CN110751548A - User loan risk prediction method applied to intelligent bank - Google Patents
User loan risk prediction method applied to intelligent bank Download PDFInfo
- Publication number
- CN110751548A CN110751548A CN201910832603.0A CN201910832603A CN110751548A CN 110751548 A CN110751548 A CN 110751548A CN 201910832603 A CN201910832603 A CN 201910832603A CN 110751548 A CN110751548 A CN 110751548A
- Authority
- CN
- China
- Prior art keywords
- user
- features
- bank
- risk
- historical behavior
- 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.)
- Withdrawn
Links
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
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/03—Credit; Loans; Processing thereof
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Accounting & Taxation (AREA)
- Computational Mathematics (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Finance (AREA)
- Mathematical Physics (AREA)
- Evolutionary Biology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Probability & Statistics with Applications (AREA)
- Databases & Information Systems (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Operations Research (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Algebra (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- Technology Law (AREA)
- General Business, Economics & Management (AREA)
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
Abstract
The invention discloses a user loan risk prediction method applied to an intelligent bank, which comprises the following steps of 100, acquiring historical behavior data of a bank user; step 101, preprocessing historical behavior data of a user and cleaning user attributes; 102, selecting features based on an xgboost model, and reserving the first n features; step 103, establishing a Random Forest, GBDT, XGBOOST and LightGBM model; step 104, obtaining 4 models W1, W2, W3 and W4 after the step 103 is executed; 105, predicting the future overdue possibility of the user according to the historical behavior data of the user; step 106, outputting the possibility of overdue payment of the user; the invention is applied to the intelligent bank loan risk early warning platform, which insists on abandoning the management concept of 'heavy loan light management', improves the post-loan management working efficiency, grasps key risk points, improves the risk identification foresight and improves the risk resolving capability.
Description
Technical Field
The invention relates to a user loan risk prediction method applied to an intelligent bank.
Background
Under the drive of the internet plus, the dissimilarity of financial demeanour, interest rate marketization, non-financial institutions and internet finance forces the traditional bank management mode, business operation mode and network point service mode to be transformed into the direct marketing bank, internet bank, intelligent bank and the like.
As early as 2009, IBM has brought forward the idea of a smart website, and in a narrow sense, a smart bank refers to the improvement and upgrading of the website layout based on intelligent equipment, self-service equipment and the like, and in a broad sense, the smart bank utilizes mature intelligent technology and equipment to realize the reconstruction and remodeling of traditional business processes, operation modes, management modes and the like of the bank so as to improve service efficiency and customer experience; with the explosive development of new-generation information technologies represented by mobile internet, cloud computing, big data, and the like, internet finance is being silently developed in a unique state of business.
In the internet financial field, in view of the characteristics that consumption type financial service objects such as investment financing and loan payment services are small in amount, large in crowd and short in period, the consumption type financial service objects are generally recognized as the subdivision field with the highest risk, so that risk control is always the core foundation of the services, loan of the traditional financial industry needs a user to provide asset information, online network consumption behaviors of the user are integrated while assets under a user line are integrated through internet finance, various financial data are actively collected, analyzed and arranged by means of technologies such as artificial intelligence and big data, so that better service experience is provided for the user, and more comprehensive understanding and evaluation are provided for financial merchants to distinguish default risk users.
Disclosure of Invention
In view of the above technical problems, the present invention aims to: the user loan risk prediction method applied to the intelligent bank is provided, and the post-loan management work efficiency, the key risk point grasping, the risk identification foresight and the risk resolving capability are improved.
The technical solution of the invention is realized as follows: a user loan risk prediction method applied to an intelligent bank comprises the following steps of 100, obtaining historical behavior data, online data and offline data of a bank user; step 101, preprocessing historical behavior data of a user, cleaning user attributes, and constructing basic features, time sequence features, business features, combination features, discrete features and other hundreds-dimensional features through engineering operation; 102, selecting features based on an xgboost model by adopting a method that a learning process and a feature selection process are performed simultaneously, and reserving the first n features; 103, establishing four models of Random Forest, GBDT, XGBOST and LightGBM, randomly dividing original data into 5 subsets which are mutually intersected and have the same size by each model, selecting one of the subsets as a verification set and taking the remaining 4 as a training set, and thus, each subset has an opportunity to be used as a training set; step 104, obtaining 4 models W1, W2, W3 and W4 after the step 103 is executed; 105, predicting the future overdue possibility of the user according to the historical behavior data of the user; and step 106, outputting the possibility of overdue payment of the user.
Preferably, the method is deployed using a docker container, and the development language is Python.
Preferably, the method uses K-fold cross-validation to obtain the validation set.
Due to the application of the technical scheme, compared with the prior art, the invention has the following advantages:
the user loan risk prediction method applied to the intelligent bank is applied to an intelligent bank loan risk early warning platform, the platform firmly abandons the management concept of 'heavy loan light management', the post-loan management working efficiency is improved, key risk points are grasped, the risk identification foresight is improved, and the risk resolving capability is improved.
Drawings
The technical scheme of the invention is further explained by combining the accompanying drawings as follows:
fig. 1 is a flow chart illustrating the operation of the method for predicting the loan risk of a user applied to an intelligent bank according to the present invention.
Detailed Description
The invention is described below with reference to the accompanying drawings.
Fig. 1 shows a user loan risk prediction method applied to an intelligent bank according to the present invention, which includes the following steps, step 100, obtaining historical behavior data, online data and offline data of a bank user; step 101, preprocessing historical behavior data of a user, cleaning user attributes, and constructing basic features, time sequence features, business features, combination features, discrete features and other hundreds-dimensional features through engineering operation; 102, selecting features based on an xgboost model by adopting a method that a learning process and a feature selection process are performed simultaneously, and reserving the first n features; 103, establishing four models of Random Forest, GBDT, XGBOST and LightGBM, randomly dividing original data into 5 subsets which are mutually intersected and have the same size by each model, selecting one of the subsets as a verification set and taking the remaining 4 as a training set, and thus, each subset has an opportunity to be used as a training set; step 104, obtaining 4 models W1, W2, W3 and W4 after the step 103 is executed; 105, predicting the future overdue possibility of the user according to the historical behavior data of the user; and step 106, outputting the possibility of overdue payment of the user.
The method is deployed by using a docker container, and the development language is Python.
The method adopts K-fold cross validation to obtain a validation set.
The user loan risk prediction method applied to the intelligent bank is applied to an intelligent bank loan risk early warning platform, the platform firmly abandons the management concept of 'heavy loan light management', the post-loan management working efficiency is improved, key risk points are grasped, the risk identification foresight is improved, and the risk resolving capability is improved.
The above-mentioned embodiments are merely illustrative of the technical idea and features of the present invention, and the purpose thereof is to enable those skilled in the art to understand the contents of the present invention and implement the present invention, and not to limit the scope of the present invention, and all equivalent changes or modifications made according to the spirit of the present invention should be covered in the scope of the present invention.
Claims (3)
1. A user loan risk prediction method applied to a smart bank is characterized by comprising the following steps: the method comprises the following steps of step 100, acquiring historical behavior data, online data and offline data of a bank user; step 101, preprocessing historical behavior data of a user, cleaning user attributes, and constructing basic features, time sequence features, business features, combination features, discrete features and other hundreds-dimensional features through engineering operation; 102, selecting features based on an xgboost model by adopting a method that a learning process and a feature selection process are performed simultaneously, and reserving the first n features; 103, establishing four models of RandomForest, GBDT, XGBOST and LightGBM, randomly dividing original data into 5 subsets which are mutually intersected and have the same size, selecting one of the subsets as a verification set each time, and taking the remaining 4 as a training set, wherein each subset is organically used as the training set; step 104, obtaining 4 models W1, W2, W3 and W4 after the step 103 is executed; 105, predicting the future overdue possibility of the user according to the historical behavior data of the user; and step 106, outputting the possibility of overdue payment of the user.
2. The method as claimed in claim 1, wherein the user loan risk prediction method applied to the intelligent bank is as follows: the method is deployed by using a docker container, and the development language is Python.
3. The method as claimed in claim 1, wherein the user loan risk prediction method applied to the intelligent bank is as follows: the method adopts K-fold cross validation to obtain a validation set.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910832603.0A CN110751548A (en) | 2019-09-04 | 2019-09-04 | User loan risk prediction method applied to intelligent bank |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910832603.0A CN110751548A (en) | 2019-09-04 | 2019-09-04 | User loan risk prediction method applied to intelligent bank |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110751548A true CN110751548A (en) | 2020-02-04 |
Family
ID=69276159
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910832603.0A Withdrawn CN110751548A (en) | 2019-09-04 | 2019-09-04 | User loan risk prediction method applied to intelligent bank |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110751548A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112288334A (en) * | 2020-11-27 | 2021-01-29 | 上海评驾科技有限公司 | Lightgbm-based car networking risk factor extraction method |
| WO2021190300A1 (en) * | 2020-03-26 | 2021-09-30 | 肾泰网健康科技(南京)有限公司 | Method for constructing ai chronic kidney disease risk screening model, and chronic kidney disease risk screening method and system |
-
2019
- 2019-09-04 CN CN201910832603.0A patent/CN110751548A/en not_active Withdrawn
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021190300A1 (en) * | 2020-03-26 | 2021-09-30 | 肾泰网健康科技(南京)有限公司 | Method for constructing ai chronic kidney disease risk screening model, and chronic kidney disease risk screening method and system |
| CN112288334A (en) * | 2020-11-27 | 2021-01-29 | 上海评驾科技有限公司 | Lightgbm-based car networking risk factor extraction method |
| CN112288334B (en) * | 2020-11-27 | 2024-04-16 | 上海评驾科技有限公司 | Method for extracting Internet of vehicles risk factors based on lightgbm |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Jin et al. | A data-driven approach to predict default risk of loan for online peer-to-peer (P2P) lending | |
| CN112561598B (en) | Customer loss prediction and retrieval method and system based on customer portrayal | |
| Corney et al. | Putting the crowd to work in a knowledge-based factory | |
| Kobren et al. | Getting more for less: Optimized crowdsourcing with dynamic tasks and goals | |
| CN110223155A (en) | Method for pushing, device and the computer equipment of investment recommendation information | |
| Krishna et al. | Artificial Intelligence Application for Effective Customer Relationship Management | |
| Kunwar | Artificial intelligence in finance: Understanding how automation and machine learning is transforming the financial industry | |
| CN112784986A (en) | Feature interpretation method, device, equipment and medium for deep learning calculation result | |
| CN108288161A (en) | The method and system of prediction result are provided based on machine learning | |
| Moutinho et al. | Quantitative modelling in marketing and management | |
| CN111708883A (en) | Credit credit limit determination method and device based on machine learning and equipment fingerprint | |
| Akinsola et al. | Artificial intelligence emergence in disruptive technology | |
| CN110751548A (en) | User loan risk prediction method applied to intelligent bank | |
| Topalović et al. | Data mining applications in SMEs: an Italian perspective | |
| CN116645251A (en) | Panoramic knowledge training method, device, equipment and medium based on artificial intelligence | |
| Ilieva et al. | Modelling in support of decision making in business intelligence | |
| CN116701658A (en) | Business intention determining method and device and computer equipment | |
| Santos et al. | Corporate bankruptcy prediction using data mining techniques | |
| Raj et al. | The Significance of Big Data for the Base of the Pyramid Segment | |
| Furness | Techniques for customer modelling in CRM | |
| CN114298118B (en) | Data processing method based on deep learning, related equipment and storage medium | |
| Nanayakkara et al. | Adaptive Stock Market portfolio management and stock prices prediction platform for Colombo Stock Exchange of sri lanka | |
| Dudycz et al. | Application of ontology in financial assessment based on real options in small and medium-sized companies | |
| Permana | The Strategy for Developing a Marketplace Promotion Model Based on Artificial Intelligence (AI) to Improve Online Marketing in Indonesia | |
| Houshmand et al. | Presenting a rule based loan evaluation expert 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 | ||
| WW01 | Invention patent application withdrawn after publication | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20200204 |