CN117726426A - Credit evaluation method, credit evaluation device, electronic equipment and storage medium - Google Patents

Abstract

The application discloses a credit evaluation method, a credit evaluation device, an electronic device and a storage medium, wherein the credit evaluation method comprises the following steps: acquiring each piece of to-be-evaluated data of the target object corresponding to each target evaluation index according to each target evaluation index predicted by the evaluation index prediction model; and executing credit rating evaluation according to the data to be evaluated of the target object corresponding to each target evaluation index by using the credit evaluation model to obtain a credit rating evaluation result of the target object. Therefore, the method and the device can obviously improve the recognition rate of the bad samples and improve the accuracy of the credit rating evaluation result under the condition that the good samples are ensured not to be misplaced.

Description

Credit evaluation method, credit evaluation device, electronic equipment and storage medium

Technical Field

The present disclosure relates to the field of data processing technologies, and in particular, to a credit evaluation method, a credit evaluation device, an electronic device, and a storage medium.

Background

Artificial intelligence (Artificial Intelligence, AI) is a comprehensive technology of computer science, and by researching the design principles and implementation methods of various intelligent machines, the machines have the functions of sensing, reasoning and decision. Artificial intelligence technology is a comprehensive subject, and relates to a wide range of fields, such as natural language processing technology, machine learning/deep learning and other directions, and with the development of technology, the artificial intelligence technology will be applied in more fields and has an increasingly important value.

In the financial services field, financial institutions provide financial support to businesses, helping them expand business, invest in development, or deal with business needs. The accurate risk control is to guarantee a safe life line of financial service business, and the accuracy of enterprise credit prediction results directly depends on the validity of data characteristics of enterprise objects corresponding to various evaluation indexes. In the prior art, the accuracy of credit evaluation results for enterprise clients is low, so that the problems that high-quality clients are misplaced and the recognition rate for bad clients is low are easily caused.

Disclosure of Invention

In view of this, the embodiments of the present application provide a credit evaluation scheme to improve the accuracy of the credit evaluation result of the client.

According to a first aspect of an embodiment of the present application, there is provided a credit evaluation method, including: according to each target evaluation index predicted from each candidate evaluation index by the evaluation index prediction model, acquiring each piece of to-be-evaluated data of the target object corresponding to each target evaluation index; and executing credit rating evaluation according to the data to be evaluated of the target object corresponding to each target evaluation index by using a credit evaluation model to obtain a credit rating evaluation result of the target object.

According to a second aspect of embodiments of the present application, there is provided a credit evaluation device, including: the index obtaining module is used for obtaining each piece of to-be-evaluated data of the target object corresponding to each target evaluation index according to each target evaluation index predicted by the evaluation index prediction model; and the credit evaluation module is used for executing credit rating evaluation according to the data to be evaluated of the target object corresponding to each target evaluation index by using a credit evaluation model to obtain a credit rating evaluation result of the target object.

According to a third aspect of embodiments of the present application, there is provided an electronic device, including: a memory and a processor, wherein the memory stores a computer program which, when executed by the processor, implements the credit assessment method as described in the first aspect.

According to a fourth aspect of embodiments of the present application, there is provided a computer storage medium storing computer program code which, when executed by a processor, causes the processor to perform the credit assessment method as described in the first aspect.

In summary, according to the credit evaluation scheme provided in the aspects of the present application, each target evaluation index having a large influence on the credit evaluation result is predicted from each candidate evaluation index through the evaluation index prediction model, and then the credit evaluation model performs the credit rating evaluation of the target object according to each to-be-evaluated data of the target object corresponding to each target evaluation index, so that the accuracy of the credit evaluation result can be effectively improved.

Drawings

In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following description will briefly introduce the drawings that are required to be used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are only some embodiments described in the embodiments of the present application, and other drawings may also be obtained according to these drawings for a person having ordinary skill in the art.

Fig. 1 is a process flow diagram of a credit evaluation method according to an exemplary embodiment of the present application.

Fig. 2 is a process flow diagram of a credit evaluation method according to another exemplary embodiment of the present application.

Fig. 3 is a process flow diagram of a credit evaluation method according to another exemplary embodiment of the present application.

Fig. 4 is a process flow diagram of a credit evaluation method according to another exemplary embodiment of the present application.

Fig. 5A is a graph of predicted effects obtained by using a conventional credit evaluation scheme, and fig. 5B is a graph of predicted effects obtained by implementing the credit evaluation method of the present application.

Fig. 6 is a block diagram of a credit evaluation device according to an exemplary embodiment of the present application.

Fig. 7 is a block diagram of an electronic device according to an exemplary embodiment of the present application.

Detailed Description

In order to better understand the technical solutions in the embodiments of the present application, the following descriptions will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only some embodiments of the present application, but not all embodiments. All other embodiments obtained by a person of ordinary skill in the art based on the embodiments in the embodiments of the present application shall fall within the scope of protection of the embodiments of the present application.

Specific implementations of various embodiments of the present application are described in detail below with reference to the accompanying drawings.

Fig. 1 shows a process flow of a credit evaluation method according to an exemplary embodiment of the present application. As shown in the figure, the method 100 of this embodiment mainly includes:

step 102, according to each target evaluation index predicted from each candidate evaluation index by the evaluation index prediction model, each piece of to-be-evaluated data of the target object corresponding to each target evaluation index is obtained.

In some embodiments, the plurality of candidate evaluation indicators may be screened to determine at least one target evaluation indicator of the plurality of candidate evaluation indicators according to the plurality of sample data, each sample data set corresponding to the plurality of candidate evaluation indicators, and the class label of each sample data set, by a trained evaluation indicator prediction model.

In this embodiment, the different sample data sets originate from different financial services institutions, including but not limited to: banks, insurance companies, investment companies, loan companies, etc.

In this embodiment, each evaluation index is used to verify whether the client is credit or not corresponding to different dimensions. By way of example, alternative assessment indicators may include "total amount of tickets for approximately 12 months", "actual business length of the business", "number of tickets for approximately 12 months", "ticket amount for approximately 3 months", and so forth (see table 1 below).

And 104, performing credit rating assessment according to the data to be assessed of the target object corresponding to each target assessment index by using a credit assessment model to obtain a credit rating assessment result of the target object.

In some embodiments, a credit rating model may be used to perform credit rating assessment according to each piece of data to be assessed corresponding to each target assessment index by using the target object, to obtain each item assessment result corresponding to each target assessment index by using the target object, and each item assessment result corresponding to each target assessment index by using the target object, each weight coefficient of each target assessment index, and a given intercept value, to obtain a comprehensive assessment result of the target object, and predict a default probability value of the target object according to the comprehensive assessment result, so as to determine a predicted score of the target object, and convert the predicted score of the target object into a corresponding credit rating according to a corresponding relationship between each predicted score range and each credit rating.

In summary, in this embodiment, each target evaluation index having a large influence on the credit evaluation result is predicted from each candidate evaluation index through the evaluation index prediction model, and then the credit evaluation model performs the credit rating evaluation of the target object according to each to-be-evaluated data of the target object corresponding to each target evaluation index, so that the accuracy of the credit evaluation result can be effectively improved.

Fig. 2 is a process flow of a credit evaluation method according to another embodiment of the application. The present embodiment shows a training scheme for evaluating an index prediction model, and as shown in the figure, the method 200 of the present embodiment mainly includes:

step 202, performing classification prediction on each sample data of each sample data set according to a plurality of sample data of each sample data set corresponding to a plurality of candidate evaluation indexes and given class labels, and predicting the class of each sample data set.

In this embodiment, the category label includes one of a positive sample label and a negative sample label. For example, the sample tag of sample data of good customers (credits above a threshold) may be set to a positive sample tag, and the sample tag of sample data of bad customers (credits below a threshold) may be set to a negative sample tag.

Step 204, calculating the recall rate of each sample data set corresponding to each category label according to the category label and the prediction category of each sample data set.

For example, for negative example labels, the recall for each sample dataset corresponding to the negative example label may be calculated using equation 1 below:

step 206, according to the recall rate of each sample data set corresponding to each category label, a plurality of preset thresholds, and the comprehensive representation value of each category label.

In some embodiments, each sample dataset corresponds to a recall for each category label according to a preset recall threshold, each sample dataset having a recall greater than the preset recall threshold is determined as a target dataset, and a composite representation value for each category label is calculated according to the recall for each target dataset corresponding to each category label, the recall weight value for each target dataset.

For example, for a negative-sample label, the comprehensive representation value of the negative-sample label can be calculated using the following equation 2:

wherein w is _i Represents the i-th target data set, n represents the total number of target data sets, and recovery _i Indicating that the ith target dataset corresponds to the recall of the negative-sample label.

And step 208, optimizing the evaluation index prediction model according to the comprehensive representation value of each sample data set corresponding to each class label.

In this embodiment, the model parameters of the evaluation index prediction model may be optimized according to the comprehensive representation value of each sample data set corresponding to each class label.

Step 210, it is determined whether the model training end condition is satisfied, if yes, step 212 is performed, and if not, step 202 is performed.

In some embodiments, when the recall rate (i.e., the calculation result of step 204) of each sample dataset corresponding to each category label is greater than a preset optimization threshold, a determination may be made that the model training end condition is met.

And 212, obtaining a trained evaluation index prediction model.

In summary, in the evaluation index prediction model trained by the scheme, the target evaluation index with a large influence on the credit rating evaluation can be predicted from the alternative evaluation indexes, so that the accuracy of the credit evaluation result of the subsequent client is improved.

Fig. 3 is a process flow of a credit evaluation method according to another embodiment of the application. The present embodiment shows a training scheme of a credit assessment model, and as shown in the figure, the method 300 of the present embodiment mainly includes:

and 302, performing the binning processing and WOE conversion on the data to be evaluated of the target object corresponding to each target evaluation index according to the multiple binning intervals, so as to obtain a WOE maximum value and a WOE minimum value of the target object corresponding to each target evaluation index.

In some embodiments, the method may include performing binning processing and WOE conversion (Weight of Evidence) on to-be-evaluated data (actual data) of the target object corresponding to each target evaluation index based on a plurality of bin intervals, obtaining a WOE value of each target evaluation index of the target object corresponding to each bin interval, normalizing the WOE value by using an extremum normalization (min-max method) method, obtaining a WOE normalized value of each target evaluation index corresponding to each bin interval, and obtaining a WOE maximum value and a WOE minimum value of each target evaluation index according to the WOE normalized value of each target evaluation index corresponding to each bin interval.

In this embodiment, each binning interval may be set or adjusted based on manual means.

Illustratively, the target evaluation index WOE maximum value and WOE minimum value are shown in Table 2 below:

(Table 2)

And step 304, predicting the credit level of the target object according to the WOE maximum value and the WOE minimum value of the target object corresponding to each target evaluation index, and obtaining the credit prediction value of the target object.

In this embodiment, the maximum value and the minimum value of WOE of the target object corresponding to each target evaluation index may be input into the credit evaluation model to be trained for prediction, so as to obtain the credit prediction value of the target object.

And 306, obtaining a model loss function of the credit evaluation model according to the credit predicted value and the credit label value of the target object, and optimizing the credit evaluation model according to the model loss function.

In this embodiment, a model loss function may be obtained according to the credit prediction value and the credit label value of the target object, and the model parameters of the credit evaluation model may be iteratively updated based on the model loss function to optimize the credit evaluation model.

Step 308, determining whether the model optimization ending condition is satisfied, if yes, proceeding to step 310, otherwise proceeding to step 302.

In some embodiments, when the model parameters of the credit assessment model tend to converge, for example, when the difference between the model parameters of the credit assessment model before and after updating is smaller than a preset threshold, a determination result that the model optimization end condition is satisfied may be obtained.

Step 310, obtaining a trained credit assessment model.

In summary, according to the model training scheme of the embodiment, the accuracy of the credit evaluation result can be effectively improved by performing credit evaluation by using the target evaluation index given by the evaluation index prediction model and performing data preprocessing of the binning and WOE conversion on the data to be evaluated.

Fig. 4 shows a process flow of a credit evaluation method according to another exemplary embodiment of the present application. This example is a specific implementation of step 104 described above. As shown, the scheme 400 of the present embodiment includes:

step 402, performing data normalization processing according to each piece of to-be-evaluated data of each target evaluation index corresponding to each target evaluation target, an evaluation coefficient value of each target evaluation index and a target intercept value, and calculating a comprehensive evaluation value of the target object.

In this embodiment, each piece of data to be evaluated corresponding to each target evaluation index of the target object may be divided according to a plurality of value ranges, and a tag value corresponding to each piece of data to be evaluated of the target object may be obtained according to the value range corresponding to each piece of data to be evaluated.

In this embodiment, the following formula 3 may be used to normalize the tag value of each data to be evaluated, so as to obtain a normalized value of the target object corresponding to each data to be evaluated.

In the above formula 3, the normalized value and the flag value are both one piece of data to be evaluated for which the processing is currently performed, and the minimum value and the maximum value correspond to the minimum flag value and the maximum flag value, respectively, in all pieces of data to be evaluated.

In this embodiment, the weighted calculation may be performed according to the normalized value of the target object corresponding to each data to be evaluated, the target evaluation index of each data to be evaluated, the evaluation coefficient value of each target evaluation index, and the target intercept value, to obtain the comprehensive evaluation value of the target object.

Comprehensive evaluation value of target object = to-be-evaluated data of target evaluation index 1 x evaluation coefficient value of target evaluation index 1+ to-be-evaluated data of target evaluation index 2 x evaluation coefficient value of target evaluation index 2 + … + target intercept value.

For example, referring to fig. 3, the comprehensive evaluation value of the target object=total amount of invoices of approximately 12 months×coefficient (2.557544) +actual business length of the enterprise×coefficient (2.555923) + … + intercept (-9.53695).

(Table 3)

And step 404, according to the comprehensive evaluation value of the target object, performing constraint breaking probability calculation to obtain a constraint breaking probability value of the target object.

In this embodiment, the following equation 4 may be used to perform the calculation of the probability of breach based on the comprehensive evaluation value of the target object, so as to obtain the value of breach probability of the target object.

p=exp (a)/(1+exp (a)) (equation 4)

In the above formula 4, p represents the offending probability value of the target object, a represents the comprehensive evaluation value of the target object, exp () means an exponent based on e.

Step 406, determining the credit evaluation level of the target object according to the default probability value of the target object.

In this embodiment, equation 5 may be used to obtain the predicted value of the target object according to the default probability value of the target object.

s=offset- (factor×ln (p/(1-p))) (equation 5

In the above formula 5, s represents the predictive value of the target object, offset=s () +factor×ln (odds), factor=pdo/ln (2), pdo=2.5, odds=1/80, s () =75.

In some embodiments, the target score range within which the target object falls may be determined from the preset score ranges according to the predicted score of the target object, and the credit rating level of the target object may be determined according to the correspondence between the preset score ranges and the credit rating levels.

In the present embodiment, the correspondence between each preset score range and each credit evaluation level is as shown in the following table 4:

grade	Score range
		5	[83，100]
4	[62，83)
		3	[0，62)

In summary, according to the embodiment, the comprehensive evaluation value of the target object is calculated according to each piece of data to be evaluated of the target object corresponding to each target evaluation index, the default probability of the target object is budgeted according to the comprehensive evaluation value, and the credit level of the target object is predicted based on the default probability, so that the accuracy of the credit level evaluation result can be effectively improved.

Specifically, reference is made to fig. 5A and 5B, where fig. 5A is a statistical diagram of prediction results obtained by using a conventional credit rating evaluation model, and fig. 5B is a statistical diagram of prediction results obtained by using the scheme of the present application. In the statistical diagram of fig. 5A, the prediction result distribution of the high-quality clients and the poor clients is more scattered (the prediction value 60 is used as a demarcation point), and in the statistical diagram of fig. 5B, the prediction result distribution of the high-quality clients and the poor clients is obviously concentrated, so that it can be seen that by adopting the technical scheme of the present application, the recognition rate of the poor clients can be obviously improved under the condition that the high-quality clients are ensured not to be misplaced.

Fig. 6 is a block diagram of a credit evaluation device 600 according to an exemplary embodiment of the present application. As shown in the figure, the credit evaluation device 600 mainly includes:

the index obtaining module 602 is configured to obtain each piece of to-be-evaluated data of the target object corresponding to each target evaluation index according to each target evaluation index predicted by the evaluation index prediction model;

the credit evaluation module 904 is configured to perform credit rating evaluation according to each data to be evaluated of the target object corresponding to each target evaluation index by using a credit evaluation model, so as to obtain a credit rating evaluation result of the target object.

Optionally, the credit assessment device 600 further comprises a training module for training the assessment index prediction model, comprising:

performing a classification prediction on each sample data of each sample data set based on a plurality of sample data, each sample data set corresponding to a plurality of candidate evaluation indicators, given class labels, a prediction class of each sample data set; calculating the recall rate of each sample data set corresponding to each category label according to the category label and the prediction category of each sample data set; calculating the comprehensive representation value of each category label according to the recall rate of each sample data set corresponding to each category label and the recall weight value of each sample data set; and optimizing the evaluation index prediction model according to the comprehensive representation value of each sample data set corresponding to each class label to obtain a trained evaluation index prediction model.

Optionally, the training module is further configured to determine, as the target data set, each sample data set having a recall greater than a preset recall threshold, according to a preset recall threshold, each sample data set corresponding to a recall of each class label; and calculating the comprehensive representation value of each category label according to the recall rate of each target data set corresponding to each category label and the recall weight value of each target data set.

Optionally, the training module is further configured to optimize model parameters of the evaluation index prediction model according to the comprehensive representation value of each sample dataset corresponding to each class label; and returning to execute the step of predicting the category of each sample data set according to the plurality of sample data corresponding to the plurality of candidate evaluation indexes and the given class label, wherein the step of predicting the category of each sample data set is executed on each sample data of each sample data set until the recall rate of each sample data set corresponding to each category label is larger than a preset optimization threshold value.

Optionally, the preset optimization threshold is 90%.

Optionally, the index obtaining module 602 is further configured to: and screening the multiple candidate evaluation indexes according to the multiple sample data of each sample data set corresponding to the multiple candidate evaluation indexes and the class label of each sample data set through the trained evaluation index prediction model, and determining at least one target evaluation index in the multiple candidate evaluation indexes.

Optionally, the training module is further configured to train a credit assessment model, which includes:

according to the multiple binning intervals, performing binning processing and WOE conversion on the data to be evaluated of the target object corresponding to each target evaluation index to obtain a WOE maximum value and a WOE minimum value of the target object corresponding to each target evaluation index; predicting the credit level of the target object according to the WOE maximum value and the WOE minimum value of the target object corresponding to each target evaluation index to obtain a credit prediction value of the target object; and obtaining a model loss function of the credit evaluation model according to the credit predicted value and the credit label value of the target object, and optimizing the credit evaluation model according to the model loss function.

Optionally, the credit evaluation module 604 is further configured to: performing data standardization processing according to each piece of to-be-evaluated data of each target evaluation index, an evaluation coefficient value of each target evaluation index and a target intercept value of each target object, and calculating a comprehensive evaluation value of the target object, wherein the evaluation coefficient value of each target evaluation index and the target intercept value are determined by using the trained credit evaluation model; according to the comprehensive evaluation value of the target object, executing the calculation of the default probability to obtain the default probability value of the target object; and determining the credit evaluation grade of the target object according to the default probability value of the target object.

Embodiments of the present application provide a computer storage medium storing computer program code that, when executed by a processor, causes the processor to perform the credit assessment method described in the embodiments of the present application.

An exemplary embodiment of the present application provides an electronic device, including: at least one processor; and a memory communicatively coupled to the at least one processor. The memory stores a computer program executable by the at least one processor for causing the electronic device to perform the credit assessment method according to the exemplary embodiments of the present application when executed by the at least one processor.

Referring to fig. 7, a block diagram of an electronic device 700 that may be a server or client of the present application, which is an example of a hardware device that may be applied to aspects of the present application, will now be described. Electronic devices are intended to represent various forms of digital electronic computer devices, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular telephones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be exemplary only, and are not meant to limit implementations of the application described and/or claimed herein.

As shown in fig. 7, the electronic device 700 includes a computing unit 701 that can perform various appropriate actions and processes according to a computer program stored in a Read Only Memory (ROM) 702 or a computer program loaded from a storage unit 708 into a Random Access Memory (RAM) 703. In the RAM 703, various programs and data required for the operation of the device 700 may also be stored. The computing unit 701, the ROM 702, and the RAM 703 are connected to each other through a bus 704. An input/output (I/O) interface 705 is also connected to bus 704.

Various components in the electronic device 700 are connected to the I/O interface 705, including: an input unit 706, an output unit 707, a storage unit 708, and a communication unit 709. The input unit 706 may be any type of device capable of inputting information to the electronic device 700, and the input unit 706 may receive input numeric or character information and generate key signal inputs related to user settings and/or function controls of the electronic device. The output unit 707 may be any type of device capable of presenting information and may include, but is not limited to, a display, speakers, video/audio output terminals, vibrators, and/or printers. Storage unit 708 may include, but is not limited to, magnetic disks, optical disks. The communication unit 709 allows the electronic device 700 to exchange information/data with other devices through computer networks, such as the internet, and/or various telecommunications networks, and may include, but is not limited to, modems, network cards, infrared communication devices, wireless communication transceivers and/or chipsets, such as bluetooth (TM) devices, wiFi devices, wiMax devices, cellular communication devices, and/or the like.

The computing unit 701 may be a variety of general and/or special purpose processing components having processing and computing capabilities. Some examples of computing unit 701 include, but are not limited to, a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), various specialized Artificial Intelligence (AI) computing chips, various computing units running machine learning model algorithms, a Digital Signal Processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 701 performs the various methods and processes described above. For example, in some embodiments, a credit assessment method as described above may be implemented as a computer software program tangibly embodied on a machine-readable medium, such as storage unit 708. In some embodiments, part or all of the computer program may be loaded and/or installed onto the electronic device 700 via the ROM 702 and/or the communication unit 709. In some embodiments, the computing unit 701 may be configured to perform the credit assessment method described above by any other suitable means (e.g., by means of firmware).

Program code for carrying out methods of the present application may be written in any combination of one or more programming languages. These program code may be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable credit assessment device such that the program code, when executed by the processor or controller, causes the functions/operations specified in the flowchart and/or block diagram to be implemented. The program code may execute entirely on the machine, partly on the machine, as a stand-alone software package, partly on the machine and partly on a remote machine or entirely on the remote machine or server.

In the context of this application, a machine-readable medium may be a tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. The machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

As used herein, the terms "machine-readable medium" and "computer-readable medium" refer to any computer program product, apparatus, and/or device (e.g., magnetic discs, optical disks, memory, programmable Logic Devices (PLDs)) used to provide machine instructions and/or data to a programmable processor, including a machine-readable medium that receives machine instructions as a machine-readable signal. The term "machine-readable signal" refers to any signal used to provide machine instructions and/or data to a programmable processor.

To provide for interaction with a user, the systems and techniques described here can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to a user; and a keyboard and pointing device (e.g., a mouse or trackball) by which a user can provide input to the computer. Other kinds of devices may also be used to provide for interaction with a user; for example, feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form, including acoustic input, speech input, or tactile input.

The systems and techniques described here can be implemented in a computing system that includes a background component (e.g., as a data service), or that includes a middleware component (e.g., an application service), or that includes a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such background, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local Area Networks (LANs), wide Area Networks (WANs), and the internet.

The computer system may include a client and a server. The client and the server are typically remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other.

It should be noted that, according to implementation requirements, each component/step described in the embodiments of the present application may be split into more components/steps, and two or more components/steps or part of operations of the components/steps may be combined into new components/steps, so as to achieve the purposes of the embodiments of the present application.

The above embodiments are only for illustrating the embodiments of the present application, but not for limiting the embodiments of the present application, and various changes and modifications can be made by one skilled in the relevant art without departing from the spirit and scope of the embodiments of the present application, so that all equivalent technical solutions also fall within the scope of the embodiments of the present application, and the scope of the embodiments of the present application should be defined by the claims.

Claims (10)

1. A method of credit assessment, comprising:

according to each target evaluation index predicted from each candidate evaluation index by the evaluation index prediction model, acquiring each piece of to-be-evaluated data of the target object corresponding to each target evaluation index;

and executing credit rating evaluation according to the data to be evaluated of the target object corresponding to each target evaluation index by using a credit evaluation model to obtain a credit rating evaluation result of the target object.

2. The method of claim 1, wherein the assessment index prediction model is trained by;

performing a classification prediction on each sample data of each sample data set based on a plurality of sample data, each sample data set corresponding to a plurality of candidate evaluation indicators, given class labels, a prediction class of each sample data set;

calculating the recall rate of each sample data set corresponding to each category label according to the category label and the prediction category of each sample data set;

calculating the comprehensive representation value of each category label according to the recall rate of each sample data set corresponding to each category label and the recall weight value of each sample data set;

and optimizing the evaluation index prediction model according to the comprehensive representation value of each sample data set corresponding to each class label to obtain a trained evaluation index prediction model.

3. The method of claim 2, wherein calculating the composite representation value for each category label based on the recall rate for each category label for each sample dataset, and the recall weight value for each sample dataset, comprises:

according to a preset recall threshold, determining each sample data set with the recall greater than the preset recall threshold as a target data set, wherein each sample data set corresponds to the recall of each category label;

and calculating the comprehensive representation value of each category label according to the recall rate of each target data set corresponding to each category label and the recall weight value of each target data set.

4. The method according to claim 2, wherein optimizing model parameters of the evaluation index prediction model according to the comprehensive representation value of each sample dataset corresponding to each class label to obtain a trained evaluation index prediction model comprises:

optimizing model parameters of the evaluation index prediction model according to comprehensive representation values of each sample data set corresponding to each class label;

returning to execute the step of executing classification prediction on each sample data of each sample data set according to the plurality of sample data corresponding to the plurality of candidate evaluation indexes and the given class labels until the recall rate of each sample data set corresponding to each class label is larger than a preset optimization threshold;

wherein the preset optimization threshold is 90%.

5. The method according to any one of claims 1 to 4, wherein each target evaluation index is obtained by:

and screening the multiple candidate evaluation indexes according to the multiple sample data of each sample data set corresponding to the multiple candidate evaluation indexes and the class label of each sample data set through the trained evaluation index prediction model, and determining at least one target evaluation index in the multiple candidate evaluation indexes.

6. The method of claim 1, wherein the credit assessment model is trained by:

according to the multiple binning intervals, performing binning processing and WOE conversion on the data to be evaluated of the target object corresponding to each target evaluation index to obtain a WOE maximum value and a WOE minimum value of the target object corresponding to each target evaluation index;

predicting the credit level of the target object according to the WOE maximum value and the WOE minimum value of the target object corresponding to each target evaluation index to obtain a credit prediction value of the target object;

and obtaining a model loss function of the credit evaluation model according to the credit predicted value and the credit label value of the target object, and optimizing the credit evaluation model according to the model loss function.

7. The method according to claim 1 or 6, wherein the performing the credit rating assessment according to each data to be assessed of the target object corresponding to each target assessment index to obtain the credit rating assessment result of the target object comprises:

performing data standardization processing according to each piece of to-be-evaluated data of each target evaluation index, an evaluation coefficient value of each target evaluation index and a target intercept value of each target object, and calculating a comprehensive evaluation value of the target object, wherein the evaluation coefficient value of each target evaluation index and the target intercept value are determined by using the trained credit evaluation model;

according to the comprehensive evaluation value of the target object, executing the calculation of the default probability to obtain the default probability value of the target object;

and determining the credit evaluation grade of the target object according to the default probability value of the target object.

8. A credit evaluation apparatus, comprising:

the index obtaining module is used for obtaining each piece of to-be-evaluated data of the target object corresponding to each target evaluation index according to each target evaluation index predicted by the evaluation index prediction model;

and the credit evaluation module is used for executing credit rating evaluation according to the data to be evaluated of the target object corresponding to each target evaluation index by using a credit evaluation model to obtain a credit rating evaluation result of the target object.

9. An electronic device, comprising:

a memory and a processor, wherein the memory has stored therein a computer program which, when executed by the processor, implements the credit assessment method of any one of claims 1 to 7.

10. A computer storage medium storing computer program code which, when executed by a processor, causes the processor to perform the credit assessment method of any of claims 1 to 7.