CN108961040A - Loan limit assessment system and method for credit extension loan - Google Patents
Loan limit assessment system and method for credit extension loan Download PDFInfo
- Publication number
- CN108961040A CN108961040A CN201810712237.0A CN201810712237A CN108961040A CN 108961040 A CN108961040 A CN 108961040A CN 201810712237 A CN201810712237 A CN 201810712237A CN 108961040 A CN108961040 A CN 108961040A
- Authority
- CN
- China
- Prior art keywords
- user
- credit
- loan
- layer
- portrait
- 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
Abstract
The invention discloses a kind of loan limit assessment systems and method for credit extension loan, it is related to financial services industry, abnormality detection model, first layer credit scoring model are constructed using user's portrait, the unusual customers that abnormality detection model excludes high fraud and low loan repayment capacity are first passed through, realize the preliminary screening to user;It recycles first layer credit scoring model to score the credit score of user and postsearch screening is carried out to user, reuse second layer credit scoring model and the division of score section is carried out to the credit score of user, realize and the amount of different score section users is quantified.The present invention solves the problems, such as that the prior art assesses business standing with loan repayment capacity inaccurate, is mainly used for carrying out precision assessment and quantization to the loan limit of the user of different credit scores.
Description
Technical field
It is the present invention relates to financial services industry, in particular to a kind of for the loan limit assessment system of credit extension loan and side
Method.
Background technique
The accrediting amount refers to that business bank is the stock control index for the short-term giving credit that client appraises and decides, and generally can be divided into
The single loan accrediting amount, borrowing enterprise's amount and borrowing enterprise of group amount.As long as credit remaining sum is no more than corresponding business
Indicator of product variety, no matter it is accumulative provide the amount of money and provide number be it is how many, commercial banking department can be quickly to client's offer
The short-term credit fund of bank can be easily recycled in short-term credit, i.e. enterprise, so that it is quick to financial service to meet client
The requirement of property and convenience.
Personal loan amount currently on the market, especially personal credit extension loan amount is commented by manually evaluating credit
Grade, it is artificial to calculate loan or the accrediting amount.Existing personal loan or credit decision need first to grade to client, then people
Work measuring and calculating loan or the accrediting amount, relatively cumbersome, human intervention factor is more, and error is larger, ultimately cause to business standing with
The assessment of loan repayment capacity is inaccurate, and causing cannot be to the fine control of the accurate credit of client and credit risk.
Summary of the invention
The invention is intended to provide a kind of loan limit assessment system and method for credit extension loan, realize to different credits
The precision of the loan limit of the user of score is assessed and quantization.
In order to solve the above technical problems, base case provided by the invention is as follows:
User, which draws a portrait, obtains module: for collecting the relevant characteristic of user and establishing user behavior and attribute portrait;
Loan limit evaluation module: for determining user's loan value according to user behavior and attribute portrait building credit model
Degree.
In technical solution of the present invention, user, which draws a portrait, to be obtained the relevant characteristic of module collection user and establishes user's row
To draw a portrait with attribute, for example, relevant characteristic may include transaction data, the commodity for obtaining user by e-commerce platform
Data, behavioral data and log-on data;Loan limit evaluation module is determined according to user behavior and attribute portrait building credit model
Determine user's loan limit, to reach through credit model precise quantification user's loan limit, realizes bank to credit risk
Control.
Further, further includes:
Abnormal user excludes module: for excluding high fraud by user behavior and attribute portrait building abnormality detection model
With the unusual customers of low loan repayment capacity;
User's screening module: for constructing first layer credit scoring model to each user by user behavior and attribute portrait
Credit scoring is carried out, and user is screened according to the height of first layer credit score.
It is drawn a portrait by user and obtains the abnormal index of the characteristic building user of module collection and built by abnormal index
Mould carries out screening to suspicious trade company;Specifically, it is assumed that characteristic includes: Merchants register, login, order, payment and commodity class
Type data, then abnormal index is then are as follows: abnormal login number, abnormal order ratio, order volume and merchant type do not meet degree,
Order volume deviates trade company's historical level, then show that every abnormal index deviates standard using data mining and machine learning method
Total abnormal coefficient of each trade company is calculated in every abnormal coefficient weighted sum by the abnormal coefficient of value;Again by the total of each user
Abnormal coefficient is compared with abnormal coefficient threshold, and excludes the user that total abnormal coefficient is greater than abnormal coefficient threshold, that is, is reached
The unusual customers for excluding high fraud and low loan repayment capacity are arrived.
User's screening module carries out each user by user behavior and attribute portrait building first layer credit scoring model
Credit scoring, and user is screened according to the height of first layer credit score, for example, to user's hair more than set score
Loan is invited out, and excludes the user of low score, the loan repayment capacity of user can be measured according to first layer credit score, to refuse
The user of exhausted riskier loans, more reasonably carries out risk control.
Further, the loan limit evaluation module includes:
User, which draws a portrait, supplements submodule: behavior and the attribute of user are supplemented according to the people's row reference and main strategies of user
Portrait;
Rating Model constructs submodule: according to the behavior of the user after supplement and attribute portrait building second layer credit scoring
Model and the second layer credit score for calculating each user;
Loan limit determines submodule: the second layer credit score of each user being divided score section and determines that user provides a loan
Amount and time limit.
Behavior and the attribute portrait that user is supplemented according to the people's row reference and main strategies of user, make user's portrait more
Effectively, second layer credit scoring model completely, is constructed according to the behavior of the user after supplement and attribute portrait and calculates each user
Second layer credit score, realize and fining scoring carried out to the credit of client, then the second layer credit score of each user is drawn
Divide score section, the division further according to score section determines user's loan limit and time limit, and the reduction accrediting amount is excessively high and to visitor
The case where family loan repayment capacity overestimate, further control credit risk.
Further, user's screening module includes:
Screening model constructs submodule: being used for maintenance data excavation and machine learning algorithm for user behavior portrait and attribute
It draws a portrait and calculates the first layer credit score of each user as characteristic index building first layer scorecard model;
User excludes submodule: credit score threshold value is preset, for by the first layer credit score and credit of user
Score threshold compares, and filters out the user that first layer credit score is higher than credit score threshold value.
Characteristic index can include: merchant type, abnormal transaction index, abnormal login index, growth trend, internet use
Habit, Platform Dependent degree, whether chain, regional ranking and regional ranking of the same trade or business etc., the characteristic index meeting basis of all modelings should
Index in a model arranges the separating capacity of fine or not client, preferential to choose that predictive ability is most strong and stability is strongest
10-16 index is for modeling.First layer scorecard model calculates each user using the algorithm that logistic regression and decision tree combine
First layer credit score;The first layer credit score of user and pre-set credit score threshold value are compared again, into
And realize the screening to user, to exclude the user of low score.
Another object of the present invention is to provide a kind of loan limit appraisal procedure for credit extension loan, this method be based on
Upper system, method includes the following steps:
User's portrait obtaining step: it collects the relevant characteristic of user and simultaneously establishes user behavior and attribute portrait;
Loan limit appraisal procedure: user's loan limit is determined according to user behavior and attribute portrait building credit model.
Further, user's portrait obtaining step and the accrediting amount determine between step and include:
Abnormal user exclusion step: it is cheated by user behavior and attribute portrait building abnormality detection model exclusion height and low
The unusual customers of loan repayment capacity;
User's screening step: each user is carried out by user behavior and attribute portrait building first layer credit scoring model
Credit scoring, and user is screened according to the height of first layer credit score.
Further, the loan limit appraisal procedure includes:
S1: the behavior of user is supplemented according to the people's row reference and main strategies of user and attribute is drawn a portrait;
S2: according to the behavior of the user after supplement and attribute portrait building second layer credit scoring model and each user is calculated
Second layer credit score;
S3: the second layer credit score of each user is divided into score section and determines user's loan limit and time limit.
Further, user's screening step specifically includes:
Step 1: maintenance data excavates and user behavior portrait and attribute are drawn a portrait and be used as characteristic index by machine learning algorithm
Building first layer scorecard model calculates the first layer credit score of each user;
Step 2: presetting credit score threshold value, and the first layer credit score of user and credit score threshold value are carried out
Comparison filters out the user that first layer credit score is higher than credit score threshold value.
Detailed description of the invention
Fig. 1 is schematic block diagram of the present invention for the loan limit assessment system embodiment of credit extension loan;
Fig. 2 is flow chart of the present invention for the loan limit appraisal procedure embodiment of credit extension loan.
Specific embodiment
It is further described below by specific embodiment:
As shown in Figure 1, the present invention is used for the loan limit assessment system of credit extension loan, comprising:
User, which draws a portrait, obtains module: for collecting the relevant characteristic of user and establishing user behavior and attribute portrait;
Abnormal user excludes module: for excluding high fraud by user behavior and attribute portrait building abnormality detection model
With the unusual customers of low loan repayment capacity;
User's screening module: for constructing first layer credit scoring model to each user by user behavior and attribute portrait
Credit scoring is carried out, and user is screened according to the height of first layer credit score.
Loan limit evaluation module: for determining user's loan value according to user behavior and attribute portrait building credit model
Degree.
In the present embodiment, user's screening module includes:
Screening model constructs submodule: being used for maintenance data excavation and machine learning algorithm for user behavior portrait and attribute
It draws a portrait and calculates the first layer credit score of each user as characteristic index building first layer scorecard model;
User excludes submodule: credit score threshold value is preset, for by the first layer credit score and credit of user
Score threshold compares, and filters out the user that first layer credit score is higher than credit score threshold value.
Loan limit evaluation module includes:
User, which draws a portrait, supplements submodule: behavior and the attribute of user are supplemented according to the people's row reference and main strategies of user
Portrait;
Rating Model constructs submodule: according to the behavior of the user after supplement and attribute portrait building second layer credit scoring
Model and the second layer credit score for calculating each user;
Loan limit determines submodule: the second layer credit score of each user being divided score section and determines that user provides a loan
Amount and time limit.
This is used for the principle and usage scenario of the loan limit assessment system of credit extension loan are as follows:
One, unusual customers screening
User, which draws a portrait, to be obtained the relevant characteristic of module collection user and establishes user behavior and attribute portrait, for example,
Relevant characteristic may include transaction data, commodity data, behavioral data and the note that user is obtained by e-commerce platform
Volumes evidence and history applicant's loaning bill amount and refund situation data etc.;It is drawn a portrait by user and obtains the feature of module collection
Data construct the abnormal index of user and are modeled by abnormal index, carry out screening to suspicious trade company;Specifically, it is assumed that characteristic
According to including: Merchants register, login, order, payment and type of merchandise data, then abnormal index is then are as follows: abnormal login number, different
Normal order ratio, order volume and merchant type do not meet degree, order volume deviates trade company's historical level, and characteristic is as judgement
Every abnormal index deviates the basis of standard value;Then show that every abnormal index is inclined using data mining and machine learning method
Total abnormal coefficient of each trade company is calculated in every abnormal coefficient weighted sum by the abnormal coefficient from standard value;Again by each use
Total abnormal coefficient at family is compared with pre-set abnormal coefficient threshold, and is excluded total abnormal coefficient and be greater than abnormal coefficient
The user of threshold value has reached the unusual customers for excluding high fraud and low loan repayment capacity;
It specifically, can using the algorithm in data mining and machine learning method are as follows: found out using K arest neighbors sorting algorithm
Sample closest sample class determines sample generic, by the influence around adjacent to sample to the sample according to weighted value
Size and the mode that distance the is inversely proportional weight that give the sample different, are finally returned;Using clustering algorithm, by sample point
Class makes similar sample homogeney maximize inhomogeneity sample heterogeneity simultaneously larger;It is divided using the decision Tree algorithms of tree-shaped
The weight of each attribute, while finding outlier, remove the serious part that peels off;Each mind is connected using the algorithm of neural network
Through member and carry out weight division.
Two, low scoring users are excluded
User's screening module carries out each user by user behavior and attribute portrait building first layer credit scoring model
Credit scoring, and user is screened according to the height of first layer credit score, specifically, maintenance data excavates and engineering
Practise user behavior portrait and attribute portrait are constructed that first layer scorecard model calculates each user as characteristic index by algorithm the
One layer of credit score;The first layer credit score of user and pre-set credit score threshold value are compared, filter out
One layer of credit score is higher than the user of credit score threshold value, for example, credit score threshold value is 70 points, the user to 70 or more is issued
Loan is invited, and excludes 70 users below, and the loan repayment capacity of user can be tentatively measured according to first layer credit score, thus
Refuse the user of riskier loans, more reasonably carries out risk control.
First layer scorecard modeling process are as follows: modeling the characteristic index used includes: merchant type, abnormal transaction
Whether chain index, abnormal login index, growth trend, internet use habit, Platform Dependent degree, regional ranking, area be same
Industry ranking etc..The characteristic index of all modelings in a model can arrange the separating capacity of fine or not client according to the index
Column, preferential selection predictive ability is most strong and the strongest 10-16 index of stability is for modeling.Model is using logistic regression and certainly
The algorithm that plan tree combines.Specifically, periodically fine or not sample is had according to history applicant's loaning bill amount and the selection of refund situation,
Fine or not client is defined, available 10-16 variable is defined according to information content and group's stability indicator value, with logistic regression and certainly
Model cut-off threshold value is arranged, using model K-S test value and receiver operating characteristic curve to model in plan tree development model
Accuracy stability is assessed, and is adjusted after assessment to model, and adjustment is operated until model K-S test value and recipient
Indicatrix generates scorecard after reaching expected, and the first layer credit score of each user is provided further according to scorecard.
Three, second layer credit score score section divides
For the user that first layer Rating Model passes through, people's row reference and main strategies data are run, according to the people of user
The behavior of row reference and main strategies supplement user and attribute portrait, the index feature supplemented, specifically can include: enterprise
Main marital status, education landscape, house vehicle mortgage situation, identity card whether high risk zone, bank card wallet position, mobile phone
Number service condition, current loan situation, if bull debt-credit, if relate to and tell, tax affairs etc. keep user's portrait more complete
It is face, complete, according to the behavior of the user after supplement and attribute portrait building second layer credit scoring model and calculate each user's
Second layer credit score;
Second layer scorecard modeling process are as follows: used the characteristic index of supplement and first layer Rating Model
Characteristic index is combined together, and is arranged again to the separating capacity of fine or not client in a model according to all characteristic indexs
Column, preferential predictive ability of choosing most are established with the strongest 10-16 index of stability for second layer Rating Model by force.Model according to
The algorithm so combined using logistic regression and decision tree, it is consistent with first layer Rating Model to establish overall process basic step.It obtains
Scorecard provide the second layer credit score of each client, and carry out the score section area more more careful than first layer credit score
Point, for example, the present embodiment divides score Duan Weiwu, 70~75 graduation are divided into the first score section, 75~80 graduation are divided into the
Two score sections, 80~85 graduation are divided into third score section, and 85~90 graduation are divided into the 4th score section, and 90~95 graduation are divided into the 5th
Score section.
Four, amount determines
User's loan limit and time limit are determined according to the division of score section, and specifically, the height of score section and bank authorize
Amount it is directly proportional.Meanwhile the client of balloon score section obtains the more time limit options of the client lower than score section.To allow client
Authorization amount and time limit reasonably determined, it is excessively high and the case where to client's loan repayment capacity overestimate to reduce the accrediting amount,
Further control credit risk.
For the clearer course of work for illustrating the loan grade assessment system for credit extension loan of the invention, this reality
It applies in example, also discloses a kind of loan grade appraisal procedure for credit extension loan, this method is based on system above, such as Fig. 2 institute
Show, method includes the following steps:
User's portrait obtaining step: it collects the relevant characteristic of user and simultaneously establishes user behavior and attribute portrait;
Abnormal user exclusion step: it is cheated by user behavior and attribute portrait building abnormality detection model exclusion height and low
The unusual customers of loan repayment capacity;
User's screening step: each user is carried out by user behavior and attribute portrait building first layer credit scoring model
Credit scoring, and user is screened according to the height of first layer credit score;
Loan limit appraisal procedure: user's loan limit is determined according to user behavior and attribute portrait building credit model.
Specifically, user's screening step specifically includes:
Step 1: maintenance data excavates and user behavior portrait and attribute are drawn a portrait and be used as characteristic index by machine learning algorithm
Building first layer scorecard model calculates the first layer credit score of each user;
Step 2: presetting credit score threshold value, and the first layer credit score of user and credit score threshold value are carried out
Comparison filters out the user that first layer credit score is higher than credit score threshold value.
Loan limit appraisal procedure includes:
S1: the behavior of user is supplemented according to the people's row reference and main strategies of user and attribute is drawn a portrait;
S2: according to the behavior of the user after supplement and attribute portrait building second layer credit scoring model and each user is calculated
Second layer credit score;
S3: the second layer credit score of each user is divided into score section and determines user's loan limit and time limit.
In conclusion the beneficial effects of the present invention are:
One, abnormality detection model, first layer credit scoring model are constructed using user's portrait, first passes through abnormality detection mould
Type excludes the unusual customers of high fraud and low loan repayment capacity, realizes the preliminary screening to user;Recycle first layer credit scoring
Model scores to the credit score of user and carries out postsearch screening to user, and screening is realized to the comprehensive of abnormal user twice
Detection, more reasonably controls credit risk.
Two, the division of score section is carried out using credit score of the second layer credit scoring model to user, realized to difference point
The amount quantization of several sections of users, divides by the score section of foundation of credit score, client authorization amount and time limit is allowed to obtain rationally
Determination, it is excessively high to the accrediting amount and the case where to client's loan repayment capacity overestimate to reduce bank, further realizes to loan
The control of risk.
What has been described above is only an embodiment of the present invention, and the common sense such as well known specific structure and characteristic are not made herein in scheme
Excessive description.It, without departing from the structure of the invention, can be with it should be pointed out that for those skilled in the art
Several modifications and improvements are made, these also should be considered as protection scope of the present invention, these all will not influence what the present invention was implemented
Effect and patent practicability.The scope of protection required by this application should be based on the content of the claims, in specification
The records such as specific embodiment can be used for explaining the content of claim.
Claims (8)
1. being used for the loan limit assessment system of credit extension loan characterized by comprising
User, which draws a portrait, obtains module: for collecting the relevant characteristic of user and establishing user behavior and attribute portrait;
Loan limit evaluation module: for determining user's loan limit according to user behavior and attribute portrait building credit model.
2. the loan limit assessment system according to claim 1 for credit extension loan, which is characterized in that further include:
Abnormal user excludes module: for excluding high fraud and low by user behavior and attribute portrait building abnormality detection model
The unusual customers of loan repayment capacity;
User's screening module: for being carried out by user behavior and attribute portrait building first layer credit scoring model to each user
Credit scoring, and user is screened according to the height of first layer credit score.
3. the loan limit assessment system according to claim 2 for credit extension loan, which is characterized in that the loan value
Spending evaluation module includes:
User, which draws a portrait, supplements submodule: supplementing the behavior of user according to the people's row reference and main strategies of user and attribute is drawn
Picture;
Rating Model constructs submodule: according to the behavior of the user after supplement and attribute portrait building second layer credit scoring model
And calculate the second layer credit score of each user;
Loan limit determines submodule: the second layer credit score of each user being divided score section and determines user's loan limit
And the time limit.
4. the loan limit assessment system according to claim 2 for credit extension loan, which is characterized in that user's sieve
Modeling block includes:
Screening model constructs submodule: excavating for maintenance data and machine learning algorithm draws a portrait user behavior portrait and attribute
The first layer credit score of each user is calculated as characteristic index building first layer scorecard model;
User excludes submodule: credit score threshold value is preset, for by the first layer credit score and credit score of user
Threshold value compares, and filters out the user that first layer credit score is higher than credit score threshold value.
5. being used for the loan limit appraisal procedure of credit extension loan, which comprises the following steps:
User's portrait obtaining step: it collects the relevant characteristic of user and simultaneously establishes user behavior and attribute portrait;
Loan limit appraisal procedure: user's loan limit is determined according to user behavior and attribute portrait building credit model.
6. the loan limit appraisal procedure according to claim 5 for credit extension loan, which is characterized in that the user draws
Include: as obtaining step and the accrediting amount determine between step
Abnormal user excludes step: excluding high fraud and low refund by user behavior and attribute portrait building abnormality detection model
The unusual customers of ability;
User's screening step: credit is carried out to each user by user behavior and attribute portrait building first layer credit scoring model
Scoring, and user is screened according to the height of first layer credit score.
7. the loan limit appraisal procedure according to claim 6 for credit extension loan, which is characterized in that the loan value
Spending appraisal procedure includes:
S1: the behavior of user is supplemented according to the people's row reference and main strategies of user and attribute is drawn a portrait;
S2: according to the behavior of the user after supplement and attribute portrait building second layer credit scoring model and the of each user is calculated
Two layers of credit score;
S3: the second layer credit score of each user is divided into score section and determines user's loan limit and time limit.
8. the loan limit appraisal procedure according to claim 6 for credit extension loan, which is characterized in that user's sieve
Step is selected to specifically include:
Step 1: maintenance data excavates and machine learning algorithm constructs user behavior portrait and attribute portrait as characteristic index
First layer scorecard model calculates the first layer credit score of each user;
Step 2: presetting credit score threshold value, and the first layer credit score of user is compared with credit score threshold value,
Filter out the user that first layer credit score is higher than credit score threshold value.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810712237.0A CN108961040A (en) | 2018-06-29 | 2018-06-29 | Loan limit assessment system and method for credit extension loan |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810712237.0A CN108961040A (en) | 2018-06-29 | 2018-06-29 | Loan limit assessment system and method for credit extension loan |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108961040A true CN108961040A (en) | 2018-12-07 |
Family
ID=64484821
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810712237.0A Pending CN108961040A (en) | 2018-06-29 | 2018-06-29 | Loan limit assessment system and method for credit extension loan |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108961040A (en) |
Cited By (20)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109685649A (en) * | 2018-12-28 | 2019-04-26 | 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 |
CN109816509A (en) * | 2018-12-14 | 2019-05-28 | 平安科技(深圳)有限公司 | Generation method, terminal device and the medium of scorecard model |
CN109859051A (en) * | 2019-01-15 | 2019-06-07 | 国网电子商务有限公司 | A kind of financing method and device |
CN110020920A (en) * | 2019-04-03 | 2019-07-16 | 爱数科技服务(北京)有限公司 | Credit estimation method, device and storage medium |
CN110135973A (en) * | 2019-04-23 | 2019-08-16 | 北京淇瑀信息科技有限公司 | A kind of intelligent credit method based on IM and intelligent credit device |
CN110135700A (en) * | 2019-04-23 | 2019-08-16 | 北京淇瑀信息科技有限公司 | Credit Risk Assessment method and device based on expandtabs data |
CN110223163A (en) * | 2019-06-06 | 2019-09-10 | 杭州奕奕网络科技有限公司 | Intermediary's management system and counter cheat data analysing method |
CN110223076A (en) * | 2019-05-24 | 2019-09-10 | 东方银谷(北京)科技发展有限公司 | For the data processing method and device of financial business risk control, server |
CN110363658A (en) * | 2019-07-16 | 2019-10-22 | 北京明略软件系统有限公司 | Processing method and processing device, storage medium and the electronic device of credit data |
CN110807527A (en) * | 2019-09-30 | 2020-02-18 | 北京淇瑀信息科技有限公司 | Line adjusting method and device based on guest group screening and electronic equipment |
CN110807653A (en) * | 2019-11-28 | 2020-02-18 | 北京淇瑀信息科技有限公司 | Method and device for screening users and electronic equipment |
CN111145006A (en) * | 2019-12-26 | 2020-05-12 | 南京三百云信息科技有限公司 | Automobile financial anti-fraud model training method and device based on user portrait |
CN111275545A (en) * | 2020-02-14 | 2020-06-12 | 中国建设银行股份有限公司 | Method, apparatus, device and medium for on-line mortgage |
CN112132677A (en) * | 2020-09-22 | 2020-12-25 | 北京思特奇信息技术股份有限公司 | Intelligent signal control and income evaluation method |
CN112348654A (en) * | 2020-09-23 | 2021-02-09 | 民生科技有限责任公司 | Automatic assessment method, system and readable storage medium for enterprise credit line |
CN112561691A (en) * | 2020-12-24 | 2021-03-26 | 中国农业银行股份有限公司 | Customer credit prediction method, device, equipment and storage medium |
CN112862298A (en) * | 2020-07-09 | 2021-05-28 | 北京睿知图远科技有限公司 | Credit assessment method for user portrait |
CN113177837A (en) * | 2021-05-12 | 2021-07-27 | 广州市全民钱包科技有限公司 | Loan amount evaluation method, device, equipment and storage medium for loan applicant |
TWI742528B (en) * | 2020-02-05 | 2021-10-11 | 玉山商業銀行股份有限公司 | Method and system for intelligently processing loan application |
CN113781198A (en) * | 2020-06-09 | 2021-12-10 | 台北富邦商业银行股份有限公司 | Enterprise loan application evaluation system |
-
2018
- 2018-06-29 CN CN201810712237.0A patent/CN108961040A/en active Pending
Cited By (22)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109816509A (en) * | 2018-12-14 | 2019-05-28 | 平安科技(深圳)有限公司 | Generation method, terminal device and the medium of scorecard model |
CN109685649A (en) * | 2018-12-28 | 2019-04-26 | 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 |
CN109859051A (en) * | 2019-01-15 | 2019-06-07 | 国网电子商务有限公司 | A kind of financing method and device |
CN110020920A (en) * | 2019-04-03 | 2019-07-16 | 爱数科技服务(北京)有限公司 | Credit estimation method, device and storage medium |
CN110135973A (en) * | 2019-04-23 | 2019-08-16 | 北京淇瑀信息科技有限公司 | A kind of intelligent credit method based on IM and intelligent credit device |
CN110135700A (en) * | 2019-04-23 | 2019-08-16 | 北京淇瑀信息科技有限公司 | Credit Risk Assessment method and device based on expandtabs data |
CN110223076A (en) * | 2019-05-24 | 2019-09-10 | 东方银谷(北京)科技发展有限公司 | For the data processing method and device of financial business risk control, server |
CN110223163A (en) * | 2019-06-06 | 2019-09-10 | 杭州奕奕网络科技有限公司 | Intermediary's management system and counter cheat data analysing method |
CN110363658A (en) * | 2019-07-16 | 2019-10-22 | 北京明略软件系统有限公司 | Processing method and processing device, storage medium and the electronic device of credit data |
CN110807527B (en) * | 2019-09-30 | 2023-11-14 | 北京淇瑀信息科技有限公司 | Credit adjustment method and device based on guest group screening and electronic equipment |
CN110807527A (en) * | 2019-09-30 | 2020-02-18 | 北京淇瑀信息科技有限公司 | Line adjusting method and device based on guest group screening and electronic equipment |
CN110807653A (en) * | 2019-11-28 | 2020-02-18 | 北京淇瑀信息科技有限公司 | Method and device for screening users and electronic equipment |
CN111145006A (en) * | 2019-12-26 | 2020-05-12 | 南京三百云信息科技有限公司 | Automobile financial anti-fraud model training method and device based on user portrait |
TWI742528B (en) * | 2020-02-05 | 2021-10-11 | 玉山商業銀行股份有限公司 | Method and system for intelligently processing loan application |
CN111275545A (en) * | 2020-02-14 | 2020-06-12 | 中国建设银行股份有限公司 | Method, apparatus, device and medium for on-line mortgage |
CN113781198A (en) * | 2020-06-09 | 2021-12-10 | 台北富邦商业银行股份有限公司 | Enterprise loan application evaluation system |
CN112862298A (en) * | 2020-07-09 | 2021-05-28 | 北京睿知图远科技有限公司 | Credit assessment method for user portrait |
CN112862298B (en) * | 2020-07-09 | 2024-02-27 | 北京睿知图远科技有限公司 | Credit evaluation method for user portrait |
CN112132677A (en) * | 2020-09-22 | 2020-12-25 | 北京思特奇信息技术股份有限公司 | Intelligent signal control and income evaluation method |
CN112348654A (en) * | 2020-09-23 | 2021-02-09 | 民生科技有限责任公司 | Automatic assessment method, system and readable storage medium for enterprise credit line |
CN112561691A (en) * | 2020-12-24 | 2021-03-26 | 中国农业银行股份有限公司 | Customer credit prediction method, device, equipment and storage medium |
CN113177837A (en) * | 2021-05-12 | 2021-07-27 | 广州市全民钱包科技有限公司 | Loan amount evaluation method, device, equipment and storage medium for loan applicant |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108961040A (en) | Loan limit assessment system and method for credit extension loan | |
CN108921689A (en) | Credit risk monitoring system and method | |
Berger et al. | Commercial imperialism? Political influence and trade during the Cold War | |
US8799150B2 (en) | System and method for predicting consumer credit risk using income risk based credit score | |
CN110009479A (en) | Credit assessment method and device, storage medium, computer equipment | |
CN108898480A (en) | Loan grade assessment system and method for credit extension loan | |
CN106845988A (en) | method and device for selecting payment channel | |
CN110458693A (en) | A kind of automatic measures and procedures for the examination and approval of business loan, device, storage medium and electronic equipment | |
Toloo et al. | Two alternative approaches for selecting performance measures in data envelopment analysis | |
Bravo et al. | Granting and managing loans for micro-entrepreneurs: New developments and practical experiences | |
CN108256691A (en) | Refund Probabilistic Prediction Model construction method and device | |
CN106651570A (en) | System and method for real-time loan approval | |
CN110135853A (en) | Clique's user identification method, device and equipment | |
CN106779996A (en) | Business data method for digging and system in a kind of supply chain finance | |
CN108596495A (en) | A kind of retail credit business points-scoring system and method | |
CN110675243A (en) | Machine learning-fused credit prediction overdue method and system | |
US20150363875A1 (en) | System and Method for Filtering and Analyzing Transaction Information | |
CN106855989A (en) | Risk monitoring and control method and system after one kind loan | |
US10929859B2 (en) | Systems and methods for determining economic impact of an event within a geographic area | |
US20210201402A1 (en) | Using psychometric analysis for determining credit risk | |
CN110009417A (en) | Target customer's screening technique, device, equipment and computer readable storage medium | |
CN107346515A (en) | A kind of credit card Forecasting Methodology and device by stages | |
CN112232950A (en) | Loan risk assessment method and device, equipment and computer-readable storage medium | |
KR101112755B1 (en) | System for managing insolvent obligation and the method therefor | |
KR20200130575A (en) | Customer credit rating system |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181207 |