CN112734568B - Credit scoring card model construction method, device, equipment and readable storage medium - Google Patents
Credit scoring card model construction method, device, equipment and readable storage medium Download PDFInfo
- Publication number
- CN112734568B CN112734568B CN202110139379.4A CN202110139379A CN112734568B CN 112734568 B CN112734568 B CN 112734568B CN 202110139379 A CN202110139379 A CN 202110139379A CN 112734568 B CN112734568 B CN 112734568B
- Authority
- CN
- China
- Prior art keywords
- index
- parameter combination
- evaluation
- evaluation index
- determining
- 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.)
- Active
Links
- 238000010276 construction Methods 0.000 title claims abstract description 26
- 238000011156 evaluation Methods 0.000 claims abstract description 315
- 238000010801 machine learning Methods 0.000 claims abstract description 129
- 238000012549 training Methods 0.000 claims abstract description 83
- 238000012795 verification Methods 0.000 claims abstract description 31
- 230000006399 behavior Effects 0.000 claims abstract description 30
- 238000000034 method Methods 0.000 claims description 22
- 238000005516 engineering process Methods 0.000 abstract description 10
- 238000012216 screening Methods 0.000 abstract description 9
- 230000000694 effects Effects 0.000 description 11
- 238000004891 communication Methods 0.000 description 6
- 238000004422 calculation algorithm Methods 0.000 description 4
- 230000001133 acceleration Effects 0.000 description 2
- 230000006835 compression Effects 0.000 description 2
- 238000007906 compression Methods 0.000 description 2
- 230000001186 cumulative effect Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 230000005484 gravity Effects 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 238000007637 random forest analysis 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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Software Systems (AREA)
- General Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Data Mining & Analysis (AREA)
- Strategic Management (AREA)
- Marketing (AREA)
- Economics (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Technology Law (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Development Economics (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
Abstract
The invention discloses a credit score card model construction method, a credit score card model construction device, credit score card model construction equipment and a readable storage medium, wherein the credit score card model construction method comprises the following steps: determining a training sample and a verification sample based on credit behavior data of the borrower; inputting a training sample into a machine learning model, and training the machine learning model based on parameter combinations to obtain a first evaluation index; inputting the verification sample into the machine learning model after training, and determining a second evaluation index; based on the first and second evaluation indexes, an optimal parameter combination of the parameter combinations is determined to determine a credit score card model corresponding to the machine learning model based on the optimal parameter combination. According to the invention, the model evaluation indexes are automatically evaluated through the index evaluation system, so that the manual screening of the model evaluation indexes is avoided, the technical problem of low modeling efficiency of the conventional credit score card model modeling technology caused by the mode of manually screening the model evaluation indexes is solved, and the modeling efficiency of the credit score card model is improved.
Description
Technical Field
The present invention relates to the technical field of financial science (Fintech), and in particular, to a credit score card model construction method, apparatus, device and readable storage medium.
Background
With the development of computer technology, more and more technologies (big data, distributed, blockchain, artificial intelligence, etc.) are applied in the financial field, and the traditional financial industry is gradually changing to the financial technology (Fintech), but due to the requirements of security and real-time performance of the financial industry, higher requirements are also put forward on the technologies.
Credit scoring card model: the grading card is a decision tool for the loan organization to analyze and mine the credit behavior data of the borrower, predict the default probability of the borrower in a certain period, and obtain the credit scores of different grades so as to conduct risk management and control. Machine learning algorithm: refers to an emerging modeling method of neural networks, random forests, GBDT and the like. In the process of applying the machine learning algorithm to the credit scoring model, because the machine learning algorithm has the fitting risk, the evaluation index is inconsistent with the indexes such as distinguishing force, stability and the like of the scoring card model, so that the machine learning algorithm is applied to the credit scoring card model, the evaluation indexes of a plurality of scoring cards are required to be additionally calculated, and then the optimal model evaluation indexes are manually screened. However, the mode of manually screening the model evaluation index causes the existing credit score card model modeling technology to have the technical problem of low modeling efficiency.
The foregoing is provided merely for the purpose of facilitating understanding of the technical solutions of the present invention and is not intended to represent an admission that the foregoing is prior art.
Disclosure of Invention
The invention mainly aims to provide a credit score card model construction method, a credit score card model construction device, credit score card model construction equipment and a credit score card model construction readable storage medium, and aims to solve the technical problem that modeling efficiency of an existing credit score card model modeling technology is low.
In order to achieve the above object, the present invention provides a credit score card model construction method, which includes the steps of:
acquiring credit behavior data of a borrower, and determining a training sample and a verification sample based on the credit behavior data;
inputting the training sample into a preset machine learning model, and training the machine learning model based on parameter combinations to obtain a first evaluation index corresponding to the parameter combinations, wherein one parameter combination correspondingly trains one machine learning model based on the same training sample;
after the machine learning model is trained, inputting the verification sample into the trained machine learning model, and determining a second evaluation index corresponding to the parameter combination;
and determining an optimal parameter combination in the parameter combinations based on the first evaluation index and the second evaluation index corresponding to the parameter combinations, so as to determine a credit score card model corresponding to the machine learning model based on the optimal parameter combination.
Optionally, the first evaluation index includes a first KS index output by the machine learning model based on the training sample, and the second evaluation index includes a second KS index and a PSI index output by the machine learning model based on the verification sample;
the step of determining the optimal parameter combination in the parameter combination based on the first evaluation index and the second evaluation index corresponding to the parameter combination includes:
determining a first target parameter combination corresponding to the first KS index and the second KS index when the first KS index and the second KS index meet a first preset condition based on the first KS index and the second KS index corresponding to the parameter combination;
determining a second target parameter combination corresponding to the PSI index corresponding to the first target parameter combination when the PSI index corresponding to the first target parameter combination meets a second preset condition based on the PSI index corresponding to the first target parameter combination;
and when a second target parameter combination corresponding to the PSI index corresponding to the second target parameter combination meets a second preset condition exists, the second target parameter combination is used as the optimal parameter combination in the parameter combinations.
Optionally, the step of determining, based on the first KS indicator and the second KS indicator corresponding to the parameter combination, a first target parameter combination corresponding to when the first KS indicator and the second KS indicator meet a first preset condition includes:
Determining that the parameter combination corresponding to the second KS index larger than a first preset threshold value is a third target parameter combination based on the second KS index corresponding to the parameter combination;
and determining a third target parameter combination corresponding to the third target parameter combination as a first target parameter combination when the difference value between the first KS index and the second KS index corresponding to the third target parameter combination is smaller than a second preset threshold value based on the first KS index and the second KS index corresponding to the third target parameter combination.
Optionally, the step of determining, based on the PSI index corresponding to the first target parameter combination, a second target parameter combination corresponding to the PSI index corresponding to the first target parameter combination when the PSI index meets a second preset condition includes:
and based on the PSI index corresponding to the first target parameter combination, taking the first target parameter combination corresponding to the smallest PSI index in the PSI indexes corresponding to the first target parameter combination as a second target parameter combination.
Optionally, the step of determining the best parameter combination in the parameter combinations based on the first evaluation index and the second evaluation index corresponding to the parameter combinations includes:
determining whether the first evaluation index and the second evaluation index corresponding to the parameter combination meet a third preset condition corresponding to the index evaluation system or not based on a preset index evaluation system;
And if the first evaluation index and the second evaluation index corresponding to the parameter combination meet a third preset condition corresponding to the index evaluation system, determining the optimal parameter combination in the parameter combination based on the first evaluation index and the second evaluation index corresponding to the parameter combination.
Optionally, the step of determining whether the first evaluation index and the second evaluation index corresponding to the parameter combination meet a third preset condition corresponding to the index evaluation system based on a preset index evaluation system includes:
determining whether the first evaluation index corresponding to the parameter combination is in a first index range corresponding to the index evaluation system or not based on a preset index evaluation system, and determining whether the second evaluation index corresponding to the parameter combination is in a second index range corresponding to the index evaluation system or not;
if the first evaluation index and the second evaluation index corresponding to the parameter combination meet a third preset condition corresponding to the index evaluation system, determining the optimal parameter combination in the parameter combination based on the first evaluation index and the second evaluation index corresponding to the parameter combination comprises:
And if the first evaluation index corresponding to the parameter combination is in a first index range corresponding to the index evaluation system and the second evaluation index corresponding to the parameter combination is in a second index range corresponding to the index evaluation system, determining the optimal parameter combination in the parameter combination based on the first evaluation index and the second evaluation index corresponding to the parameter combination.
Optionally, after the step of determining whether the first evaluation index and the second evaluation index corresponding to the parameter combination meet a third preset condition corresponding to the index evaluation system based on a preset index evaluation system, the method further includes:
and if the first evaluation index and the second evaluation index corresponding to the parameter combination do not meet a third preset condition corresponding to the index evaluation system, acquiring a new parameter combination, taking the new parameter combination as the parameter combination, executing the training sample to be input into a preset machine learning model, and training the machine learning model based on the parameter combination to obtain a first evaluation index corresponding to the parameter combination.
In addition, in order to achieve the above object, the present invention also provides a credit card model construction apparatus, including:
The acquisition module is used for acquiring credit behavior data of the borrower and determining a training sample and a verification sample based on the credit behavior data;
the training module is used for inputting the training sample into a preset machine learning model, training the machine learning model based on parameter combinations, and obtaining a first evaluation index corresponding to the parameter combinations, wherein one parameter combination correspondingly trains one machine learning model based on the same training sample;
the first determining module is used for inputting the verification sample into the trained machine learning model after the machine learning model is trained, and determining a second evaluation index corresponding to the parameter combination;
and the second determining module is used for determining the optimal parameter combination in the parameter combinations based on the first evaluation index and the second evaluation index corresponding to the parameter combinations so as to determine a credit score card model corresponding to the machine learning model based on the optimal parameter combination.
In addition, in order to achieve the above object, the present invention also provides a credit card model construction apparatus including: the credit card model building system comprises a memory, a processor and a credit card model building program which is stored in the memory and can run on the processor, wherein the credit card model building program realizes the steps of the credit card model building method when being executed by the processor.
In addition, in order to achieve the above object, the present invention also provides a readable storage medium having stored thereon a credit card model building program which, when executed by a processor, implements the steps of the credit card model building method as described above.
The invention determines a training sample and a verification sample by acquiring credit behavior data of a borrower and based on the credit behavior data; inputting the training sample into a preset machine learning model, and training the machine learning model based on parameter combinations to obtain a first evaluation index corresponding to the parameter combinations, wherein one parameter combination correspondingly trains one machine learning model based on the same training sample; after the machine learning model is trained, inputting the verification sample into the trained machine learning model, and determining a second evaluation index corresponding to the parameter combination; and determining an optimal parameter combination in the parameter combinations based on the first evaluation index and the second evaluation index corresponding to the parameter combinations, so as to determine a credit score card model corresponding to the machine learning model based on the optimal parameter combination. According to the invention, machine learning is trained according to different parameter combinations to obtain the first evaluation index and the second evaluation index corresponding to the parameter combinations, and then the corresponding machine learning model is evaluated according to the first evaluation index and the second evaluation index, so that the training model training effect of the parameter combinations is evaluated, the optimal parameter combinations in the parameter combinations can be determined according to the first evaluation index and the second evaluation index, the manual screening model evaluation index is avoided, the technical problem that the modeling efficiency of the traditional credit score card model modeling technology is low due to the mode of manually screening the model evaluation index is solved, and the modeling efficiency of the credit score card model is improved.
Drawings
FIG. 1 is a schematic diagram of a credit score card model building device of a hardware running environment according to an embodiment of the present invention;
FIG. 2 is a flowchart of a credit score card model building method according to a first embodiment of the present invention;
FIG. 3 is a flowchart of a credit score card model building method according to a second embodiment of the present invention.
The achievement of the objects, functional features and advantages of the present invention will be further described with reference to the accompanying drawings, in conjunction with the embodiments.
Detailed Description
It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
Referring to fig. 1, fig. 1 is a schematic structural diagram of a credit score card model building device of a hardware running environment according to an embodiment of the present invention.
The credit score card model building device of the embodiment of the invention can be a PC, and also can be mobile terminal devices with display functions, such as a smart phone, a tablet personal computer, an electronic book reader, an MP3 (Moving Picture Experts Group Audio Layer III, dynamic image expert compression standard audio layer 3) player, an MP4 (Moving Picture Experts Group Audio Layer IV, dynamic image expert compression standard audio layer 4) player, a portable computer and the like.
As shown in fig. 1, the credit score card model building apparatus may include: a processor 1001, such as a CPU, a network interface 1004, a user interface 1003, a memory 1005, a communication bus 1002. Wherein the communication bus 1002 is used to enable connected communication between these components. The user interface 1003 may include a Display, an input unit such as a Keyboard (Keyboard), and the optional user interface 1003 may further include a standard wired interface, a wireless interface. The network interface 1004 may optionally include a standard wired interface, a wireless interface (e.g., WI-FI interface). The memory 1005 may be a high-speed RAM memory or a stable memory (non-volatile memory), such as a disk memory. The memory 1005 may also optionally be a storage device separate from the processor 1001 described above.
Optionally, the credit card model building device may further include a camera, an RF (Radio Frequency) circuit, a sensor, an audio circuit, a WiFi module, and the like. Among other sensors, such as light sensors, motion sensors, and other sensors. In particular, the light sensor may include an ambient light sensor that may adjust the brightness of the display screen according to the brightness of ambient light, and a proximity sensor that may turn off the display screen and/or backlight when the credit card model building device is moved to the ear. As one of the motion sensors, the gravity acceleration sensor can detect the acceleration in all directions (generally three axes), and can detect the gravity and the direction when the device is stationary, and the device can be used for identifying the application of the gesture of credit score card model building equipment (such as horizontal and vertical screen switching, related games and magnetometer gesture calibration), vibration identification related functions (such as pedometer and knocking) and the like; of course, the mobile terminal may also be configured with other sensors such as a gyroscope, a barometer, a hygrometer, a thermometer, an infrared sensor, and the like, which are not described herein.
It will be appreciated by those skilled in the art that the credit card model building device structure shown in fig. 1 does not constitute a limitation of the credit card model building device, and may include more or fewer components than shown, or may combine certain components, or a different arrangement of components.
As shown in fig. 1, an operating system, a network communication module, a user interface module, and a credit card model building program may be included in a memory 1005, which is a type of computer storage medium.
In the credit scoring card model building device shown in fig. 1, the network interface 1004 is mainly used for connecting to a background server and performing data communication with the background server; the user interface 1003 is mainly used for connecting a client (user side) and performing data communication with the client; and processor 1001 may be used to invoke a credit score card model building program stored in memory 1005.
In the present embodiment, the credit score card model construction apparatus includes: the system comprises a memory 1005, a processor 1001 and a credit score card model building program which is stored in the memory 1005 and can run on the processor 1001, wherein when the processor 1001 calls the credit score card model building program stored in the memory 1005, the processor 1001 performs the following operations:
Acquiring credit behavior data of a borrower, and determining a training sample and a verification sample based on the credit behavior data;
inputting the training sample into a preset machine learning model, and training the machine learning model based on parameter combinations to obtain a first evaluation index corresponding to the parameter combinations, wherein one parameter combination correspondingly trains one machine learning model based on the same training sample;
after the machine learning model is trained, inputting the verification sample into the trained machine learning model, and determining a second evaluation index corresponding to the parameter combination;
and determining an optimal parameter combination in the parameter combinations based on the first evaluation index and the second evaluation index corresponding to the parameter combinations, so as to determine a credit score card model corresponding to the machine learning model based on the optimal parameter combination.
Further, the processor 1001 may call the credit score card model building program stored in the memory 1005, and further perform the following operations:
determining a first target parameter combination corresponding to the first KS index and the second KS index when the first KS index and the second KS index meet a first preset condition based on the first KS index and the second KS index corresponding to the parameter combination;
Determining a second target parameter combination corresponding to the PSI index corresponding to the first target parameter combination when the PSI index corresponding to the first target parameter combination meets a second preset condition based on the PSI index corresponding to the first target parameter combination;
and when a second target parameter combination corresponding to the PSI index corresponding to the second target parameter combination meets a second preset condition exists, the second target parameter combination is used as the optimal parameter combination in the parameter combinations.
Further, the processor 1001 may call the credit score card model building program stored in the memory 1005, and further perform the following operations:
determining that the parameter combination corresponding to the second KS index larger than a first preset threshold value is a third target parameter combination based on the second KS index corresponding to the parameter combination;
and determining a third target parameter combination corresponding to the third target parameter combination as a first target parameter combination when the difference value between the first KS index and the second KS index corresponding to the third target parameter combination is smaller than a second preset threshold value based on the first KS index and the second KS index corresponding to the third target parameter combination.
Further, the processor 1001 may call the credit score card model building program stored in the memory 1005, and further perform the following operations: and based on the PSI index corresponding to the first target parameter combination, taking the first target parameter combination corresponding to the smallest PSI index in the PSI indexes corresponding to the first target parameter combination as a second target parameter combination.
Further, the processor 1001 may call the credit score card model building program stored in the memory 1005, and further perform the following operations:
determining whether the first evaluation index and the second evaluation index corresponding to the parameter combination meet a third preset condition corresponding to the index evaluation system or not based on a preset index evaluation system;
and if the first evaluation index and the second evaluation index corresponding to the parameter combination meet a third preset condition corresponding to the index evaluation system, determining the optimal parameter combination in the parameter combination based on the first evaluation index and the second evaluation index corresponding to the parameter combination.
Further, the processor 1001 may call the credit score card model building program stored in the memory 1005, and further perform the following operations:
determining whether the first evaluation index corresponding to the parameter combination is in a first index range corresponding to the index evaluation system or not based on a preset index evaluation system, and determining whether the second evaluation index corresponding to the parameter combination is in a second index range corresponding to the index evaluation system or not;
if the first evaluation index and the second evaluation index corresponding to the parameter combination meet a third preset condition corresponding to the index evaluation system, determining the optimal parameter combination in the parameter combination based on the first evaluation index and the second evaluation index corresponding to the parameter combination comprises:
And if the first evaluation index corresponding to the parameter combination is in a first index range corresponding to the index evaluation system and the second evaluation index corresponding to the parameter combination is in a second index range corresponding to the index evaluation system, determining the optimal parameter combination in the parameter combination based on the first evaluation index and the second evaluation index corresponding to the parameter combination.
Further, the processor 1001 may call the credit score card model building program stored in the memory 1005, and further perform the following operations:
and if the first evaluation index and the second evaluation index corresponding to the parameter combination do not meet a third preset condition corresponding to the index evaluation system, acquiring a new parameter combination, taking the new parameter combination as the parameter combination, executing the training sample to be input into a preset machine learning model, and training the machine learning model based on the parameter combination to obtain a first evaluation index corresponding to the parameter combination.
The invention also provides a credit score card model construction method, referring to fig. 2, fig. 2 is a flow chart of a first embodiment of the credit score card model construction method of the invention.
In this embodiment, the credit score card model construction method includes the steps of:
step S10, credit behavior data of borrowers are obtained, and training samples and verification samples are determined based on the credit behavior data;
the credit score card model is a decision tool for analyzing and mining credit behavior data of borrowers, predicting default probabilities of the borrowers in a certain period according to the credit behavior data of the borrowers, and obtaining credit scores of different grades so as to conduct risk management and control.
In this embodiment, the credit data is a credit history record of the borrower and a business performance record of the borrower in a certain period, wherein the credit history record is a personal credit record recorded by the borrower at a people bank, and the business performance record is recorded data of the behavior of the borrower about loan business at the loan institution or other loan institutions, including loan amount, borrowing time, repayment time, and the like. Firstly, acquiring credit behavior data of a borrower, extracting preset characteristic data in the credit behavior data as training samples and verification samples for modeling, training a machine learning model based on the training samples corresponding to the credit behavior data of the borrower, and verifying the machine learning model based on the verification samples. The credit line data includes various characteristic data of the borrower, such as age, sex, loan business processed, loan amount, borrowing time or repayment time, etc.
Step S20, inputting the training sample into a preset machine learning model, and training the machine learning model based on a parameter combination to obtain a first evaluation index corresponding to the parameter combination, wherein one parameter combination correspondingly trains one machine learning model based on the same training sample;
in this embodiment, after obtaining a training sample corresponding to credit behavior data of a borrower, the training sample corresponding to the credit behavior data is input into a machine learning model, so as to train the machine learning model based on the training sample, and obtain a first evaluation index corresponding to a parameter combination. The first evaluation index is a model evaluation index output by the machine learning model based on training samples. Further, it should be noted that the parameter combinations include a plurality of parameters, and one machine learning model is trained based on one parameter combination by using the same training sample, that is, a plurality of parameter combinations correspondingly train a plurality of machine learning models.
Specifically, after the training sample is input into the machine learning model, for each model parameter of the machine learning model, the model parameter includes a plurality of undetermined model parameters which can be selected, and one target undetermined model parameter is arbitrarily selected from the plurality of undetermined model parameters corresponding to the model parameter, so as to obtain a parameter combination, wherein the parameter combination is a combination corresponding to the target undetermined model parameter selected from the undetermined model parameters. After the parameter combinations corresponding to the machine learning model are obtained, training the machine learning model based on different parameter combinations. Specifically, setting model parameters of a machine learning model as target undetermined parameters corresponding to the parameter combination, so as to train the machine learning model based on the parameter combination, and obtaining a first evaluation index corresponding to the parameter combination.
Step S30, after the machine learning model is trained, inputting the verification sample into the trained machine learning model, and determining a second evaluation index corresponding to the parameter combination;
in this embodiment, after training the machine learning model based on the training sample is completed, the verification sample is input into the trained machine learning model to perform model verification on the trained machine learning model, so that the machine learning model outputs a second evaluation index corresponding to the parameter combination based on the verification sample. The second evaluation index is a model evaluation index output by the machine learning model based on the verification sample. Further, it should be noted that, one machine learning model is trained based on one parameter combination, that is, a plurality of parameter combinations correspondingly train a plurality of machine learning models, and the first evaluation index and the second evaluation index output by the machine learning model correspond to each other.
And step S40, determining the optimal parameter combination in the parameter combinations based on the first evaluation index and the second evaluation index corresponding to the parameter combination, so as to determine a credit score card model corresponding to the machine learning model based on the optimal parameter combination.
In this embodiment, after the first evaluation index and the second evaluation index corresponding to the parameter combination are obtained, the machine learning model is evaluated according to the first evaluation index and the second evaluation index corresponding to the parameter combination, so that the machine learning model obtained by training different parameter combinations is evaluated. In this embodiment, the corresponding machine learning model is evaluated according to the first evaluation index and the second evaluation index corresponding to the parameter combination, so as to evaluate the model effect of the machine learning model obtained by training based on the parameter combination, thereby obtaining the effect of the parameter combination. Therefore, after the machine learning model corresponding to the parameter combination is evaluated, determining the optimal parameter combination from the plurality of parameter combinations according to the first evaluation index and the second evaluation index corresponding to the parameter combination, and finally taking the machine learning model trained by the optimal parameter combination as a credit score card model.
Further, after the credit score card model is obtained, credit behavior data of the client to be predicted is obtained; the credit behavior data of the clients to be predicted are input into a credit score card model to determine credit risk prediction results of the clients to be predicted based on the credit score card model.
The credit score card model construction method provided by the embodiment is characterized in that credit behavior data of borrowers are obtained, and training samples and verification samples are determined based on the credit behavior data; inputting the training sample into a preset machine learning model, and training the machine learning model based on parameter combinations to obtain a first evaluation index corresponding to the parameter combinations, wherein one parameter combination correspondingly trains one machine learning model based on the same training sample; after the machine learning model is trained, inputting the verification sample into the trained machine learning model, and determining a second evaluation index corresponding to the parameter combination; and determining an optimal parameter combination in the parameter combinations based on the first evaluation index and the second evaluation index corresponding to the parameter combinations, so as to determine a credit score card model corresponding to the machine learning model based on the optimal parameter combination. According to the invention, machine learning is trained according to different parameter combinations to obtain the first evaluation index and the second evaluation index corresponding to the parameter combinations, and then the corresponding machine learning model is evaluated according to the first evaluation index and the second evaluation index, so that the training model training effect of the parameter combinations is evaluated, the optimal parameter combinations in the parameter combinations can be determined according to the first evaluation index and the second evaluation index, the manual screening model evaluation index is avoided, the technical problem that the modeling efficiency of the traditional credit score card model modeling technology is low due to the mode of manually screening the model evaluation index is solved, and the modeling efficiency of the credit score card model is improved.
Based on the first embodiment, a second embodiment of the credit score card model building method of the present invention is proposed, referring to fig. 3, in this embodiment, step S40 includes:
step S41, determining a first target parameter combination corresponding to the first KS index and the second KS index meeting a first preset condition based on the first KS index and the second KS index corresponding to the parameter combination;
step S42, based on PSI indexes corresponding to the first target parameter combinations, determining second target parameter combinations corresponding to PSI indexes corresponding to the first target parameter combinations when the PSI indexes meet second preset conditions;
and step S43, when a second target parameter combination corresponding to the PSI index corresponding to the second target parameter combination meets a second preset condition exists, the second target parameter combination is used as the optimal parameter combination in the parameter combinations.
In this embodiment, the first evaluation index and the second evaluation index are model evaluation indexes of the machine learning model, the first evaluation index includes a first KS index output by the machine learning model based on the training sample, and the second evaluation index includes a second KS index and a PSI index output by the machine learning model based on the verification sample. Wherein KS (Kolmogorov-Smirnov) is the difference between cumulative distributions of good and bad samples. The larger the cumulative difference between the good and bad samples, the larger the KS index, and the stronger the risk distinguishing capability of the model. The PSI (Population Stability Index, population stability) index, which in this embodiment is used to measure the stability of the machine learning model under parameter combination training, is the feature stability.
Specifically, for each parameter combination, firstly, based on a first KS index and a second KS index, whether the first KS index and the second KS index corresponding to each parameter combination meet a first preset condition is detected, and the parameter combination corresponding to the first KS index and the second KS index meeting the first preset condition is used as a first target parameter combination. When the first target parameter combination exists, determining whether the PSI index corresponding to the first target parameter combination meets a second preset condition according to the PSI index corresponding to the first target parameter combination, and taking the first target parameter combination, of which the PSI index corresponding to the first target parameter combination meets the second preset condition, as a second target parameter combination. When the second target parameter combination exists, the second target parameter combination is taken as the optimal parameter combination in the parameter combinations.
Further, the step of determining, based on the first KS indicator and the second KS indicator corresponding to the parameter combination, a first target parameter combination corresponding to when the first KS indicator and the second KS indicator satisfy a first preset condition includes:
step S411, based on the second KS indicator corresponding to the parameter combination, determining that the parameter combination corresponding to the second KS indicator being greater than the first preset threshold is a third target parameter combination;
Step S412, based on the first KS index and the second KS index corresponding to the third target parameter combination, determines the third target parameter combination corresponding to the third target parameter combination as the first target parameter combination when the difference between the first KS index and the second KS index corresponding to the third target parameter combination is smaller than the second preset threshold.
In this embodiment, specifically, for each parameter combination, based on the second KS index corresponding to each parameter combination, whether the second KS index corresponding to the parameter combination is greater than the first preset threshold is detected, and the parameter combination corresponding to the second KS index greater than the first preset threshold is used as the third target parameter combination. When the third target parameter combination exists, calculating a difference value between the first KS index and the second KS index corresponding to the third target parameter combination based on the first KS index and the second KS index corresponding to the third target parameter combination, and taking the third target parameter combination corresponding to the third target parameter combination when the difference value between the first KS index and the second KS index corresponding to the third target parameter combination is smaller than a second preset threshold value as the first target parameter combination.
It should be noted that, in this embodiment, by detecting whether the second KS index corresponding to the parameter combination is greater than the first preset threshold, to detect the model effect of the machine learning model trained based on the parameter combination, the greater the second KS index, the better the model effect of the machine learning model; and comparing the difference value between the first KS index and the second KS index corresponding to the third target parameter combination with a second preset threshold value by calculating the difference value between the first KS index and the second KS index corresponding to the third target parameter combination so as to detect whether the machine learning model is over-fitted, and specifically, if the difference value between the first KS index and the second KS index corresponding to the third target parameter combination is greater than or equal to the second preset threshold value, the corresponding machine learning model is over-fitted.
Further, the step of determining, based on the PSI index corresponding to the first target parameter combination, a second target parameter combination corresponding to the PSI index corresponding to the first target parameter combination when the PSI index meets a second preset condition includes:
step S421, based on the PSI index corresponding to the first target parameter combination, uses the first target parameter combination corresponding to the smallest PSI index in the PSI indexes corresponding to the first target parameter combination as the second target parameter combination.
In this embodiment, after evaluating the machine learning model according to the first KS index and the second KS index to determine a first target parameter combination satisfying a first preset condition in the parameter combinations, determining, according to the PSI index corresponding to the first target parameter combination, a second target parameter combination corresponding to the PSI index corresponding to the first target parameter combination satisfying a second preset condition. Specifically, according to the PSI indexes corresponding to the first target parameter combination, determining the smallest PSI index in the PSI indexes corresponding to the first target parameter combination, and taking the first target parameter combination corresponding to the smallest PSI index in the PSI indexes corresponding to the first target parameter combination as the second target parameter combination. In this embodiment, the PSI indexes corresponding to the parameter combinations are compared with each other to detect the model effect of the machine learning model trained based on the parameter combinations, and the smaller the PSI index, the better the model effect of the machine learning model.
Further, the step of determining the optimal parameter combination of the parameter combinations based on the first evaluation index and the second evaluation index corresponding to the parameter combinations includes:
step S44, determining whether the first evaluation index and the second evaluation index corresponding to the parameter combination meet a third preset condition corresponding to the index evaluation system or not based on a preset index evaluation system;
step S45, if the first evaluation index and the second evaluation index corresponding to the parameter combination meet a third preset condition corresponding to the index evaluation system, determining an optimal parameter combination in the parameter combinations based on the first evaluation index and the second evaluation index corresponding to the parameter combination.
In this embodiment, after obtaining the model evaluation indexes corresponding to the different parameter combinations, that is, after obtaining the first evaluation index and the second evaluation index, it is detected whether the first evaluation index and the second evaluation index corresponding to the parameter combinations satisfy a third preset condition corresponding to a preset index evaluation system. And if the first evaluation index and the second evaluation index corresponding to the parameter combination meet a third preset condition, indicating that the error of the model evaluation index is within an acceptable range, evaluating the machine learning model according to the first evaluation index and the second evaluation index output by the machine learning model so as to evaluate different parameter combinations. And if the first evaluation index and the second evaluation index corresponding to the parameter combination do not meet the third preset condition, indicating that the error of the model evaluation index is too large, eliminating the parameter combination corresponding to the model evaluation index which does not meet the third preset condition from the parameter combination, and executing the parameter combination corresponding to the model evaluation index which meets the third preset condition, inputting the training sample into a preset machine learning model, and training the machine learning model based on the parameter combination to obtain the first evaluation index corresponding to the parameter combination.
Further, the step of determining whether the first evaluation index and the second evaluation index corresponding to the parameter combination meet a third preset condition corresponding to the index evaluation system based on a preset index evaluation system includes:
step S441, based on a preset index evaluation system, determining whether the first evaluation index corresponding to the parameter combination is within a first index range corresponding to the index evaluation system, and determining whether the second evaluation index corresponding to the parameter combination is within a second index range corresponding to the index evaluation system;
if the first evaluation index and the second evaluation index corresponding to the parameter combination meet a third preset condition corresponding to the index evaluation system, determining the optimal parameter combination in the parameter combination based on the first evaluation index and the second evaluation index corresponding to the parameter combination comprises:
step S451, if the first evaluation index corresponding to the parameter combination is within a first index range corresponding to the index evaluation system and the second evaluation index corresponding to the parameter combination is within a second index range corresponding to the index evaluation system, determining an optimal parameter combination in the parameter combinations based on the first evaluation index and the second evaluation index corresponding to the parameter combination.
In this embodiment, after obtaining the model evaluation indexes corresponding to the different parameter combinations, the first evaluation index and the second evaluation index are obtained, it is determined whether the first evaluation index corresponding to the parameter combination is within the first index range corresponding to the index evaluation system, and it is determined whether the second evaluation index corresponding to the parameter combination is within the second index range corresponding to the index evaluation system. If the first evaluation index is in a first index range corresponding to the index evaluation system and the second evaluation index is in a second index range corresponding to the index evaluation system, and the model effect is in an acceptance range, evaluating the machine learning model according to the first evaluation index and the second evaluation index output by the machine learning model so as to evaluate different parameter combinations.
Further, after the step of determining whether the first evaluation index and the second evaluation index corresponding to the parameter combination meet a third preset condition corresponding to the index evaluation system based on a preset index evaluation system, the method further includes:
step S46, if the first evaluation index and the second evaluation index corresponding to the parameter combination do not meet the third preset condition corresponding to the index evaluation system, obtaining a new parameter combination, using the new parameter combination as the parameter combination, executing the training sample to be input to a preset machine learning model, and training the machine learning model based on the parameter combination to obtain the first evaluation index corresponding to the parameter combination.
In this embodiment, further, if the first evaluation index and the second evaluation index corresponding to the parameter combination do not meet the third preset condition, it is indicated that the training effect of the machine learning model based on the parameter combination is not up to standard, a new parameter combination may be obtained, then the machine model is trained based on the new parameter combination and the original training sample and verification sample, so as to determine the first evaluation index and the second evaluation index corresponding to the training sample, and the step of determining the best parameter combination in the parameter combination based on the first evaluation index and the second evaluation index corresponding to the parameter combination is performed, so as to determine the credit score card model corresponding to the machine learning model based on the best parameter combination.
According to the credit score card model construction method, a first target parameter combination corresponding to the first KS index and the second KS index meeting a first preset condition is determined based on the first KS index and the second KS index corresponding to the parameter combination; determining a second target parameter combination corresponding to the PSI index corresponding to the first target parameter combination when the PSI index corresponding to the first target parameter combination meets a second preset condition based on the PSI index corresponding to the first target parameter combination; and when a second target parameter combination corresponding to the PSI index corresponding to the second target parameter combination meets a second preset condition exists, the second target parameter combination is used as the optimal parameter combination in the parameter combinations. According to the invention, the corresponding machine learning model is evaluated according to the first evaluation index and the second evaluation index, so that the training model training effect of the parameter combination is evaluated, the optimal parameter combination in the parameter combination can be determined according to the first evaluation index and the second evaluation index, the manual screening of the model evaluation index is avoided, the technical problem of low modeling efficiency of the conventional credit score card model modeling technology caused by the manual screening of the model evaluation index is solved, and the modeling efficiency of the credit score card model is improved.
In addition, the embodiment of the invention also provides a credit score card model construction device, which comprises:
the acquisition module is used for acquiring credit behavior data of the borrower and determining a training sample and a verification sample based on the credit behavior data;
the training module is used for inputting the training sample into a preset machine learning model, training the machine learning model based on parameter combinations, and obtaining a first evaluation index corresponding to the parameter combinations, wherein one parameter combination correspondingly trains one machine learning model based on the same training sample;
the first determining module is used for inputting the verification sample into the trained machine learning model after the machine learning model is trained, and determining a second evaluation index corresponding to the parameter combination;
and the second determining module is used for determining the optimal parameter combination in the parameter combinations based on the first evaluation index and the second evaluation index corresponding to the parameter combinations so as to determine a credit score card model corresponding to the machine learning model based on the optimal parameter combination.
Further, the second determining module is further configured to:
Determining a first target parameter combination corresponding to the first KS index and the second KS index when the first KS index and the second KS index meet a first preset condition based on the first KS index and the second KS index corresponding to the parameter combination;
determining a second target parameter combination corresponding to the PSI index corresponding to the first target parameter combination when the PSI index corresponding to the first target parameter combination meets a second preset condition based on the PSI index corresponding to the first target parameter combination;
and when a second target parameter combination corresponding to the PSI index corresponding to the second target parameter combination meets a second preset condition exists, the second target parameter combination is used as the optimal parameter combination in the parameter combinations.
Further, the second determining module is further configured to:
determining that the parameter combination corresponding to the second KS index larger than a first preset threshold value is a third target parameter combination based on the second KS index corresponding to the parameter combination;
and determining a third target parameter combination corresponding to the third target parameter combination as a first target parameter combination when the difference value between the first KS index and the second KS index corresponding to the third target parameter combination is smaller than a second preset threshold value based on the first KS index and the second KS index corresponding to the third target parameter combination.
Further, the second determining module is further configured to:
and based on the PSI index corresponding to the first target parameter combination, taking the first target parameter combination corresponding to the smallest PSI index in the PSI indexes corresponding to the first target parameter combination as a second target parameter combination.
Further, the second determining module is further configured to:
determining whether the first evaluation index and the second evaluation index corresponding to the parameter combination meet a third preset condition corresponding to the index evaluation system or not based on a preset index evaluation system;
and if the first evaluation index and the second evaluation index corresponding to the parameter combination meet a third preset condition corresponding to the index evaluation system, determining the optimal parameter combination in the parameter combination based on the first evaluation index and the second evaluation index corresponding to the parameter combination.
Further, the second determining module is further configured to:
determining whether the first evaluation index corresponding to the parameter combination is in a first index range corresponding to the index evaluation system or not based on a preset index evaluation system, and determining whether the second evaluation index corresponding to the parameter combination is in a second index range corresponding to the index evaluation system or not;
And if the first evaluation index corresponding to the parameter combination is in a first index range corresponding to the index evaluation system and the second evaluation index corresponding to the parameter combination is in a second index range corresponding to the index evaluation system, determining the optimal parameter combination in the parameter combination based on the first evaluation index and the second evaluation index corresponding to the parameter combination.
Further, the second determining module is further configured to:
and if the first evaluation index and the second evaluation index corresponding to the parameter combination do not meet a third preset condition corresponding to the index evaluation system, acquiring a new parameter combination, taking the new parameter combination as the parameter combination, executing the training sample to be input into a preset machine learning model, and training the machine learning model based on the parameter combination to obtain a first evaluation index corresponding to the parameter combination.
In addition, the embodiment of the invention also provides a readable storage medium, wherein the readable storage medium stores a credit score card model building program, and the credit score card model building program realizes the steps of the credit score card model building method when being executed by a processor.
The specific embodiments of the readable storage medium of the present invention are substantially the same as the embodiments of the credit score card model building method described above, and will not be described in detail herein.
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or system. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or system that comprises the element.
The foregoing embodiment numbers of the present invention are merely for the purpose of description, and do not represent the advantages or disadvantages of the embodiments.
From the above description of the embodiments, it will be clear to those skilled in the art that the above-described embodiment method may be implemented by means of software plus a necessary general hardware platform, but of course may also be implemented by means of hardware, but in many cases the former is a preferred embodiment. Based on such understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art in the form of a software product stored in a storage medium (e.g. ROM/RAM, magnetic disk, optical disk) as described above, comprising instructions for causing a terminal device (which may be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to perform the method according to the embodiments of the present invention.
The foregoing description is only of the preferred embodiments of the present invention, and is not intended to limit the scope of the invention, but rather is intended to cover any equivalents of the structures or equivalent processes disclosed herein or in the alternative, which may be employed directly or indirectly in other related arts.
Claims (6)
1. The credit score card model construction method is characterized by comprising the following steps of:
acquiring credit behavior data of a borrower, and determining a training sample and a verification sample based on the credit behavior data;
inputting the training sample into a preset machine learning model, and training the machine learning model based on parameter combinations to obtain a first evaluation index corresponding to the parameter combinations, wherein one parameter combination correspondingly trains one machine learning model based on the same training sample;
wherein the first evaluation index comprises a first KS index output by the machine learning model based on the training samples;
after the machine learning model is trained, inputting the verification sample into the trained machine learning model, and determining a second evaluation index corresponding to the parameter combination;
Wherein the second evaluation index includes a second KS index and a PSI index that the machine learning model outputs based on the verification sample;
determining an optimal parameter combination in the parameter combinations based on the first evaluation index and the second evaluation index corresponding to the parameter combinations, so as to determine a credit score card model corresponding to the machine learning model based on the optimal parameter combination;
wherein the step of determining the optimal parameter combination in the parameter combination based on the first evaluation index and the second evaluation index corresponding to the parameter combination includes:
determining a first target parameter combination corresponding to the first KS index and the second KS index when the first KS index and the second KS index meet a first preset condition based on the first KS index and the second KS index corresponding to the parameter combination;
wherein the step of determining, based on the first KS indicator and the second KS indicator corresponding to the parameter combination, a first target parameter combination corresponding to when the first KS indicator and the second KS indicator satisfy a first preset condition includes:
determining that the parameter combination corresponding to the second KS index larger than a first preset threshold value is a third target parameter combination based on the second KS index corresponding to the parameter combination;
Based on a first KS index and a second KS index corresponding to the third target parameter combination, determining the third target parameter combination corresponding to the third target parameter combination as a first target parameter combination when the difference value between the first KS index and the second KS index corresponding to the third target parameter combination is smaller than a second preset threshold value;
determining a second target parameter combination corresponding to the PSI index corresponding to the first target parameter combination when the PSI index corresponding to the first target parameter combination meets a second preset condition based on the PSI index corresponding to the first target parameter combination;
wherein, based on the PSI index corresponding to the first target parameter combination, the step of determining the second target parameter combination corresponding to the PSI index corresponding to the first target parameter combination when the PSI index meets a second preset condition includes:
based on PSI indexes corresponding to the first target parameter combinations, taking a first target parameter combination corresponding to the smallest PSI index in the PSI indexes corresponding to the first target parameter combinations as a second target parameter combination;
when a second target parameter combination corresponding to the PSI index corresponding to the second target parameter combination meets a second preset condition exists, the second target parameter combination is used as an optimal parameter combination in the parameter combinations;
Wherein the step of determining the optimal parameter combination in the parameter combination based on the first evaluation index and the second evaluation index corresponding to the parameter combination includes:
determining whether the first evaluation index and the second evaluation index corresponding to the parameter combination meet a third preset condition corresponding to the index evaluation system or not based on a preset index evaluation system;
and if the first evaluation index and the second evaluation index corresponding to the parameter combination meet a third preset condition corresponding to the index evaluation system, determining the optimal parameter combination in the parameter combination based on the first evaluation index and the second evaluation index corresponding to the parameter combination.
2. The credit card model construction method according to claim 1, wherein the step of determining whether the first and second evaluation indexes corresponding to the parameter combinations satisfy a third preset condition corresponding to the index evaluation system based on a preset index evaluation system includes:
determining whether the first evaluation index corresponding to the parameter combination is in a first evaluation index range corresponding to the index evaluation system or not based on a preset index evaluation system, and determining whether the second evaluation index corresponding to the parameter combination is in a second evaluation index range corresponding to the index evaluation system or not;
If the first evaluation index and the second evaluation index corresponding to the parameter combination meet a third preset condition corresponding to the index evaluation system, determining the optimal parameter combination in the parameter combination based on the first evaluation index and the second evaluation index corresponding to the parameter combination comprises:
and if the first evaluation index corresponding to the parameter combination is in a first evaluation index range corresponding to the index evaluation system and the second evaluation index corresponding to the parameter combination is in a second evaluation index range corresponding to the index evaluation system, determining the optimal parameter combination in the parameter combination based on the first evaluation index and the second evaluation index corresponding to the parameter combination.
3. The credit rating card model construction method according to claim 1, wherein after the step of determining whether the first rating index and the second rating index corresponding to the parameter combination satisfy a third preset condition corresponding to the index rating system based on a preset index rating system, further comprising:
and if the first evaluation index and the second evaluation index corresponding to the parameter combination do not meet a third preset condition corresponding to the index evaluation system, acquiring a new parameter combination, taking the new parameter combination as the parameter combination, executing the training sample to be input into a preset machine learning model, and training the machine learning model based on the parameter combination to obtain a first evaluation index corresponding to the parameter combination.
4. A credit score card model construction apparatus, characterized in that the credit score card model construction apparatus comprises:
the acquisition module is used for acquiring credit behavior data of the borrower and determining a training sample and a verification sample based on the credit behavior data;
the training module is used for inputting the training sample into a preset machine learning model, training the machine learning model based on parameter combinations, and obtaining a first evaluation index corresponding to the parameter combinations, wherein one parameter combination correspondingly trains one machine learning model based on the same training sample;
wherein the first evaluation index comprises a first KS index output by the machine learning model based on the training samples;
the first determining module is used for inputting the verification sample into the trained machine learning model after the machine learning model is trained, and determining a second evaluation index corresponding to the parameter combination;
wherein the second evaluation index includes a second KS index and a PSI index that the machine learning model outputs based on the verification sample;
the second determining module is used for determining an optimal parameter combination in the parameter combinations based on the first evaluation index and the second evaluation index corresponding to the parameter combinations so as to determine a credit score card model corresponding to the machine learning model based on the optimal parameter combination;
Wherein the second determining module is further configured to:
determining a first target parameter combination corresponding to the first KS index and the second KS index when the first KS index and the second KS index meet a first preset condition based on the first KS index and the second KS index corresponding to the parameter combination;
determining that the parameter combination corresponding to the second KS index larger than a first preset threshold value is a third target parameter combination based on the second KS index corresponding to the parameter combination;
based on a first KS index and a second KS index corresponding to the third target parameter combination, determining the third target parameter combination corresponding to the third target parameter combination as a first target parameter combination when the difference value between the first KS index and the second KS index corresponding to the third target parameter combination is smaller than a second preset threshold value;
determining a second target parameter combination corresponding to the PSI index corresponding to the first target parameter combination when the PSI index corresponding to the first target parameter combination meets a second preset condition based on the PSI index corresponding to the first target parameter combination;
based on PSI indexes corresponding to the first target parameter combinations, taking a first target parameter combination corresponding to the smallest PSI index in the PSI indexes corresponding to the first target parameter combinations as a second target parameter combination;
when a second target parameter combination corresponding to the PSI index corresponding to the second target parameter combination meets a second preset condition exists, the second target parameter combination is used as an optimal parameter combination in the parameter combinations;
Determining whether the first evaluation index and the second evaluation index corresponding to the parameter combination meet a third preset condition corresponding to the index evaluation system or not based on a preset index evaluation system;
and if the first evaluation index and the second evaluation index corresponding to the parameter combination meet a third preset condition corresponding to the index evaluation system, determining the optimal parameter combination in the parameter combination based on the first evaluation index and the second evaluation index corresponding to the parameter combination.
5. A credit score card model building apparatus, characterized in that the credit score card model building apparatus comprises: a memory, a processor and a credit card model building program stored on the memory and executable on the processor, which credit card model building program when executed by the processor implements the steps of the credit card model building method according to any one of claims 1 to 3.
6. A readable storage medium, wherein a credit card model building program is stored on the readable storage medium, which, when executed by a processor, implements the steps of the credit card model building method according to any one of claims 1 to 3.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110139379.4A CN112734568B (en) | 2021-01-29 | 2021-01-29 | Credit scoring card model construction method, device, equipment and readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110139379.4A CN112734568B (en) | 2021-01-29 | 2021-01-29 | Credit scoring card model construction method, device, equipment and readable storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112734568A CN112734568A (en) | 2021-04-30 |
| CN112734568B true CN112734568B (en) | 2024-01-12 |
Family
ID=75595164
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110139379.4A Active CN112734568B (en) | 2021-01-29 | 2021-01-29 | Credit scoring card model construction method, device, equipment and readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112734568B (en) |
Citations (20)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106897918A (en) * | 2017-02-24 | 2017-06-27 | 上海易贷网金融信息服务有限公司 | A kind of hybrid machine learning credit scoring model construction method |
| CN108154430A (en) * | 2017-12-28 | 2018-06-12 | 上海氪信信息技术有限公司 | A kind of credit scoring construction method based on machine learning and big data technology |
| CN108648068A (en) * | 2018-05-16 | 2018-10-12 | 长沙农村商业银行股份有限公司 | A kind of assessing credit risks method and system |
| CN109447461A (en) * | 2018-10-26 | 2019-03-08 | 北京三快在线科技有限公司 | User credit appraisal procedure and device, electronic equipment, storage medium |
| CN109670940A (en) * | 2018-11-12 | 2019-04-23 | 深圳壹账通智能科技有限公司 | Credit Risk Assessment Model generation method and relevant device based on machine learning |
| CN109993412A (en) * | 2019-03-01 | 2019-07-09 | 百融金融信息服务股份有限公司 | The construction method and device of risk evaluation model, storage medium, computer equipment |
| CN110046984A (en) * | 2019-03-01 | 2019-07-23 | 安徽省优质采科技发展有限责任公司 | Enterprise credit risk system and evaluation method |
| CN110163743A (en) * | 2019-04-28 | 2019-08-23 | 钛镕智能科技(苏州)有限公司 | A kind of credit-graded approach based on hyperparameter optimization |
| CN110472743A (en) * | 2019-07-31 | 2019-11-19 | 北京百度网讯科技有限公司 | Processing method and processing device, equipment and the readable medium that feature is passed through in sample set |
| CN110909970A (en) * | 2018-09-17 | 2020-03-24 | 北京京东金融科技控股有限公司 | Credit scoring method and device |
| CN111080397A (en) * | 2019-11-18 | 2020-04-28 | 支付宝(杭州)信息技术有限公司 | Credit evaluation method and device and electronic equipment |
| CN111080442A (en) * | 2019-12-21 | 2020-04-28 | 湖南大学 | Credit scoring model construction method, device, equipment and storage medium |
| CN111311400A (en) * | 2020-03-30 | 2020-06-19 | 百维金科(上海)信息科技有限公司 | Modeling method and system of grading card model based on GBDT algorithm |
| CN111583024A (en) * | 2020-05-08 | 2020-08-25 | 南京甄视智能科技有限公司 | Credit evaluation method, device, storage medium and server |
| CN111652710A (en) * | 2020-06-03 | 2020-09-11 | 北京化工大学 | Personal credit risk assessment method based on ensemble tree feature extraction and Logistic regression |
| CN111768285A (en) * | 2019-04-01 | 2020-10-13 | 杭州金智塔科技有限公司 | Credit wind control model construction system and method, wind control system and storage medium |
| CN111931876A (en) * | 2020-10-12 | 2020-11-13 | 支付宝(杭州)信息技术有限公司 | Target data side screening method and system for distributed model training |
| CN112037005A (en) * | 2020-07-21 | 2020-12-04 | 苏宁金融科技(南京)有限公司 | Fusion method and device of score cards, computer equipment and storage medium |
| CN112085595A (en) * | 2020-09-27 | 2020-12-15 | 中国建设银行股份有限公司 | Credit scoring model monitoring method and device |
| CN112257873A (en) * | 2020-11-11 | 2021-01-22 | 深圳前海微众银行股份有限公司 | Training method, device, system, equipment and storage medium of machine learning model |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107704495B (en) * | 2017-08-25 | 2018-08-10 | 平安科技（深圳）有限公司 | Training method, device and the computer readable storage medium of subject classification device |
| CN107798390B (en) * | 2017-11-22 | 2023-03-21 | 创新先进技术有限公司 | Training method and device of machine learning model and electronic equipment |
| US10417528B2 (en) * | 2018-02-18 | 2019-09-17 | Sas Institute Inc. | Analytic system for machine learning prediction model selection |
-
2021
- 2021-01-29 CN CN202110139379.4A patent/CN112734568B/en active Active
Patent Citations (20)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106897918A (en) * | 2017-02-24 | 2017-06-27 | 上海易贷网金融信息服务有限公司 | A kind of hybrid machine learning credit scoring model construction method |
| CN108154430A (en) * | 2017-12-28 | 2018-06-12 | 上海氪信信息技术有限公司 | A kind of credit scoring construction method based on machine learning and big data technology |
| CN108648068A (en) * | 2018-05-16 | 2018-10-12 | 长沙农村商业银行股份有限公司 | A kind of assessing credit risks method and system |
| CN110909970A (en) * | 2018-09-17 | 2020-03-24 | 北京京东金融科技控股有限公司 | Credit scoring method and device |
| CN109447461A (en) * | 2018-10-26 | 2019-03-08 | 北京三快在线科技有限公司 | User credit appraisal procedure and device, electronic equipment, storage medium |
| CN109670940A (en) * | 2018-11-12 | 2019-04-23 | 深圳壹账通智能科技有限公司 | Credit Risk Assessment Model generation method and relevant device based on machine learning |
| CN109993412A (en) * | 2019-03-01 | 2019-07-09 | 百融金融信息服务股份有限公司 | The construction method and device of risk evaluation model, storage medium, computer equipment |
| CN110046984A (en) * | 2019-03-01 | 2019-07-23 | 安徽省优质采科技发展有限责任公司 | Enterprise credit risk system and evaluation method |
| CN111768285A (en) * | 2019-04-01 | 2020-10-13 | 杭州金智塔科技有限公司 | Credit wind control model construction system and method, wind control system and storage medium |
| CN110163743A (en) * | 2019-04-28 | 2019-08-23 | 钛镕智能科技(苏州)有限公司 | A kind of credit-graded approach based on hyperparameter optimization |
| CN110472743A (en) * | 2019-07-31 | 2019-11-19 | 北京百度网讯科技有限公司 | Processing method and processing device, equipment and the readable medium that feature is passed through in sample set |
| CN111080397A (en) * | 2019-11-18 | 2020-04-28 | 支付宝(杭州)信息技术有限公司 | Credit evaluation method and device and electronic equipment |
| CN111080442A (en) * | 2019-12-21 | 2020-04-28 | 湖南大学 | Credit scoring model construction method, device, equipment and storage medium |
| CN111311400A (en) * | 2020-03-30 | 2020-06-19 | 百维金科(上海)信息科技有限公司 | Modeling method and system of grading card model based on GBDT algorithm |
| CN111583024A (en) * | 2020-05-08 | 2020-08-25 | 南京甄视智能科技有限公司 | Credit evaluation method, device, storage medium and server |
| CN111652710A (en) * | 2020-06-03 | 2020-09-11 | 北京化工大学 | Personal credit risk assessment method based on ensemble tree feature extraction and Logistic regression |
| CN112037005A (en) * | 2020-07-21 | 2020-12-04 | 苏宁金融科技(南京)有限公司 | Fusion method and device of score cards, computer equipment and storage medium |
| CN112085595A (en) * | 2020-09-27 | 2020-12-15 | 中国建设银行股份有限公司 | Credit scoring model monitoring method and device |
| CN111931876A (en) * | 2020-10-12 | 2020-11-13 | 支付宝(杭州)信息技术有限公司 | Target data side screening method and system for distributed model training |
| CN112257873A (en) * | 2020-11-11 | 2021-01-22 | 深圳前海微众银行股份有限公司 | Training method, device, system, equipment and storage medium of machine learning model |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112734568A (en) | 2021-04-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108717663B (en) | Facial tag fraud judging method, device, equipment and medium based on micro expression | |
| CN111932269B (en) | Equipment information processing method and device | |
| CN106293074B (en) | Emotion recognition method and mobile terminal | |
| CN110264093B (en) | Credit model establishing method, device, equipment and readable storage medium | |
| CN110246032A (en) | Risk monitoring and control method, apparatus and computer readable storage medium after loan | |
| CN112529679A (en) | Construction method, device and equipment of enterprise trust model and readable storage medium | |
| CN112100431A (en) | Evaluation method, device and equipment of OCR system and readable storage medium | |
| CN111061394B (en) | Touch force identification method, training method and device of model thereof and electronic system | |
| CN110472162B (en) | Evaluation method, system, terminal and readable storage medium | |
| CN112053363A (en) | Retinal vessel segmentation method and device and model construction method | |
| CN111046742B (en) | Eye behavior detection method, device and storage medium | |
| CN113538070A (en) | User life value cycle detection method and device and computer equipment | |
| CN112906772A (en) | Sample processing method, device, equipment and computer readable storage medium | |
| CN114418748A (en) | Vehicle credit evaluation method, device, equipment and storage medium | |
| CN117115596B (en) | Training method, device, equipment and medium of object action classification model | |
| CN113919432A (en) | Classification model construction method, data classification method and device | |
| CN111476144B (en) | Pedestrian attribute identification model determining method and device and computer readable storage medium | |
| CN113642462A (en) | Driving behavior assessment method and device, terminal equipment and storage medium | |
| CN117422905A (en) | Target identification method, device, equipment and storage medium | |
| CN112734568B (en) | Credit scoring card model construction method, device, equipment and readable storage medium | |
| CN111784053A (en) | Transaction risk detection method, device and readable storage medium | |
| CN116167454A (en) | Intelligent two-classification model training method and device | |
| CN113269433B (en) | Tax risk prediction method, apparatus, medium and computer program product | |
| CN115762530A (en) | Voiceprint model training method and device, computer equipment and storage medium | |
| CN112529699A (en) | Construction method, device and equipment of enterprise trust model and readable storage medium |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |