CN113793132A - Automatic approval method, system and terminal based on machine learning - Google Patents
Automatic approval method, system and terminal based on machine learning Download PDFInfo
- Publication number
- CN113793132A CN113793132A CN202111118597.6A CN202111118597A CN113793132A CN 113793132 A CN113793132 A CN 113793132A CN 202111118597 A CN202111118597 A CN 202111118597A CN 113793132 A CN113793132 A CN 113793132A
- Authority
- CN
- China
- Prior art keywords
- user
- approval
- data
- fraud
- application
- 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
- 238000000034 method Methods 0.000 title claims abstract description 74
- 238000010801 machine learning Methods 0.000 title claims abstract description 20
- 230000002159 abnormal effect Effects 0.000 claims description 7
- 238000009825 accumulation Methods 0.000 claims description 3
- 230000002776 aggregation Effects 0.000 claims description 2
- 238000004220 aggregation Methods 0.000 claims description 2
- 230000005856 abnormality Effects 0.000 abstract 1
- 239000000284 extract Substances 0.000 description 2
- 238000013475 authorization Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/10—Office automation; Time management
- G06Q10/103—Workflow collaboration or project management
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- 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
- G06Q40/04—Trading; Exchange, e.g. stocks, commodities, derivatives or currency exchange
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Strategic Management (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- General Physics & Mathematics (AREA)
- Economics (AREA)
- Marketing (AREA)
- Entrepreneurship & Innovation (AREA)
- Data Mining & Analysis (AREA)
- General Business, Economics & Management (AREA)
- Finance (AREA)
- Software Systems (AREA)
- Accounting & Taxation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Technology Law (AREA)
- Development Economics (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
Abstract
The invention provides an automatic approval method, system and terminal based on machine learning, wherein the method comprises the following steps: sending the user data of the user applying for approval to a big data rule engine system, judging whether the user data is a financial fraud transaction or a fraud application or not by the big data rule engine system according to the user data of the user applying for, the accumulated anti-fraud data and the user data of the user applying for, and entering manual approval or refusing the application if the user data is the user data of the user applying for, and otherwise, sending the user data of the user applying for approval to the big data rule engine system; if not, entering an automatic approval process, rating the user according to the user data, grouping the user according to the rating, grading the user according to the user group to which the user belongs and the user data and providing a pre-granted credit limit, performing different process approvals on different user groups, extracting data related to corresponding process nodes according to a third-party data source and user application information, and approving the corresponding process nodes, and if an abnormality occurs in a certain process node, entering manual approval.
Description
Technical Field
The invention belongs to the technical field of automatic office work, and particularly relates to an automatic approval method, system and terminal based on machine learning.
Background
With the development of the times, science and technology are changing the living and consumption habits of people step by step, and the payment habits of people are also changed from cash payment to digital payment. With the increase of payment scenes and the complication of consumer groups, the credit review system has higher requirements. The existing approval method can not meet the requirements on speed and precision.
Disclosure of Invention
The present invention has been made to solve the above-described problems, and an object of the present invention is to provide a machine learning-based automated approval method, system, and terminal that have a high approval speed and can improve the approval accuracy in various scenes.
The invention provides an automatic approval method based on machine learning, which is characterized by comprising the following steps of:
sending the user data of the user applying for approval to a big data rule engine system, wherein the big data rule engine system obtains the fraud information reversely hit by the user according to the user data of the user already applying for, the accumulated anti-fraud data and the user data of the user applying for the time, judges whether the user is financial fraud transaction or fraud application according to the fraud information reversely hit by the user, and enters manual approval or application rejection if the user is financial fraud transaction or fraud application; if the transaction is not financial fraud transaction or fraud application, entering an automatic approval process, grading the user according to the user data by the big data rule engine system, grouping the user according to the grading, grading the user according to the user group to which the user belongs and the user data, providing corresponding pre-approval credit limit for the user according to the grading interval, approving different user groups in different processes, effectively integrating a third-party data source and user application card information at different process nodes, extracting ordered and effective data related to the corresponding process nodes from the fragmented information of the user, approving the corresponding process nodes according to the data, entering manual approval if the process nodes are abnormal, and determining whether the approval passes through in the last process node of the automatic approval process or the manual approval link in combination with the pre-approval credit limit,
the big data rule engine system is established according to a third-party data source for mastering credit records and personal conditions of the user and a wind control model strategy, and the wind control model strategy is established according to five principles of status of a claimant, fund usage, repayment source, debt guarantee and prospect of a borrowed user.
Further, in the automatic approval method based on machine learning provided by the present invention, the method may further have the following characteristics: the user data comprises basic data filled by the user and authorized basic information.
Further, in the automatic approval method based on machine learning provided by the present invention, the method may further have the following characteristics: the third-party data source comprises a public security system, a pedestrian report, a Bai Rou network, a Hodun, a GEO set and an Olympic aggregation and a virtual operator.
Further, in the automatic approval method based on machine learning provided by the present invention, the method may further have the following characteristics: and the wind control model is continuously adjusted according to the accumulation of the application quantity of the examination and approval users.
The invention also provides an automatic approval system, which is characterized by comprising the following components: the input module is used for inputting user data of a user applying for approval;
the anti-fraud module is used for obtaining anti-hit fraud information of the user according to the user data of the applied user, the accumulated anti-fraud data and the user data of the user approved by the application, and judging financial fraud transactions and fraud applications of the user according to the anti-hit fraud information of the user;
the grouping module is used for judging that the anti-fraud module is not a fraudulent transaction or an applied user and grouping the user into a corresponding user group according to the user data;
the pre-credit limit module is used for scoring the user according to the user group to which the user belongs and the user data and providing corresponding pre-credit limit for the user according to a scored interval;
the automatic approval module is used for carrying out approval of different processes aiming at different user groups, effectively integrating third-party data sources and user application card information at different process nodes, extracting ordered and effective data related to corresponding process nodes from fragmented information of users, carrying out approval on the corresponding process nodes according to the data, carrying out manual approval on the application of the user if any process node is abnormal, and determining whether the approval passes or not by combining with the pre-granted credit limit of the user at the last process node; and
and the manual approval module is used for manually approving the application users who have abnormity during the approval of the automatic approval module or the anti-fraud module judges the application users to be financial fraud transactions or fraud applications.
The invention also provides a terminal, comprising a memory and a processor, wherein the memory stores computer instructions capable of running on the processor, and the terminal is characterized in that: and when the processor runs the computer instruction, the automatic approval method based on the machine learning is executed.
The invention has the following advantages:
the automatic approval method based on machine learning has the advantages of high approval speed and high approval precision, and can reduce the cost of automatic approval.
Drawings
FIG. 1 is a flow chart of the automated machine learning-based approval method of the present invention.
FIG. 2 is a block diagram of the automated approval system of the present invention.
Detailed Description
In order to make the technical means, the creation features, the achievement objectives and the efficacy of the present invention easy to understand, the following embodiments specifically describe the automatic approval method, the system and the terminal based on machine learning in accordance with the present invention with reference to the accompanying drawings.
As shown in fig. 1, the automatic approval method based on machine learning includes the following steps:
and sending the user data of the user applying for approval to the big data rule engine system. The big data rule engine system is established according to a third-party data source for mastering credit records and personal conditions of the user and a wind control model strategy, wherein the wind control model strategy is established according to five principles of the status of a card applicant, the use of funds, the source of repayment, debt guarantee and the prospect of a borrowed user. Specifically, the third-party data source includes a public security system, a pedestrian report, a Bai Ji network, a Hodun, a GEO set and an Olympic, a virtual operator, and the like.
The big data rule engine system obtains fraud information which is reversely hit by the user according to the user data of the applied user, the accumulated anti-fraud data and the user data of the user applying for the time, judges whether the user is financial fraud transaction or fraud application or not according to the fraud information which is reversely hit by the user, and enters manual approval or application rejection if the user is financial fraud transaction or fraud application; if the transaction is not a financial fraud transaction or a fraud application, an automatic approval process is entered, the big data rule engine system grades the user according to the user data, grouping the users according to the rating, scoring the users according to the user group to which the users belong and the user data, providing corresponding pre-granted credit limit for the user according to the scored interval, carrying out different processes of examination and approval on different user groups, at different process nodes, the big data rule engine system effectively integrates the third-party data source and the user application information, extracts ordered and effective data related to the corresponding process nodes from the fragmented information of the user, examining and approving corresponding process nodes according to the data, if a certain process node is abnormal, and entering manual examination and approval, and determining whether the examination and approval pass or not by combining the final process node of the automatic examination and approval process or the manual examination and approval link with the pre-granted credit line. The method has the advantages of high approval speed and high approval precision. The grouping approval can reduce the cost of automatic approval and reduce the nodes of automatic approval. Providing different users with corresponding credit limits can reduce loss caused by default of special users.
Specifically, the user data includes basic data filled by the user and basic information of authorization.
Specifically, the wind control model is continuously adjusted according to the accumulation of the number of application of the approval users, so that the manual approval amount can be greatly reduced, instructive data can be provided for the manually approved approval business personnel, and the possible default risk can be evaluated and optimized.
Specifically, in the automatic approval process, progressive approval is adopted, the wind control model is refined, indexes of output are calculated through a big data rule engine system, and various data sources are subjected to hierarchical operation and are approved in times.
As shown in fig. 2, the embodiment of the present invention further discloses an automated approval system, which includes: the system comprises an input module 10, an anti-fraud module 20, a grouping module 30, a credit line pre-granting module 40, an automatic approval module 50 and a manual approval module 60.
The input module 10 is used for inputting user data of a user applying for approval.
The anti-fraud module 20 is configured to obtain fraud information that is anti-hit by the user according to the user data of the user who has applied for the transaction, the accumulated anti-fraud data, and the user data of the user who has approved the application, and determine financial fraud transactions and fraud applications of the user according to the fraud information that is anti-hit by the user.
The grouping module 30 determines that the anti-fraud module is not a fraudulent transaction or an applied user, and classifies the user into a corresponding user group according to the user data.
The pre-granted credit line module 40 scores the user according to the user group to which the user belongs and the user data, and provides the corresponding pre-granted credit line for the user according to the scored interval.
The automatic approval module 50 performs approval of different processes for different user groups, effectively integrates the third-party data source and the user application information at different process nodes, extracts ordered and effective data related to the corresponding process nodes from the fragmented information of the user, approves the corresponding process nodes according to the data, enters the manual approval module if any process node is abnormal, and determines whether the approval passes or not by combining the pre-granted credit limit of the user at the last process node.
The manual approval module 60 performs manual approval on an application user who determines that the automatic approval module is abnormal or the fraud-resisting module is judged to be financial fraud transaction or fraud application when the automatic approval module is approved, and determines whether the approval is passed or not by combining with the pre-granted credit limit of the user when the automatic approval module is approved.
The embodiment of the invention also discloses a terminal, which comprises a memory and a processor, wherein the memory stores computer instructions capable of running on the processor, and the terminal is characterized in that: and when the processor runs the computer instruction, the technical scheme of the automatic approval method based on the machine learning is executed.
The above embodiments are preferred examples of the present invention, and are not intended to limit the scope of the present invention.
Claims (6)
1. An automatic approval method based on machine learning is characterized by comprising the following steps:
sending the user data of the user applying for approval to a big data rule engine system, wherein the big data rule engine system obtains the fraud information reversely hit by the user according to the user data of the user already applying for, the accumulated anti-fraud data and the user data of the user applying for the time, judges whether the user is financial fraud transaction or fraud application according to the fraud information reversely hit by the user, and enters manual approval or application rejection if the user is financial fraud transaction or fraud application; if the transaction is not financial fraud transaction or fraud application, entering an automatic approval process, grading the user according to the user data by the big data rule engine system, grouping the user according to the grading, grading the user according to the user group to which the user belongs and the user data, providing corresponding pre-approval credit limit for the user according to the grading interval, approving different user groups in different processes, effectively integrating a third-party data source and user application card information at different process nodes, extracting ordered and effective data related to the corresponding process nodes from the fragmented information of the user, approving the corresponding process nodes according to the data, entering manual approval if the process nodes are abnormal, and determining whether the approval passes through in the last process node of the automatic approval process or the manual approval link in combination with the pre-approval credit limit,
the big data rule engine system is established according to a third-party data source for mastering credit records and personal conditions of the user and a wind control model strategy, and the wind control model strategy is established according to five principles of status of a claimant, fund usage, repayment source, debt guarantee and prospect of a borrowed user.
2. The machine learning-based automated approval method of claim 1, wherein:
the user data comprises basic data filled by the user and authorized basic information.
3. The machine learning-based automated approval method of claim 1, wherein:
the third-party data source comprises a public security system, a pedestrian report, a Bai Rou network, a Hodun, a GEO set and an Olympic aggregation and a virtual operator.
4. The machine learning-based automated approval method of claim 1, wherein:
and the wind control model is continuously adjusted according to the accumulation of the application quantity of the examination and approval users.
5. An automated approval system, comprising:
the input module is used for inputting user data of a user applying for approval;
the anti-fraud module is used for obtaining anti-hit fraud information of the user according to the user data of the applied user, the accumulated anti-fraud data and the user data of the user approved by the application, and judging financial fraud transactions and fraud applications of the user according to the anti-hit fraud information of the user;
the grouping module is used for judging that the anti-fraud module is not a fraudulent transaction or an applied user and grouping the user into a corresponding user group according to the user data;
the pre-credit limit module is used for scoring the user according to the user group to which the user belongs and the user data and providing corresponding pre-credit limit for the user according to a scored interval;
the automatic approval module is used for carrying out approval of different processes aiming at different user groups, effectively integrating third-party data sources and user application card information at different process nodes, extracting ordered and effective data related to corresponding process nodes from fragmented information of users, carrying out approval on the corresponding process nodes according to the data, carrying out manual approval on the application of the user if any process node is abnormal, and determining whether the approval passes or not by combining with the pre-granted credit limit of the user at the last process node; and
and the manual approval module is used for manually approving the application users who have abnormity during the approval of the automatic approval module or the anti-fraud module judges the application users to be financial fraud transactions or fraud applications.
6. A terminal comprising a memory and a processor, the memory having stored thereon computer instructions executable on the processor, the terminal comprising: the processor, when executing the computer instructions, performs the machine learning-based automated approval method of any one of claims 1-4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111118597.6A CN113793132A (en) | 2021-09-24 | 2021-09-24 | Automatic approval method, system and terminal based on machine learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111118597.6A CN113793132A (en) | 2021-09-24 | 2021-09-24 | Automatic approval method, system and terminal based on machine learning |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113793132A true CN113793132A (en) | 2021-12-14 |
Family
ID=79184253
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111118597.6A Pending CN113793132A (en) | 2021-09-24 | 2021-09-24 | Automatic approval method, system and terminal based on machine learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113793132A (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109146670A (en) * | 2018-08-27 | 2019-01-04 | 深圳前海微众银行股份有限公司 | It provides a loan anti-rogue processes method, apparatus and readable storage medium storing program for executing |
| CN109300028A (en) * | 2018-09-11 | 2019-02-01 | 上海天旦网络科技发展有限公司 | Real-time anti-fraud method and system and storage medium based on network data |
| CN109409502A (en) * | 2018-09-26 | 2019-03-01 | 深圳壹账通智能科技有限公司 | Generation method, device, equipment and the storage medium of anti-fraud model |
| CN109816390A (en) * | 2019-01-03 | 2019-05-28 | 深圳壹账通智能科技有限公司 | Anti- fraud analysis processing method, device and computer equipment based on transaction data |
| CN111523996A (en) * | 2020-04-21 | 2020-08-11 | 北京易点淘网络技术有限公司 | Approval method and system |
| CN112365341A (en) * | 2020-11-26 | 2021-02-12 | 四川享宇金信金融科技有限公司 | Credit agency anti-fraud method, apparatus, device and storage medium |
| CN112734247A (en) * | 2021-01-14 | 2021-04-30 | 上海尧信惠达信息科技有限公司 | Method, system, storage medium and electronic device for automatic approval of guarantee and credit |
-
2021
- 2021-09-24 CN CN202111118597.6A patent/CN113793132A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109146670A (en) * | 2018-08-27 | 2019-01-04 | 深圳前海微众银行股份有限公司 | It provides a loan anti-rogue processes method, apparatus and readable storage medium storing program for executing |
| CN109300028A (en) * | 2018-09-11 | 2019-02-01 | 上海天旦网络科技发展有限公司 | Real-time anti-fraud method and system and storage medium based on network data |
| CN109409502A (en) * | 2018-09-26 | 2019-03-01 | 深圳壹账通智能科技有限公司 | Generation method, device, equipment and the storage medium of anti-fraud model |
| CN109816390A (en) * | 2019-01-03 | 2019-05-28 | 深圳壹账通智能科技有限公司 | Anti- fraud analysis processing method, device and computer equipment based on transaction data |
| CN111523996A (en) * | 2020-04-21 | 2020-08-11 | 北京易点淘网络技术有限公司 | Approval method and system |
| CN112365341A (en) * | 2020-11-26 | 2021-02-12 | 四川享宇金信金融科技有限公司 | Credit agency anti-fraud method, apparatus, device and storage medium |
| CN112734247A (en) * | 2021-01-14 | 2021-04-30 | 上海尧信惠达信息科技有限公司 | Method, system, storage medium and electronic device for automatic approval of guarantee and credit |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11599939B2 (en) | System, method and computer program for underwriting and processing of loans using machine learning | |
| CN109035003A (en) | Anti- fraud model modelling approach and anti-fraud monitoring method based on machine learning | |
| US7689506B2 (en) | System and method for rapid updating of credit information | |
| US8799150B2 (en) | System and method for predicting consumer credit risk using income risk based credit score | |
| CN111476660B (en) | Intelligent wind control system and method based on data analysis | |
| Koh et al. | A two-step method to construct credit scoring models with data mining techniques | |
| CN110458693A (en) | A kind of automatic measures and procedures for the examination and approval of business loan, device, storage medium and electronic equipment | |
| Gatti et al. | Measuring value‐at‐risk in project finance transactions | |
| US20080021813A1 (en) | Method for scoring accounts for retention and marketing accounts based on retention and profitability | |
| CN111008896A (en) | Financial risk early warning method and device, electronic equipment and storage medium | |
| CN106296389A (en) | The appraisal procedure of a kind of user credit degree and device | |
| Eze et al. | Electronic banking and profitability of commercial banks in Nigeria | |
| US11715106B2 (en) | Systems and methods for real-time institution analysis based on message traffic | |
| CN112801529B (en) | Financial data analysis method and device, electronic equipment and medium | |
| CN113989019A (en) | Method, device, equipment and storage medium for identifying risks | |
| CN107657529A (en) | IFRS9 credit card is expected depreciation method | |
| CN113935844A (en) | Financial wind control system based on big data and artificial intelligence | |
| CN110910002B (en) | Account receivables default risk identification method and system | |
| US11410178B2 (en) | Systems and methods for message tracking using real-time normalized scoring | |
| CN112766814A (en) | Training method, device and equipment for credit risk pressure test model | |
| KR101142132B1 (en) | Calculation system of value at risk | |
| CN112037013A (en) | Pedestrian credit variable derivation method and device | |
| CN110458684A (en) | A kind of anti-fraud detection method of finance based on two-way shot and long term Memory Neural Networks | |
| CN113793132A (en) | Automatic approval method, system and terminal based on machine learning | |
| CN115564591A (en) | Financing product determination method and related equipment |
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 |