * | 2016-08-18 | 2016-12-07 | 易联众信息技术股份有限公司 | A kind of BP neutral net credit estimation method based on PCA |
| CN106296389A (en) * | 2016-07-28 | 2017-01-04 | 联动优势科技有限公司 | The appraisal procedure of a kind of user credit degree and device |
| CN106919706A (en) * | 2017-03-10 | 2017-07-04 | 广州视源电子科技股份有限公司 | Data updating method and device |
-
2017
- 2017-09-04 CN CN201710785991.2A patent/CN107633265B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070198401A1 (en) * | 2006-01-18 | 2007-08-23 | Reto Kunz | System and method for automatic evaluation of credit requests |
| CN101145171A (en) * | 2007-09-15 | 2008-03-19 | 中国科学院合肥物质科学研究院 | Gene microarray data predication method based on independent component integrated study |
| CN101944122A (en) * | 2010-09-17 | 2011-01-12 | 浙江工商大学 | Incremental learning-fused support vector machine multi-class classification method |
| CN104574220A (en) * | 2015-01-30 | 2015-04-29 | 国家电网公司 | Power customer credit assessment method based on least square support vector machine |
| CN104820716A (en) * | 2015-05-21 | 2015-08-05 | 中国人民解放军海军工程大学 | Equipment reliability evaluation method based on data mining |
| CN105354210A (en) * | 2015-09-23 | 2016-02-24 | 深圳市爱贝信息技术有限公司 | Mobile game payment account behavior data processing method and apparatus |
| CN106296389A (en) * | 2016-07-28 | 2017-01-04 | 联动优势科技有限公司 | The appraisal procedure of a kind of user credit degree and device |
| CN106204246A (en) * | 2016-08-18 | 2016-12-07 | 易联众信息技术股份有限公司 | A kind of BP neutral net credit estimation method based on PCA |
| CN106919706A (en) * | 2017-03-10 | 2017-07-04 | 广州视源电子科技股份有限公司 | Data updating method and device |
Cited By (37)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
| CN110472802A (en) * | 2018-05-09 | 2019-11-19 | 阿里巴巴集团控股有限公司 | A kind of data characteristics appraisal procedure, device and equipment |
| CN110472802B (en) * | 2018-05-09 | 2023-12-01 | 创新先进技术有限公司 | Data characteristic evaluation method, device and equipment |
| CN108960505A (en) * | 2018-05-31 | 2018-12-07 | 试金石信用服务有限公司 | Quantitative estimation method, device, system and the storage medium of personal finance credit |
| CN111164633B (en) * | 2018-05-31 | 2024-01-05 | 重庆小雨点小额贷款有限公司 | Method and device for adjusting scoring card model, server and storage medium |
| CN111164633A (en) * | 2018-05-31 | 2020-05-15 | 重庆小雨点小额贷款有限公司 | Method and device for adjusting grading card model, server and storage medium |
| CN108711103A (en) * | 2018-06-04 | 2018-10-26 | 中国平安人寿保险股份有限公司 | Personal loan repays Risk Forecast Method, device, computer equipment and medium |
| CN108985341A (en) * | 2018-06-26 | 2018-12-11 | 四川斐讯信息技术有限公司 | A kind of the training set appraisal procedure and system of neural network model |
| CN108898479A (en) * | 2018-06-28 | 2018-11-27 | 中国农业银行股份有限公司 | The construction method and device of Credit Evaluation Model |
| CN109325020A (en) * | 2018-08-20 | 2019-02-12 | 中国平安人寿保险股份有限公司 | Small sample application method, device, computer equipment and storage medium |
| CN109389030A (en) * | 2018-08-23 | 2019-02-26 | 平安科技（深圳）有限公司 | Facial feature points detection method, apparatus, computer equipment and storage medium |
| CN110909970A (en) * | 2018-09-17 | 2020-03-24 | 北京京东金融科技控股有限公司 | Credit scoring method and device |
| CN109408583A (en) * | 2018-09-25 | 2019-03-01 | 平安科技（深圳）有限公司 | Data processing method and device, computer readable storage medium, electronic equipment |
| CN109409672A (en) * | 2018-09-25 | 2019-03-01 | 深圳市元征科技股份有限公司 | A kind of auto repair technician classifies grading modeling method and device |
| CN109408583B (en) * | 2018-09-25 | 2023-04-07 | 平安科技（深圳）有限公司 | Data processing method and device, computer readable storage medium and electronic equipment |
| CN109480864A (en) * | 2018-10-26 | 2019-03-19 | 首都医科大学附属北京安定医院 | A kind of schizophrenia automatic evaluation system based on nervous functional defects and machine learning |
| CN109583590A (en) * | 2018-11-29 | 2019-04-05 | 深圳和而泰数据资源与云技术有限公司 | Data processing method and data processing equipment |
| CN109583590B (en) * | 2018-11-29 | 2020-11-13 | 深圳和而泰数据资源与云技术有限公司 | Data processing method and data processing device |
| CN110162995A (en) * | 2019-04-22 | 2019-08-23 | 阿里巴巴集团控股有限公司 | Assess the method and device thereof of contribution data degree |
| CN110162995B (en) * | 2019-04-22 | 2023-01-10 | 创新先进技术有限公司 | Method and device for evaluating data contribution degree |
| CN110363077A (en) * | 2019-06-05 | 2019-10-22 | 平安科技（深圳）有限公司 | Sign Language Recognition Method, device, computer installation and storage medium |
| CN112115955A (en) * | 2019-06-20 | 2020-12-22 | 中科聚信信息技术（北京）有限公司 | Data processing method, data processing device and electronic equipment |
| CN110458383A (en) * | 2019-06-24 | 2019-11-15 | 平安国际智慧城市科技股份有限公司 | Demand handles implementation method, device and the computer equipment of serviceization, storage medium |
| CN110348722A (en) * | 2019-07-01 | 2019-10-18 | 百维金科(上海)信息科技有限公司 | A kind of internet finance air control model based on XGBoost |
| 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 |
| CN111047542A (en) * | 2019-12-31 | 2020-04-21 | 成都奥伦达科技有限公司 | Power line point supplementing method |
| CN111047542B (en) * | 2019-12-31 | 2021-04-27 | 成都奥伦达科技有限公司 | Power line point supplementing method |
| CN111724374A (en) * | 2020-06-22 | 2020-09-29 | 林晨 | Evaluation method of analysis result and terminal |
| CN111724374B (en) * | 2020-06-22 | 2024-03-01 | 智眸医疗(深圳)有限公司 | Evaluation method and terminal of analysis result |
| CN111949640A (en) * | 2020-08-04 | 2020-11-17 | 上海微亿智造科技有限公司 | Intelligent parameter adjusting method and system based on industrial big data |
| CN112085595A (en) * | 2020-09-27 | 2020-12-15 | 中国建设银行股份有限公司 | Credit scoring model monitoring method and device |
| CN112258312A (en) * | 2020-10-16 | 2021-01-22 | 银联商务股份有限公司 | Personal credit scoring method and system, electronic device and storage medium |
| CN112580252A (en) * | 2020-11-26 | 2021-03-30 | 北京航天智造科技发展有限公司 | Rotor drop-out fault diagnosis and analysis method and device |
| CN112613157A (en) * | 2020-11-26 | 2021-04-06 | 北京航天智造科技发展有限公司 | Rotor fault analysis method and device |
| CN112365186A (en) * | 2020-11-27 | 2021-02-12 | 中国电建集团海外投资有限公司 | Health degree evaluation method and system for electric power information system |
| CN112365104A (en) * | 2020-12-07 | 2021-02-12 | 杭州师范大学 | Marital matching method for predicting maximum marital satisfaction |
| CN112700280A (en) * | 2020-12-31 | 2021-04-23 | 上海竞动科技有限公司 | Short-term discontinuous user behavior evaluation method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107633265B (en) | 2021-03-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107633265A (en) | For optimizing the data processing method and device of credit evaluation model | |
| Xie et al. | Blockchain financial investment based on deep learning network algorithm | |
| Keramati et al. | A proposed classification of data mining techniques in credit scoring | |
| CN107590735A (en) | Data digging method and device for credit evaluation | |
| CN110110981A (en) | A kind of credit rating Default Probability estimates and method for prewarning risk | |
| Chen | Classifying credit ratings for Asian banks using integrating feature selection and the CPDA-based rough sets approach | |
| CN109191282A (en) | Methods of marking and system are monitored in a kind of loan of Behavior-based control model | |
| CN108960833A (en) | A kind of abnormal transaction identification method based on isomery finance feature, equipment and storage medium | |
| CN111178675A (en) | LR-Bagging algorithm-based electric charge recycling risk prediction method, system, storage medium and computer equipment | |
| CN107633030A (en) | Credit estimation method and device based on data model | |
| CN110415111A (en) | Merge the method for logistic regression credit examination & approval with expert features based on user data | |
| CN107633455A (en) | Credit estimation method and device based on data model | |
| CN114048436A (en) | Construction method and construction device for forecasting enterprise financial data model | |
| CN111882420A (en) | Generation method of response rate, marketing method, model training method and device | |
| CN111709826A (en) | Target information determination method and device | |
| Ruyu et al. | A comparison of credit rating classification models based on spark-evidence from lending-club | |
| CN109102396A (en) | A kind of user credit ranking method, computer equipment and readable medium | |
| CN114202243A (en) | Engineering project management risk early warning method and system based on random forest | |
| CN111951050A (en) | Financial product recommendation method and device | |
| Yang et al. | AComNN: Attention enhanced Compound Neural Network for financial time-series forecasting with cross-regional features | |
| CN112241916A (en) | Personal credit risk default early warning method, device, equipment and storage medium | |
| Bae et al. | Combining models from neural networks and inductive learning algorithms | |
| Jiang et al. | On the build and application of bank customer churn warning model | |
| Puglia et al. | Neural networks, the treasury yield curve, and recession forecasting | |
| Hui et al. | The model and empirical research of application scoring based on data mining methods |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CP02 | Change in the address of a patent holder | ||
| CP02 | Change in the address of a patent holder |
Address after: 518000 2203/2204, Building 1, Huide Building, Beizhan Community, Minzhi Street, Longhua District, Shenzhen, Guangdong Patentee after: SHENZHEN AUDAQUE DATA TECHNOLOGY Ltd. Address before: 518000 units J and K, 12 / F, block B, building 7, Baoneng Science Park, Qinghu Industrial Zone, Qingxiang Road, Longhua New District, Shenzhen City, Guangdong Province Patentee before: SHENZHEN AUDAQUE DATA TECHNOLOGY Ltd. |