CN116523628A - Credit model definition method based on public credit big data - Google Patents
Credit model definition method based on public credit big data Download PDFInfo
- Publication number
- CN116523628A CN116523628A CN202310527944.3A CN202310527944A CN116523628A CN 116523628 A CN116523628 A CN 116523628A CN 202310527944 A CN202310527944 A CN 202310527944A CN 116523628 A CN116523628 A CN 116523628A
- Authority
- CN
- China
- Prior art keywords
- data
- credit
- decision tree
- public
- method based
- 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 31
- 238000003066 decision tree Methods 0.000 claims abstract description 34
- 238000012549 training Methods 0.000 claims abstract description 14
- 238000004140 cleaning Methods 0.000 claims abstract description 12
- 238000012216 screening Methods 0.000 claims abstract description 5
- 238000004458 analytical method Methods 0.000 claims description 11
- 230000008569 process Effects 0.000 claims description 5
- 230000002159 abnormal effect Effects 0.000 claims description 4
- 238000011985 exploratory data analysis Methods 0.000 claims description 4
- 238000010219 correlation analysis Methods 0.000 claims description 3
- 238000012545 processing Methods 0.000 claims description 3
- 238000013139 quantization Methods 0.000 claims description 3
- 238000013515 script Methods 0.000 claims description 3
- 238000012163 sequencing technique Methods 0.000 claims description 3
- 238000012360 testing method Methods 0.000 claims description 3
- 238000010801 machine learning Methods 0.000 abstract description 4
- 238000011835 investigation Methods 0.000 abstract description 3
- 238000010586 diagram Methods 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 238000012217 deletion Methods 0.000 description 2
- 230000037430 deletion Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000012417 linear regression Methods 0.000 description 2
- 208000025174 PANDAS Diseases 0.000 description 1
- 208000021155 Paediatric autoimmune neuropsychiatric disorders associated with streptococcal infection Diseases 0.000 description 1
- 240000000220 Panda oleosa Species 0.000 description 1
- 235000016496 Panda oleosa Nutrition 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000000295 complement effect Effects 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000012502 risk assessment Methods 0.000 description 1
- 238000012800 visualization Methods 0.000 description 1
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
- G06F18/00—Pattern recognition
- G06F18/10—Pre-processing; Data cleansing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/211—Selection of the most significant subset of features
- G06F18/2113—Selection of the most significant subset of features by ranking or filtering the set of features, e.g. using a measure of variance or of feature cross-correlation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/24323—Tree-organised classifiers
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
- G06N20/20—Ensemble learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/01—Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound
-
- 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
- G06Q40/02—Banking, e.g. interest calculation or account maintenance
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Software Systems (AREA)
- Evolutionary Biology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Development Economics (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Technology Law (AREA)
- Computational Linguistics (AREA)
- Medical Informatics (AREA)
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
Abstract
The invention relates to a credit model definition method based on public credit big data, which comprises the specific steps of collecting and integrating user credit data to obtain financial credit related information; defining the type and data attribute of the credit data of the user, and cleaning the data to construct a corresponding credit data set; analyzing a credit data set through exploratory data, knowing the interrelationship among variables and the relation between the variables and predicted values, and screening factor data influencing the credit; constructing a model, sorting the importance of the data features, selecting important data features, and building a training data set decision tree; the invention collects user credit data through public big data, and better fits the nonlinear relation between the interpretation variable and the default risk through a machine learning method to predict the rating of the follow-up client credit, thereby achieving the purposes of reducing credit investigation cost and improving working efficiency.
Description
Technical Field
The invention relates to the technical field of credit risk assessment, in particular to a credit model definition method based on public credit big data, which is characterized in that individual data which look weakly associated are regenerated by splicing, restoring and associating fragmented data which are collected by banks or third parties and distributed at different positions.
Background
With the change of financial market environment and the development of credit derived products, commercial banks are exposed to more market risks. Where the credit risk is one of the major risks faced by financial institutions (particularly banks), most importantly, the credit risk has uncertainty, mainly referring to the risk of economic loss caused by the fact that the opponent of the trade fails to fulfill the obligation in the contract, i.e. the possibility that the trusted person cannot fulfill the responsibility of paying the present payment and the expected benefit of the trusted person deviates from the actual benefit. In order to solve the crisis problem faced by banks, research credit models are increasingly important to asset pricing and risk management work, the credit models are key technologies for accurately evaluating the credit level and risk level of objects, and the relative uniqueness of credit analysis model design is determined by the differences of industries where enterprises are located and the differences of customer groups. Enterprises need to develop and apply specialized credit analysis models. When enterprises develop credit analysis work, the requirements of accurately identifying and evaluating risks are often not met only by experience of management staff and a traditional method.
The models established at the present stage are a Z scoring model, a Crodit Risk+ model, a KMV model, a scoring card model and the like. Wherein, the Z scoring model lacks analysis on market data, so that the prediction result of the model has hysteresis and the credit change of an evaluation company can not be reflected in time; the Risk of default of the debtor of the Credit Risk + model is set randomly, unchanged during the calculation period, which does not correspond to the fact that the model use has certain drawbacks; the KMV model is suitable for evaluating credit risks of the marketing companies and has evaluation limitation; as a scoring card model of the traditional credit giving model, credit scoring can be carried out on a client according to various attribute and behavior data of the client, the dimension of reference is not perfect, and optimization through machine learning is needed.
Disclosure of Invention
Aiming at the problems, the invention provides a credit model definition method based on public credit big data, which optimizes a credit model from multidimensional attribute and business scene by machine learning, improves wind control effect, reflects credit risk degree of bank credit more truly and greatly reduces financial risk facing banks.
In order to achieve the above purpose, the following technical scheme is adopted: a credit model definition method based on public credit big data comprises the following specific steps,
s1, collecting and integrating credit data of users to obtain relevant information of financial credit;
s2, defining the type and the data attribute of the credit data of the user, and cleaning the data to construct a corresponding credit data set;
s3, analyzing a credit data set through exploratory data, knowing the interrelationship among variables and the relation between the variables and predicted values, and screening factor data influencing the credit;
s4, taking the data screened in the step S3 as a data set, constructing a model by using XGboost, sequencing the importance of the data features, selecting important data features, and establishing a training data set decision tree;
and S5, fitting a nonlinear relation between the interpretation variable and the default risk, and evaluating the performance of the model.
Further, the data cleaning in step S2 is not limited to quantization of the index, missing value processing, correlation analysis, data normalization.
Further, the data cleaning method comprises the following steps:
s200, selecting a data column in the credit data set to be analyzed, and hiding other data columns which do not participate in analysis;
s201, the same data column name appears in the credit data set, or two data column names with the same meaning, and the column name of one data column is selected for renaming;
s202, deleting repeated data values in the data, and only reserving a first piece of data of the repeated data;
s203, if the credit data set has data cells without data, inserting data values and complementing the data values;
s204, normalizing the public credit big data original data set into a data set with the mean value of 0 and the variance of 1, and splitting the data values in the inconsistent data columns if the data value standards of the data columns are inconsistent or naming rules are inconsistent in the original credit data set;
s205, analyzing the cleaned data, identifying abnormal values and deleting or interpolating.
Further, the exploratory data analysis procedure of step S3 includes:
s300, checking the whole situation of the credit data set, including checking the specific form of the data, checking the type and the data quantity of the data and checking the statistical indexes in the data, wherein the statistical indexes are not limited to the extreme value, the mean value and the variance of the data, and the quantity level of each index is ensured to be consistent in regression problems;
s301, data summarizing is carried out, data structures, data attributes and data recording modes of different data sources are checked, index indexes are selected according to the structural commonality of analysis, the data structures which are required to be modified are changed in batches by writing scripts, and data are combined;
s302, a decision tree preliminary model is established, training set data and testing set data are provided for the step S4, and setting of optimal factors is determined, wherein the optimal factors are factors influencing credit.
Further, the decision tree construction process of step S4 is as follows,
s400, training the factor data influencing the credit in the step S3;
s401, selecting data characteristics by adopting an information gain standard, and dividing a training data set into subsets according to the data characteristics;
s404, constructing a decision tree, selecting data features on each node of the decision tree corresponding to the information gain criteria, and recursively constructing the decision tree;
s405, classifying according to the data characteristic attribute of each node of the decision tree, and judging the credit score of the client according to the classification.
Further, the specific method of recursively constructing the decision tree of step S403 includes,
1) Calculating information gains of all possible data features from the root node;
2) Selecting the characteristic with the maximum information gain as the data characteristic of the node, and establishing the child node according to different values of the characteristic;
3) Recursively calling the method to the sub-nodes to construct a decision tree;
4) Until the information gain of all features is small or no features can be selected, a decision tree is obtained.
Further, in the step S5, the interpretation variable and the risk of default are both factor data affecting credit.
The invention has the beneficial effects that:
1. the invention collects user credit data through public big data, establishes a training data set decision tree through integrating, cleaning and screening factor data influencing credit, better fits the nonlinear relation between the interpretation variable and the default risk through a machine learning method, predicts the rating of the subsequent client credit, thereby achieving the purposes of reducing credit investigation cost and improving working efficiency.
2. The application of the credit model definition can penetrate through partial or all links of credit business before, during and after credit, so that the decision making efficiency is improved and the operation cost is reduced.
3. The credit model defined by the method can optimize the credit model according to the multidimensional attribute and business scene of different users, thereby improving the wind control effect, more truly reflecting the credit risk degree of the bank credit and greatly reducing the financial risk facing the bank.
Drawings
FIG. 1 is a schematic flow chart of the present invention;
FIG. 2 is a schematic diagram of a data cleaning process;
FIG. 3 is a schematic diagram of a exploratory data analysis flow;
FIG. 4 is a schematic diagram of a decision tree construction process.
Detailed Description
The technical solutions of the embodiments of the present disclosure will be clearly and completely described below with reference to the drawings in the embodiments of the present disclosure, and it is apparent that the described embodiments are some embodiments of the present disclosure, but not all embodiments. All other embodiments, which can be made by one of ordinary skill in the art without inventive effort, based on the embodiments of the present disclosure are intended to be within the scope of the embodiments of the present disclosure.
Examples
A credit model definition method based on public credit big data is characterized by comprising the following specific steps,
s1, collecting and integrating user credit data from a public data set to obtain financial credit related information;
s2, defining the type and the data attribute of the credit data of the user, and cleaning the data to construct a corresponding credit data set;
s3, analyzing a credit data set through exploratory data, knowing the interrelationship among variables and the relation between the variables and predicted values, and screening factor data influencing the credit;
s4, taking the data screened in the step S3 as a data set, constructing a model by using XGboost, sequencing the importance of the data features, selecting important data features, and establishing a training data set decision tree;
and S5, fitting a nonlinear relation between the interpretation variable and the default risk, and evaluating the performance of the model.
The step S2 defines the type of the user credit data including the user basic information, the user credit related information, and the user credit related information, the data attribute defining the user credit data includes the data attribute defining the type of the data, for example, the user basic information defines the basic information such as "user unique identification", "license number", "highest school", etc., the user credit related information defines the information such as "credit product type", "pre-credit amount", "basic rating", etc., the user credit information defines the information such as "dead credit card number", "unpaid person loan amount", etc., the specific definition is as shown in tables 1-3, the data cleaning includes quantization of index, deletion value processing, correlation analysis, data standardization, etc.
TABLE 1 user basic information
| Attributes of | |
| loanProduct | Loan product type |
| lmt | Amount of pre-credit |
| basicLevel | Basic rating |
| bankCard | Money release card number |
| residentAddr | Residence ground |
| linkRela | Contact relationship |
| setupHour | Application period |
| weekday | Date of filling |
TABLE 2 user loan related information
| Attributes of | |
| ncloseCreditCard | Number of disabled credit cards |
| unpayIndvLoan | Unpaid person loan amount |
| unpayOtherLoan | Not paying other loan amounts |
| unpayNormalLoan | Average amount of unpaid loan |
| 5yearBadloan | Pays no loan amount for five years |
TABLE 3 user credit investigation related information
The step S2 of data cleaning comprises the following steps:
s200, selecting a data column in the credit data set to be analyzed, and hiding other data columns which do not participate in analysis to avoid interference;
s201, the same data column name appears in the credit data set, or two data column names with the same meaning, and renaming is needed for the column name of a certain data column in order to avoid interference analysis results;
s202, deleting repeated data values in the data, and only reserving a first piece of data of the repeated data;
s203, data value deletion may occur in the original public credit big data, namely, data cells without data exist in the credit data set, the result is affected during data analysis, and a function is required to be manually inserted to complement the deleted data value;
s204, normalizing the public credit big data original data set into a data set with the mean value of 0 and the variance of 1, thereby eliminating errors caused by different quantity levels of various indexes, and splitting data values in inconsistent data columns if the data value standards of the data columns are inconsistent or naming rules are inconsistent in the original credit data set;
s205, analyzing the cleaned data, using a matplotlib module in Python to realize the visualization of the data, drawing a box plot by adopting a box plot function, identifying abnormal points in the box plot by using the quantile of the data, and directly deleting or interpolating the abnormal values. The exploratory data analysis flow and steps in step S3 are as follows:
s300, checking the whole situation of the credit data set, and checking the specific form of the data by using a head () method of pandas; checking the type and the data amount of the data with info (); the numerical control method comprises the following steps of normalizing variables, namely, data_trace (). Application (). Data_trace () and checking statistical indexes such as extremum, mean value, variance and the like of data, and ensuring that the quantity levels of all indexes are consistent in regression problems
data_train.describe()
data_train.info()；
S301, data summarizing is carried out, data structures, data attributes and data recording modes of different data sources are checked, index indexes are selected according to the structural commonality of analysis, the data structures which are required to be modified are changed in batches by writing scripts, and data are combined;
s303, establishing a decision tree preliminary model, providing training set and testing set data for the step S4, and determining the setting of optimal factors, wherein the optimal factors are factors influencing credit.
The process of constructing the decision tree in the step S4 is as follows:
s400, training the factor data influencing the credit in the step S3;
s401, for data feature selection, the feature selection criterion of the invention is information gain, and the training data set is divided into subsets according to the feature;
s402, constructing a decision tree by adopting an ID3 algorithm, selecting data features on each node of the decision tree corresponding to an information gain criterion, and recursively constructing the decision tree;
s403, classifying according to the data characteristic attribute of each node of the decision tree, and judging the credit score of the client according to the classification, wherein the credit score result is shown in the table 4.
Table 4 scoring credit results
The specific way to recursively construct the decision tree at step S403 is as follows,
1) Calculating information gains of all possible data features from root nodes;
2) Selecting the characteristic with the maximum information gain as the data characteristic of the node, and establishing the child node according to different values of the characteristic;
3) Recursively calling the method to the sub-nodes to construct a decision tree;
4) Until the information gain of all features is small or no features can be selected, a decision tree is obtained.
According to the characteristic attribute of the data of each node of the decision tree, the Scikit-Learn module of Python is used for multiple linear regression, and the 'M (user characteristic)' such as gender, age, academy and the like and the 'N (credit characteristic)' are used as variables of the linear regression, and two variable columns are connected into a single column by means of the np.c_command of the Numpy library:
X＝pd.DataFrame(np.c_[boston['M'],boston['N']],columns＝['M','N']Y＝boston['jieguo']
finally, a nonlinear relation between the interpretation variable and the default risk is realized, and the rating of the follow-up client credit is predicted.
The foregoing description is only illustrative of the preferred embodiments of the present invention, and is not intended to limit the scope of the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention are included in the protection scope of the present invention.
Claims (7)
1. A credit model definition method based on public credit big data is characterized by comprising the following specific steps,
s1, collecting and integrating credit data of users to obtain relevant information of financial credit;
s2, defining the type and the data attribute of the credit data of the user, and cleaning the data to construct a corresponding credit data set;
s3, analyzing a credit data set through exploratory data, knowing the interrelationship among variables and the relation between the variables and predicted values, and screening factor data influencing the credit;
s4, taking the data screened in the step S3 as a data set, constructing a model by using XGboost, sequencing the importance of the data features, selecting important data features, and establishing a training data set decision tree;
and S5, fitting a nonlinear relation between the interpretation variable and the default risk, and evaluating the performance of the model.
2. The credit model definition method based on public credit big data according to claim 1, wherein the data cleaning in the step 2 is not limited to quantization of indexes, missing value processing, correlation analysis, and data standardization.
3. The credit model definition method based on public credit big data according to claim 1 or 2, wherein said step S2 of data cleaning step comprises:
s200, selecting a data column in the credit data set to be analyzed, and hiding other data columns which do not participate in analysis;
s201, the same data column name appears in the credit data set, or two data column names with the same meaning, and the column name of one data column is selected for renaming;
s202, deleting repeated data values in the data, and only reserving a first piece of data of the repeated data;
s203, if the credit data set has data cells without data, inserting data values and complementing the data values;
s204, normalizing the public credit big data original data set into a data set with the mean value of 0 and the variance of 1, and splitting the data values in the inconsistent data columns if the data value standards of the data columns are inconsistent or naming rules are inconsistent in the original credit data set;
s205, analyzing the cleaned data, identifying abnormal values and deleting or interpolating.
4. The credit model definition method based on public credit big data according to claim 1, wherein the exploratory data analysis procedure of step S3 includes:
s300, checking the whole situation of the credit data set, including checking the specific form of the data, checking the type and the data quantity of the data and checking the statistical indexes in the data, wherein the statistical indexes are not limited to the extreme value, the mean value and the variance of the data, and the quantity level of each index is ensured to be consistent in regression problems;
s301, data summarizing is carried out, data structures, data attributes and data recording modes of different data sources are checked, index indexes are selected according to the structural commonality of analysis, the data structures which are required to be modified are changed in batches by writing scripts, and data are combined;
s302, a decision tree preliminary model is established, training set data and testing set data are provided for the step S4, and setting of optimal factors is determined, wherein the optimal factors are factors influencing credit.
5. The credit model definition method based on public credit big data according to claim 1, wherein the process of constructing the decision tree in step S4 is as follows:
s400, training the factor data influencing the credit in the step S3;
s401, selecting data characteristics by adopting an information gain standard, and dividing a training data set into subsets according to the data characteristics;
s402, constructing a decision tree, selecting data features on each node of the decision tree corresponding to the information gain criteria, and recursively constructing the decision tree;
s403, classifying according to the data characteristic attribute of each node of the decision tree, and judging the credit score of the client according to the classification.
6. The credit model definition method based on public credit big data according to claim 5, wherein the specific method of recursively constructing the decision tree in step S403 includes,
1) Calculating information gains of all possible data features from the root node;
2) Selecting the characteristic with the maximum information gain as the data characteristic of the node, and establishing the child node according to different values of the characteristic;
3) Recursively calling the method to the sub-nodes to construct a decision tree;
4) Until the information gain of all features is small or no features can be selected, a decision tree is obtained.
7. The credit model defining method based on public credit big data according to claim 1, wherein the interpretation variable and the risk of default in step S5 are both factor data affecting credit.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310527944.3A CN116523628A (en) | 2023-05-11 | 2023-05-11 | Credit model definition method based on public credit big data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310527944.3A CN116523628A (en) | 2023-05-11 | 2023-05-11 | Credit model definition method based on public credit big data |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116523628A true CN116523628A (en) | 2023-08-01 |
Family
ID=87395736
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310527944.3A Pending CN116523628A (en) | 2023-05-11 | 2023-05-11 | Credit model definition method based on public credit big data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116523628A (en) |
-
2023
- 2023-05-11 CN CN202310527944.3A patent/CN116523628A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Sarma | Predictive modeling with SAS enterprise miner: Practical solutions for business applications | |
| US7930242B2 (en) | Methods and systems for multi-credit reporting agency data modeling | |
| CN110956273A (en) | Credit scoring method and system integrating multiple machine learning models | |
| US20060085325A1 (en) | System, method, and computer program for assessing risk within a predefined market | |
| AU2014202660C1 (en) | A system and method using multi-dimensional rating to determine an entity's future commercial viability | |
| US20060100957A1 (en) | Electronic data processing system and method of using an electronic data processing system for automatically determining a risk indicator value | |
| CN111861698B (en) | Pre-loan approval early warning method and system based on loan multi-head data | |
| CN112700319A (en) | Enterprise credit line determination method and device based on government affair data | |
| CN112598500A (en) | Credit processing method and system for non-limit client | |
| CN110689437A (en) | Communication construction project financial risk prediction method based on random forest | |
| JP6251383B2 (en) | Calculating the probability of a defaulting company | |
| CN115660834A (en) | Individual loan risk assessment method based on decision tree | |
| CN116523628A (en) | Credit model definition method based on public credit big data | |
| WO1992017853A2 (en) | Direct data base analysis, forecasting and diagnosis method | |
| Andersson et al. | Bankruptcy determinants among Swedish SMEs:-The predictive power of financial measures | |
| Chen et al. | Construction of Bank Credit White List Access System Based on Grey Clustering Algorithm | |
| Huang | Recognition Algorithm of Financial Data Validity of Listed Enterprises Based on Classification Regression Decision Tree | |
| Ertuğrul | Customer Transaction Predictive Modeling via Machine Learning Algorithms | |
| CN118071482A (en) | Method for constructing retail credit risk prediction model and consumer credit business Scorebetad model | |
| CN118071483A (en) | Method for constructing retail credit risk prediction model and personal credit business Scorepsi model | |
| CN117994017A (en) | Method for constructing retail credit risk prediction model and online credit service Scoredelta model | |
| KR20020054362A (en) | System and method for valuing loan portfolios using fuzzy clustering | |
| CN115237970A (en) | Data prediction method, device, equipment, storage medium and program product | |
| KR20230094936A (en) | Activist alternative credit scoring system model using work behavior data and method for providing the same | |
| CN117764692A (en) | Method for predicting credit risk default probability |
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 |