CN115860926A - Wind control decision method and system based on decision tree - Google Patents
Wind control decision method and system based on decision tree Download PDFInfo
- Publication number
- CN115860926A CN115860926A CN202310133837.2A CN202310133837A CN115860926A CN 115860926 A CN115860926 A CN 115860926A CN 202310133837 A CN202310133837 A CN 202310133837A CN 115860926 A CN115860926 A CN 115860926A
- Authority
- CN
- China
- Prior art keywords
- data
- decision
- decision tree
- wind control
- variable
- 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
Images
Abstract
The invention provides a wind control decision method and a system based on a decision tree, wherein the method comprises the following steps: acquiring original data of a decision object, and preprocessing the original data; cleaning the preprocessed data, and deleting abnormal values in the data; performing feature extraction on the data to acquire variable information of the data; establishing a decision tree model according to the variable information, and optimizing the decision tree model; and displaying the risk evaluation result of the decision object through a visual decision tree. The invention has the beneficial effects that: the decision speed and the decision efficiency of the wind control are improved, the decision result is intuitive, the workload of wind control personnel is greatly reduced, and the working efficiency is improved.
Description
Technical Field
The invention relates to the technical field of internet finance, in particular to a wind control decision method and a wind control decision system based on a decision tree.
Background
The internet finance is a novel financial business mode for realizing fund financing, payment, investment and information intermediary service by using the internet technology and the information communication technology by traditional financial institutions and internet enterprises. The deep integration of the internet and finance is a trend, and will have more profound influence on the aspects of financial products, businesses, organizations, services and the like. The Internet finance plays a positive role in promoting the development and the expansion of small and micro enterprises, and the existing financial institutions are difficult to replace, thereby opening a door for the entrepreneurship and the innovation of the masses.
However, for risk assessment of a loan by an entity (e.g., person, institution), in various cases, a bank may determine whether to approve the borrower's loan by manually analyzing the borrower qualifications. In particular, loan risk may be determined based on analysis of certain behavior (e.g., debt tendering) and/or profile information (e.g., income, occupation) of an entity, based on which the professional requirements for risk assessment personnel are high and the efficiency of risk assessment is relatively low due to the long period of risk assessment resulting from the complexity of the assessment materials.
Disclosure of Invention
Based on this, the present invention provides a wind control decision method and system based on decision tree to solve the deficiencies in the prior art.
In a first aspect, the present application provides a wind control decision method based on a decision tree, including the following steps:
acquiring original data of a decision object, and preprocessing the original data;
cleaning the preprocessed data, and deleting abnormal values in the data;
extracting the characteristics of the data to obtain variable information of the data;
establishing a decision tree model according to the variable information, and optimizing the decision tree model;
and displaying the risk evaluation result of the decision object through a visual decision tree.
According to the wind control decision method based on the decision tree, the original data are preprocessed, the data are cleaned, necessary and valuable data are extracted, the data are extracted in a characteristic mode, the data are processed in a variable mode and can be identified and analyzed by the model, the model is optimized, the model is simplified, overfitting of the model is avoided, and the data after the model is analyzed are displayed through the visual decision tree. The method has the advantages that risk analysis and control of decision-making objects (loan intention customers) can be efficiently realized, compared with a mode that professionals manually carry out risk control, the method is high in decision-making speed and decision-making efficiency, the decision-making result is visualized, the workload of wind control personnel is greatly reduced, and the working efficiency is improved.
The wind control decision method based on the decision tree further has the following additional technical characteristics:
preferably, in the decision tree-based wind control decision method according to the present application, the step of preprocessing the raw data specifically includes: redundant variable processing, missing value processing, local interpolation variable processing, and outlier processing.
Preferably, in the decision tree-based wind control decision method according to the present application, the step of processing the redundant variables includes:
and deleting variables which do not influence the decision tree model in the original data.
Preferably, in the wind control decision method based on the decision tree according to the present application, the missing value processing step includes:
judging whether the original data has missing values or not;
if yes, inquiring whether the original data has complete data in the last 3 years;
if the original data has complete data in the previous 3 years, averaging and filling according to the data in the previous 3 years;
and if the original data does not have complete data in the last 3 years, performing 0 filling on the missing value.
Preferably, in the decision tree-based wind control decision method according to the present application, the step of extracting the features of the data and obtaining the variable information of the data specifically includes: deletion rate screening, IV value screening and correlation screening.
Preferably, in the decision tree-based wind control decision method according to the present application, the step of screening the deficiency rate specifically includes:
calculating the deletion rate of all variables in the following way:
miss rate = number of variable values missed/total number of variable values;
and if the deletion rate of any variable is greater than a preset value, deleting the corresponding variable.
Preferably, in the decision tree-based wind control decision method according to the present application, the step of screening the IV value specifically includes:
calculating IV values of all variables;
and screening and retaining variables with IV values larger than a preset value.
Preferably, in the decision tree-based wind control decision method according to the present application, the step of relevance screening specifically includes:
performing relevance evaluation on all variables by using a Pearson correlation coefficient;
and deleting the variable with the correlation larger than the preset value.
Preferably, the step of establishing a decision tree model according to the variable information and tuning the decision tree model specifically includes:
adjusting parameters to the limit that the decision tree model is extremely tested;
performing parameter tuning on the decision tree model, wherein parameters related to the tuning comprise the minimum sample number of leaf nodes, the maximum depth and random seeds of tree branches, the minimum weight sum of the leaf nodes, the minimum purity increment of the node branches and the maximum leaf node number;
and pruning the decision tree model.
In a second aspect, the present application provides a decision tree-based wind control decision system, including:
a data acquisition module: the system comprises a data acquisition module, a decision-making module and a data processing module, wherein the data acquisition module is used for acquiring raw data of a decision-making object;
a preprocessing module: the system is used for preprocessing the original data;
a cleaning module: the system is used for cleaning the preprocessed data and deleting abnormal values in the data;
a feature extraction module: the system is used for extracting the characteristics of the data and acquiring the variable information of the data;
a modeling module: the decision tree model is established according to the variable information;
an optimization module: for tuning the decision tree model;
a visualization module: and the risk assessment result of the decision object is displayed through a visual decision tree.
Additional features and advantages of the present application will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by the practice of the present application. The objectives and other advantages of the application may be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.
Drawings
Fig. 1 is a flowchart of a decision tree-based wind control decision method according to an embodiment of the present invention;
fig. 2 is a flowchart of a method for missing value processing of original data in a decision tree-based wind control decision method according to an embodiment of the present invention;
fig. 3 is a flowchart of a method for missing rate screening in a decision tree-based wind control decision method according to an embodiment of the present invention;
fig. 4 is a flowchart of a method for screening an IV value in a decision tree-based wind control decision method according to an embodiment of the present invention;
fig. 5 is a flowchart of a method for relevance screening in a decision tree-based wind control decision method according to an embodiment of the present invention;
fig. 6 is a flowchart of a method for tuning a decision tree model in a decision tree-based wind control decision method according to an embodiment of the present invention;
fig. 7 is a schematic structural diagram of a decision tree-based wind control decision system according to a second embodiment of the present invention.
The following detailed description further illustrates the invention in conjunction with the above-described figures.
Detailed Description
To facilitate an understanding of the invention, the invention will now be described more fully with reference to the accompanying drawings. Several embodiments of the invention are presented in the drawings. This invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete.
It will be understood that when an element is referred to as being "secured to" another element, it can be directly on the other element or intervening elements may also be present. When an element is referred to as being "connected" to another element, it can be directly connected to the other element or intervening elements may also be present. The terms "vertical," "horizontal," "left," "right," and the like as used herein are for illustrative purposes only.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. The terminology used in the description of the invention herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used herein, the term "and/or" includes any and all combinations of one or more of the associated listed items.
In current risk assessment for physical (e.g., personal, institutional) loans, in various cases, banks determine whether to approve the borrower's loan by manually analyzing the borrower qualifications. In particular, loan risk may be determined based on analysis of certain behavioral (e.g., tendering debts) and/or profile information (e.g., income, occupation) of the entity, based on the condition that the professional requirements for the risk assessment personnel are high, and the risk assessment period is long due to the complexity of the assessment materials, and the efficiency of the risk assessment is relatively low.
Therefore, the invention provides a wind control decision method based on a decision tree to overcome the problems in the prior art.
Referring to fig. 1, a decision tree-based wind control decision method according to a first embodiment of the present invention is shown. Specifically, the wind control decision method based on the decision tree comprises the following steps:
and S11, acquiring original data of the decision object, and preprocessing the original data.
In the embodiment of the invention, the original data acquisition technology is simply to utilize a data analysis technology, construct a model based on mass data resources, analyze hidden and high-value information in the model, and provide certain decision support for a decision maker. There are many methods of data collection that are commonly used to retrieve and crawl data from a database. By way of example and not limitation. The embodiment of the invention can extract the data according to the customer data stored by the bank. This raw data is a very good data source; when the basic plane of a company needs to be analyzed, the required enterprise information can be captured through the crawler technology.
And S12, cleaning the preprocessed data, and deleting abnormal values in the data.
It should be noted that, if the original data is directly used for modeling, some problems may occur, for example, variables in the data contain a large number of missing values, which may seriously affect the final result; the variables also contain a large number of repeated values, which are of little interest for modeling; there are also data that are not numeric and require characterization. By deleting these meaningless data, the data source employed for subsequent modeling is made better.
And S13, performing feature extraction on the data to acquire variable information of the data.
For example, but not by way of limitation, since part of data is not variable information during modeling and cannot be directly analyzed by the model, for example, when the influence of comments on the internet on the stock profitability is analyzed, the languages of the comments cannot be directly used, keywords need to be extracted, certain values (variable information) are given to the keywords, and the texts are digitized to enable later analysis. Therefore, in the present application, by performing feature extraction on data, enumerated variables not supported by a model are processed into numerical variables that can be processed by model analysis.
And S14, establishing a decision tree model according to the variable information, and optimizing the decision tree model.
The model parameters are adjusted to achieve the purposes of avoiding model overfitting and model simplification as much as possible when the model analyzes data variables.
And S15, displaying the risk evaluation result of the decision object through a visual decision tree.
To sum up, according to the wind control decision method based on the decision tree provided by the embodiment of the invention, the original data is preprocessed, the data is cleaned, necessary and valuable data is extracted, the data is subjected to characteristic extraction, the data processing is in a variable form and can be identified and analyzed by the model, the model is optimized, the model is simplified, overfitting of the model is avoided, and the data after the model analysis is displayed through the visual decision tree. The method has the advantages that risk analysis and control of decision-making objects (loan intention customers) can be efficiently realized, compared with a mode that professionals manually carry out risk control, the method is high in decision-making speed and decision-making efficiency, the decision-making result is visualized, the workload of wind control personnel is reduced, and the working efficiency is improved.
In the embodiment of the present invention, the preprocessing step of the raw data specifically includes: redundant variable processing, missing value processing, local interpolation variable processing, and outlier processing. By adopting the method, the original data can be effectively optimized.
Further, in the wind control decision method based on the decision tree provided in the embodiment of the present invention, the method for processing the redundancy variable of the original data specifically includes:
and deleting variables which do not influence the decision tree model in the original data.
It is understood that the redundant variables mainly refer to variables which have no practical meaning and do not have any influence on the model result according to the credit business common knowledge, so that the deletion processing is directly carried out before the model is built.
It should be further noted that, when the system of the bank and the loan institution operates and stores data, it is also a relatively normal situation that some variable values have missing values, mainly because the data is stored by an operation error or the system source data is incomplete, such as the information retained by the customer is not perfect. The existence of missing values has a great influence on the modeling analysis, so the missing values are processed correspondingly. In view of the above, please refer to fig. 2, which is a flowchart of a method for processing missing values of original data in a decision tree-based wind control decision method according to an embodiment of the present invention, specifically, the method includes:
and S21, judging whether the original data has missing values or not.
And S22, if yes, inquiring whether the original data has complete data in the last 3 years.
And S23, if the original data has complete data in the previous 3 years, averaging and filling according to the data in the previous 3 years.
And S24, if the original data does not have complete data in the last 3 years, performing 0 filling on the missing value.
By the missing value processing method, the situation that the original data of the system is incomplete to cause modeling analysis deviation is greatly reduced.
Further, in the embodiment of the present invention, the step of performing feature extraction on the data to obtain the variable information of the data specifically includes: deletion rate screening, IV value screening and correlation screening. By the data screening method, the analysis value of the data is further improved.
Referring to fig. 3, a flow chart of a method for missing rate screening in a decision tree-based wind control decision method according to an embodiment of the present invention is shown, and specifically, the method includes:
step S31, calculating the missing rate of all variables, wherein the missing rate is calculated in the following mode: miss rate = number of misses at variable value/total number of variable values.
And S32, if the loss rate of any variable is greater than a preset value, deleting the corresponding variable.
It can be understood that when the data volume for mining analysis is small and the proportion of the missing value in all values of the variable is also large, when the variable is used for modeling analysis, the variable has a disadvantage of being far more beneficial to achieving the effect of model analysis. At the same time, however, the possibility of a variable generating a large amount of data missing represents some special meaning, and the possibility of directly deleting the variable without analyzing the variable at all may also lose some relatively important information. By way of example and not limitation, in the embodiment of the present invention, the default value for determining the loss rate is set to 10%. The situation is effectively overcome by deleting the partial data with larger deletion rate in the variable.
Referring to fig. 4, a flowchart of a method for screening an IV value in a decision tree-based wind control decision method according to an embodiment of the present invention is shown, and specifically, the method includes:
and step S41, calculating IV values of all variables.
And S42, screening and reserving the variable with the IV value larger than the preset value.
In a financial scenario, the IV value is a measure that is often used to measure the strength of a feature. If the IV value is larger, the discrimination of the characteristic is stronger, and if the IV value is smaller, the discrimination of the characteristic is weaker. In the embodiment of the present invention, the IV values of all variables are calculated and compared with the set threshold to determine, so as to screen out the variables whose prediction capabilities meet the requirements, for example and without limitation, the preset value (determination threshold) of the IV value in the embodiment of the present invention is set to 0.02.
Referring to fig. 5, a flowchart of a method for relevance screening in a decision tree-based wind control decision method according to an embodiment of the present invention is shown, and specifically, the method includes:
and S51, performing relevance evaluation on all variables by using the Pearson correlation coefficient.
And S52, deleting the variables with the correlation larger than the preset value.
In summary, all the variables are subjected to correlation comparison through the pearson correlation coefficient, and the variables with the correlation larger than the preset value (in the embodiment of the present invention, the preset value is set to be 0.7) are deleted. And screening and removing part of the repeatability variable information. So as to improve the accuracy of the decision tree model analysis.
Referring to fig. 6, a flowchart of a method for tuning a decision tree model in a decision tree-based wind control decision method according to an embodiment of the present invention is shown, and specifically, the method includes:
and S61, adjusting the parameters to the limit of the test decision tree model.
And S62, performing parameter optimization on the decision tree model.
The parameters related to tuning comprise the minimum sample number of leaf nodes, the maximum depth and random seeds of tree branches, the minimum weight sum of the leaf nodes, the minimum purity increment of the node branches and the maximum leaf node number.
And S63, pruning the decision tree model.
In conclusion, parameter adjustment is performed during modeling of the decision tree, so that the phenomenon of overfitting of the model can be effectively avoided, and pruning is performed on the model. The model is effectively simplified, the operation intensity of the model is reduced, and the operation efficiency is improved. In the embodiment of the present invention, the pruning processing is specifically implemented by configuring the minimum number of samples required at the leaf nodes or the maximum depth of the tree.
In conclusion, a technical means of decision tree for pre-credit analysis is adopted. The beneficial effects are as follows: the decision tree is easy to understand and annotate, can be visually analyzed, and is easy to extract rules; both categorical and numerical data can be handled simultaneously; there are advantages in processing data with missing attributes; irrelevant characteristics can be well processed; when the data set is tested, the running speed is higher; and the processing speed is higher when large data samples are processed. And combines the technical means of the invention for the raw data processing. The decision speed and the decision efficiency are greatly improved, the decision result is intuitive, the workload of wind control personnel is greatly reduced, and the working efficiency is improved.
Referring to fig. 7, a decision tree-based wind control decision system according to a second embodiment of the present invention includes:
the data acquisition module 71: for obtaining raw data of the decision object.
The preprocessing module 72: for preprocessing the raw data.
The cleaning module 73: the method is used for cleaning the preprocessed data and deleting abnormal values in the data.
The feature extraction module 74: and the data processing device is used for extracting the characteristics of the data and acquiring the variable information of the data.
The modeling module 75: and the decision tree model is established according to the variable information.
The optimization module 76: for tuning the decision tree model.
The visualization module 77: and the risk assessment result of the decision object is displayed through a visual decision tree.
To sum up, the decision tree-based wind control decision system provided by the embodiment of the invention is used for implementing the decision tree-based wind control decision method, and the data is preprocessed, cleaned, extracted to obtain necessary and valuable data, and then subjected to characteristic extraction, so that the data processing is in a variable form, can be identified and analyzed by a model, and then the model is optimized, so that the model is simplified, overfitting of the model is avoided, and the data after the model analysis is displayed through a visual decision tree. The method has the advantages that risk analysis and control of decision-making objects (loan intention customers) can be efficiently realized, compared with a mode that professionals manually carry out risk control, the method is high in decision-making speed and decision-making efficiency, the decision-making result is visualized, the workload of wind control personnel is greatly reduced, and the working efficiency is improved.
In addition, the wind control decision method based on the decision tree of the embodiment of the application described in conjunction with the drawings can be realized by computer equipment. The computer device may include a processor and a memory storing computer program instructions.
The computer device may execute the decision tree-based wind control decision method in the embodiment of the present application based on the acquired data information, thereby implementing the decision tree-based wind control decision method described in conjunction with fig. 1.
In addition, in combination with the decision tree-based wind control decision method in the foregoing embodiments, embodiments of the present application may provide a computer-readable storage medium to implement. The computer readable storage medium having stored thereon computer program instructions; the computer program instructions, when executed by a processor, implement any of the decision tree based wind control decision methods of the above embodiments.
In the description herein, references to the description of the term "one embodiment," "some embodiments," "an example," "a specific example," or "some examples," etc., mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, the schematic representations of the terms used above do not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
The above-mentioned embodiments only express several embodiments of the present invention, and the description thereof is more specific and detailed, but not construed as limiting the scope of the present invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the inventive concept, which falls within the scope of the present invention. Therefore, the protection scope of the present patent shall be subject to the appended claims.
Claims (10)
1. A wind control decision method based on a decision tree is characterized by comprising the following steps:
acquiring original data of a decision object, and preprocessing the original data;
cleaning the preprocessed data, and deleting abnormal values in the data;
extracting the characteristics of the data to obtain variable information of the data;
establishing a decision tree model according to the variable information, and optimizing the decision tree model;
and displaying the risk evaluation result of the decision object through a visual decision tree.
2. The decision tree-based wind control decision method according to claim 1, wherein the preprocessing step of the raw data specifically comprises: redundant variable processing, missing value processing, local interpolation variable processing, and outlier processing.
3. The decision tree based wind control decision method of claim 2, wherein the step of redundant variable processing comprises:
and deleting variables which do not influence the decision tree model in the original data.
4. The decision tree based wind control decision method of claim 2, wherein the missing value processing step comprises:
judging whether the original data has missing values or not;
if yes, inquiring whether the original data has complete data in the last 3 years;
if the original data has complete data in the previous 3 years, calculating the average value according to the data in the previous 3 years for filling;
and if the original data does not have complete data in the last 3 years, performing 0 filling on the missing value.
5. The decision tree-based wind control decision method according to claim 1, wherein the step of extracting the features of the data and obtaining the variable information of the data specifically comprises: deletion rate screening, IV value screening and correlation screening.
6. The decision tree-based wind control decision method according to claim 5, wherein the step of missing rate screening specifically comprises:
calculating the deletion rate of all variables in the following way:
miss rate = number of variable values missed/total number of variable values;
and if the deletion rate of any variable is greater than a preset value, deleting the corresponding variable.
7. The decision tree based wind control decision method according to claim 5, wherein the step of screening the IV value specifically comprises:
calculating IV values of all variables;
and screening and reserving the variable with the IV value larger than the preset value.
8. The decision tree based wind control decision method according to claim 5, wherein the step of relevance screening specifically comprises:
performing relevance evaluation on all variables by using a Pearson correlation coefficient;
and deleting the variable with the correlation larger than the preset value.
9. The decision tree-based wind control decision method according to claim 1, wherein the step of building a decision tree model according to the variable information and tuning the decision tree model specifically comprises:
adjusting parameters to the limit that the decision tree model is extremely tested;
performing parameter tuning on the decision tree model, wherein parameters related to tuning comprise the minimum sample number of leaf nodes, the maximum depth and random seeds of tree branches, the minimum weight sum of the leaf nodes, the minimum purity increment of the node branches and the maximum leaf node number;
and pruning the decision tree model.
10. A decision tree based wind control decision system, the system comprising:
a data acquisition module: the system comprises a data acquisition module, a decision-making module and a data processing module, wherein the data acquisition module is used for acquiring raw data of a decision-making object;
a preprocessing module: the system is used for preprocessing the original data;
a cleaning module: the system is used for cleaning the preprocessed data and deleting abnormal values in the data;
a feature extraction module: the system is used for extracting the characteristics of the data and acquiring the variable information of the data;
a modeling module: the decision tree model is established according to the variable information;
an optimization module: for tuning the decision tree model;
a visualization module: and the risk assessment result of the decision object is displayed through a visual decision tree.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310133837.2A CN115860926A (en) | 2023-02-20 | 2023-02-20 | Wind control decision method and system based on decision tree |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310133837.2A CN115860926A (en) | 2023-02-20 | 2023-02-20 | Wind control decision method and system based on decision tree |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115860926A true CN115860926A (en) | 2023-03-28 |
Family
ID=85658370
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310133837.2A Pending CN115860926A (en) | 2023-02-20 | 2023-02-20 | Wind control decision method and system based on decision tree |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115860926A (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111951097A (en) * | 2020-08-12 | 2020-11-17 | 深圳微众信用科技股份有限公司 | Enterprise credit risk assessment method, device, equipment and storage medium |
| CN113064883A (en) * | 2020-09-28 | 2021-07-02 | 开鑫金服(南京)信息服务有限公司 | Method for constructing logistics wind control model, computer equipment and storage medium |
| CN113705904A (en) * | 2021-08-31 | 2021-11-26 | 国网上海市电力公司 | Chemical plant area power utilization fault prediction method based on random forest algorithm |
| CN114155880A (en) * | 2021-12-06 | 2022-03-08 | 上海欣方智能系统有限公司 | Illegal voice recognition method and system based on GBDT algorithm model |
| CN114638688A (en) * | 2022-03-21 | 2022-06-17 | 江苏城乡建设职业学院 | Interception strategy derivation method and system for credit anti-fraud |
| CN115660834A (en) * | 2022-12-23 | 2023-01-31 | 河北雄安舜耕数据科技有限公司 | Individual loan risk assessment method based on decision tree |
-
2023
- 2023-02-20 CN CN202310133837.2A patent/CN115860926A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111951097A (en) * | 2020-08-12 | 2020-11-17 | 深圳微众信用科技股份有限公司 | Enterprise credit risk assessment method, device, equipment and storage medium |
| CN113064883A (en) * | 2020-09-28 | 2021-07-02 | 开鑫金服(南京)信息服务有限公司 | Method for constructing logistics wind control model, computer equipment and storage medium |
| CN113705904A (en) * | 2021-08-31 | 2021-11-26 | 国网上海市电力公司 | Chemical plant area power utilization fault prediction method based on random forest algorithm |
| CN114155880A (en) * | 2021-12-06 | 2022-03-08 | 上海欣方智能系统有限公司 | Illegal voice recognition method and system based on GBDT algorithm model |
| CN114638688A (en) * | 2022-03-21 | 2022-06-17 | 江苏城乡建设职业学院 | Interception strategy derivation method and system for credit anti-fraud |
| CN115660834A (en) * | 2022-12-23 | 2023-01-31 | 河北雄安舜耕数据科技有限公司 | Individual loan risk assessment method based on decision tree |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10083263B2 (en) | Automatic modeling farmer | |
| US20110022513A1 (en) | Intelligent collections models | |
| US20110246385A1 (en) | Automatically recalibrating risk models | |
| Holmes et al. | Financial market impact on the real economy: An assessment of asymmetries and volatility linkages between the stock market and unemployment rate | |
| CN111738331A (en) | User classification method and device, computer-readable storage medium and electronic device | |
| CN114048436A (en) | Construction method and construction device for forecasting enterprise financial data model | |
| CN111369344A (en) | Method and device for dynamically generating early warning rule | |
| CN112700321A (en) | Multi-rule anti-fraud prediction method and system based on user behavior data | |
| CN110689437A (en) | Communication construction project financial risk prediction method based on random forest | |
| JP2000259719A (en) | Method and device for calculating probability of default on obligation | |
| Tsagem et al. | Cash conversion cycle and profitability of Nigerian small and medium-sized entities: an empirical analysis | |
| CN108510393A (en) | Intelligence based on big data, which is thrown, cares for system and method | |
| CN116596674A (en) | External trade risk assessment method based on big data analysis | |
| CN117132383A (en) | Credit data processing method, device, equipment and readable storage medium | |
| CN115860926A (en) | Wind control decision method and system based on decision tree | |
| Saragih et al. | Bankruptcy prediction by using the Grover method | |
| Nguyen | The credit risk evaluation models: An application of data mining techniques | |
| CN113869973A (en) | Product recommendation method, product recommendation system, and computer-readable storage medium | |
| CN113781210A (en) | Automatic characteristic engineering method and device based on customer financial transaction data structure | |
| Zang | Construction of Mobile Internet Financial Risk Cautioning Framework Based on BP Neural Network | |
| Warganegara et al. | Tobit Regression Analysis on Factors Influencing Dividend Policy of Indonesian Manufacturing Firms. | |
| KR20220074327A (en) | Loan regular auditing system using artificia intellicence | |
| Haoru et al. | Risk Assessment of Internet Credit Based on Big Data Analysis | |
| CN111951099A (en) | Credit card issuing model and application method thereof | |
| Parnes | Approximating default probabilities with soft information |
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 |