Debit and credit repayment capability prediction method and related device

Abstract

The invention provides a loan repayment capability prediction method and a related device, and relates to the technical field of data analysis. K neighbors of personal loan information are determined by using a distance value of a feature space matrix based on a loan application condition and a loan repayment condition, and the k neighbors are screened based on a Jaccard similarity coefficient; the method comprises the following steps: screening a loan application, constructing a priori knowledge matrix based on a screening result, training an initial prediction model by adopting the priori knowledge matrix to obtain a target prediction model, and predicting the newly obtained loan application by adopting the target model, so that an output result of the target prediction model is more reliable and higher in accuracy.

Description

Technical Field

[0001] The present invention relates to the technical field of data analysis, and in particular to a loan repayment capacity prediction method and related devices. Background Art

[0002] For lending institutions, predicting the repayment ability of loan applicants and deciding whether to approve the corresponding loan application is an important business to control loan risks. In the traditional loan review model, loan application materials are mainly reviewed manually, which requires high experience of reviewers. With the development of information technology, mining and utilizing hidden information in data has provided a new direction for predicting loan repayment ability.

[0003] The existing scheme for assessing the repayment ability of loan applicants is as follows:

[0004] (1) Obtain multi-dimensional personal data related to the loan, including real estate information, annual income information, etc.;

[0005] (2) preprocessing the feature information to be used, including converting the feature information into a feature vector, performing K-nearest neighbor averaging on the feature vector, and constructing a training set;

[0006] (3) Using a decision tree algorithm to train and generate a decision tree based on the training set;

[0007] (4) Obtain lender data to construct a test set, and use the trained decision tree to predict the lender repayment risk for the test set.

[0008] Traditional solutions usually create prediction models based on unsupervised learning methods. During the model creation process, unlabeled training data is used to train the model, and the reliability of the trained model is low. Summary of the invention

[0009] In view of this, an embodiment of the present invention provides a loan repayment capacity prediction method and related devices to achieve this.

[0010] To achieve the above objectives, the embodiments of the present invention provide the following technical solutions:

[0011] A method for predicting loan repayment capacity, comprising:

[0012] Obtaining personal loan information, including loan application conditions and loan repayment status;

[0013] Generate and store a feature space matrix matching the loan application conditions and a label space matrix matching the loan repayment conditions;

[0014] Determine k neighbors of the personal loan information based on the distance value of the feature space matrix, and screen the k neighbors based on the Jaccard similarity coefficient;

[0015] Construct a priori knowledge matrix based on the screening results;

[0016] The initial prediction model is trained using the prior knowledge matrix to obtain the target prediction model;

[0017] The target prediction model is used to predict the newly acquired loan applications.

[0018] Optionally, in the above loan repayment capacity prediction method, before using the prior knowledge matrix to train the initial prediction model, the method further includes:

[0019] Construct an initial prediction model, the objective function of which is:

[0020]

[0021] Said S is a matrix revealing the similarity of loan applications;

[0022] The W is a parameter of the prediction model;

[0023] The P is a label confidence matrix that reveals the loan repayment ability;

[0024] The p ij represents the confidence that the i-th loan application belongs to the j-th label;

[0025] The Φ is a feature mapping function that maps the original feature space X to a higher dimensional space;

[0026] The L s is the semi-positive definite graph Laplacian matrix based on the similarity matrix S;

[0027] The a, b, c, and λ are all trade-off parameters.

[0028] Optionally, in the above loan repayment capacity prediction method, the initial prediction model is trained using a priori knowledge matrix to obtain a target prediction model, including:

[0029] The objective function of the initial prediction model is solved based on the prior knowledge matrix by an alternating direction multiplier method until the objective function converges or reaches a maximum number of iterations, thereby obtaining a target prediction model.

[0030] Optionally, in the above loan repayment capacity prediction method, solving the objective function of the initial prediction model based on the prior knowledge matrix by an alternating direction multiplier method includes:

[0031] The values ​​of the fixed variables W and P are based on the formula Solve for the value of S;

[0032] The values ​​of the fixed variables S and W are based on the formula Solve for the value of P;

[0033] The values ​​of the fixed variables S and P are based on the formula Solve for the value of W.

[0034] Optionally, the above loan repayment capacity prediction method includes: when solving the value of S and the value of W, using the Lagrange multiplier method for solution;

[0035] When solving the value of W, the augmented Lagrangian method is used.

[0036] Optionally, in the above loan repayment capacity prediction method, the loan application conditions at least include: the loan applicant's work unit, education background, health status, loan amount and loan purpose;

[0037] The loan repayment status includes: the repayment status of the loan projects that the lender has already generated, and the repayment status includes repaying the loan on time, repaying the loan after the due date, not repaying the loan after the due date, and being unable to repay the overdue loan.

[0038] A loan repayment capacity prediction device, comprising:

[0039] An information collection unit, used to obtain personal loan information, wherein the personal loan information includes loan application conditions and loan repayment status;

[0040] An information processing unit, used to generate and store a feature space matrix matching the loan application conditions and a label space matrix matching the loan repayment conditions;

[0041] A screening unit, used for determining k neighbors of the personal loan information based on the distance value of the feature space matrix, and screening the k neighbors based on the Jaccard similarity coefficient;

[0042] A priori knowledge construction unit, used to construct a priori knowledge matrix based on the screening results;

[0043] A training unit, used to train the initial prediction model using a priori knowledge matrix to obtain a target prediction model;

[0044] The prediction unit is used to use the target prediction model to predict the newly obtained loan application.

[0045] A computer program product includes computer-readable instructions. When the computer-readable instructions are executed on an electronic device, the electronic device implements any one of the loan repayment capacity prediction methods described above.

[0046] An electronic device comprises at least one processor and a memory connected to the processor, wherein:

[0047] The memory is used to store computer programs;

[0048] The processor is used to execute the computer program so that the electronic device can implement any one of the loan repayment capacity prediction methods described above.

[0049] A computer storage medium, the storage medium carrying one or more computer programs, when the one or more computer programs are executed by an electronic device, can enable the electronic device to implement any one of the above loan repayment capacity prediction methods

[0050] Based on the above technical solution, the above solution provided by the embodiment of the present invention determines the k neighbors of personal loan information based on the distance value of the feature space matrix based on the loan application conditions and loan repayment status, and screens the k neighbors based on the Jaccard similarity coefficient, and then constructs a priori knowledge matrix based on the screening results, and finally uses the priori knowledge matrix to train the initial prediction model to obtain the target prediction model, and uses the target model to predict the newly acquired loan application, so that the output result of the target prediction model is more reliable and has a higher accuracy rate. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0052] Figure 1 A schematic diagram of a system framework for implementing a loan repayment capacity prediction method provided in this application;

[0053] Figure 2 A schematic diagram of the structure of a device for implementing a loan repayment ability prediction method provided by the present application;

[0054] Figure 3 A schematic diagram of the structure of a server for implementing a loan repayment capacity prediction method provided in this application;

[0055] Figure 4 A flowchart of a loan repayment capacity prediction method provided in an embodiment of the present application;

[0056] Figure 5 A schematic diagram of the construction process of the prior knowledge matrix provided in the embodiment of the present application;

[0057] Figure 6 A schematic diagram of the structure of a loan repayment ability prediction device provided in an embodiment of the present application;

[0058] Figure 7 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0059] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0060] First, the professional counting vocabulary used in this application is explained:

[0061] Supervised learning, unsupervised learning, and weakly supervised learning all infer classifiers based on a set of training data. The biggest difference between them is the different data sets they use.

[0062] Supervised learning learns from data sets with accurate and complete labels and has achieved good results in various applications. However, collecting or labeling a large amount of data with accurate and complete labels requires huge manpower and material resources. In addition, in the process of manual labeling, due to the different experiences of workers, inconsistent labeling and labeling errors may occur, making the preparation work for supervised learning full of obstacles.

[0063] In unsupervised learning, the machine uses unlabeled data and learns on its own without any supervision. The machine tries to cluster or associate unlabeled data based on similarities and differences without any prior training on the data. Unsupervised learning is more time-saving and labor-saving in data collection, but due to the complete loss of the guiding role of labels, the process of unsupervised learning is more complicated and the effect is poor.

[0064] Weakly supervised learning uses datasets with less precise annotations to build predictive models. Since it does not require accurate and complete training datasets, it is easier to obtain datasets and can be guided by labeled information during the training process. It combines the advantages of supervised learning and unsupervised learning, achieving a balance between preparation work and training results.

[0065] Partially labeled learning is an important branch of the weakly supervised learning framework. Each instance in the partially labeled dataset is associated with a set of candidate labels, of which only one is the true label. The goal of partially labeled learning is to learn a multi-class classifier from the above data with noisy label information, and to predict the true label of an unknown instance.

[0066] This application can be applied in the field of prediction models. Taking the bank lending scenario as an example, multiple application scenarios implemented in the product will be introduced below.

[0067] First, the application scenario of this application is introduced.

[0068] This application can be applied to, but is not limited to, applications with data processing functions or cloud services provided by cloud-side servers, etc., which are introduced below:

[0069] See also Figure 1 , Figure 1 A schematic diagram of a system architecture for implementing a loan repayment capacity prediction method is shown. The system may include a terminal 100 and a server 200. The server 200 may include one or more servers ( Figure 1 In the example, a server is included, and the server 200 can provide the method provided in the embodiment of the present application for one or more terminals.

[0070] Among them, an application for implementing the loan repayment ability prediction method can be installed on the terminal 100. The above application and web page can provide a human-computer interface. The terminal 100 can receive relevant parameters input by the user on the human-computer interface and send the above parameters to the server 200. The server 200 can obtain processing results based on the received parameters and return the processing results to the terminal 100.

[0071] It should be understood that in some optional implementations, the terminal 100 can also complete the action of obtaining the processing result based on the received parameters by itself without the cooperation of the server, and the embodiments of the present application are not limited to this.

[0072] Next describe Figure 1 The product form of the mid-terminal 100;

[0073] The terminal 100 in the embodiment of the present application can be a mobile phone, a tablet computer, a wearable device, a vehicle-mounted device, an augmented reality (AR) / virtual reality (VR) device, a laptop computer, an ultra-mobile personal computer (UMPC), a netbook, a personal digital assistant (PDA), etc., and the embodiment of the present application does not impose any restrictions on this.

[0074] Figure 2 An optional hardware structure diagram of the terminal 100 is shown.

[0075] refer to Figure 2 As shown, the terminal 100 may include a radio frequency unit 110, a memory 120, an input unit 130, a display unit 140, a camera 150 (optional), an audio circuit 160 (optional), a speaker 161 (optional), a microphone 162 (optional), an earphone jack 163 (optional), a processor 170, an external interface 180, a power supply 190 and other components. Those skilled in the art will appreciate that Figure 2 These are merely examples of terminals or multi-function devices and do not constitute limitations on the terminals or multi-function devices, which may include more or fewer components than those shown in the figures, or combinations of certain components, or different components.

[0076] The input unit 130 can be used to receive input digital or character information, and generate key signal input related to the user settings and function control of the portable multifunctional device. Specifically, the input unit 130 may include a touch screen 131 (optional) and / or other input devices 132. The touch screen 131 can collect the user's touch operations on or near it (such as the user's operation on or near the touch screen using any suitable object such as fingers, joints, stylus, etc.), and drive the corresponding connection device according to a pre-set program. The touch screen can detect the user's touch action on the touch screen, convert the touch action into a touch signal and send it to the processor 170, and can receive and execute the command sent by the processor 170; the touch signal at least includes the touch point coordinate information. The touch screen 131 can provide an input interface and an output interface between the terminal 100 and the user. In addition, the touch screen can be implemented using multiple types such as resistive, capacitive, infrared and surface acoustic wave. In addition to the touch screen 131, the input unit 130 can also include other input devices. Specifically, other input devices 132 may include, but are not limited to, one or more of a physical keyboard, function keys (such as a volume control key, a switch key, etc.), a trackball, a mouse, a joystick, and the like.

[0077] Among them, the input device 132 can receive input data and the like.

[0078] The display unit 140 may be used to display information input by the user or provided to the user, various menus of the terminal 100, interactive interfaces, file displays, and / or playback of any multimedia file. In the embodiment of the present application, the display unit 140 may be used to display a human-computer interaction interface, prediction results, and the like.

[0079] The memory 120 can be used to store instructions and data. The memory 120 can mainly include an instruction storage area and a data storage area. The data storage area can store various data, such as multimedia files, texts, etc.; the instruction storage area can store software units such as operating systems, applications, instructions required for at least one function, or their subsets and extensions. It can also include a non-volatile random access memory; provide the processor 170 with hardware, software and data resources including management of computing and processing equipment, and support control software and applications. It is also used for the storage of multimedia files, and the storage of running programs and applications.

[0080] The processor 170 is the control center of the terminal 100. It uses various interfaces and lines to connect various parts of the entire terminal 100. By running or executing instructions stored in the memory 120 and calling data stored in the memory 120, it executes various functions of the terminal 100 and processes data, thereby controlling the terminal device as a whole. Optionally, the processor 170 may include one or more processing units; preferably, the processor 170 may integrate an application processor and a modem processor, wherein the application processor mainly processes the operating system, user interface and application program, and the modem processor mainly processes wireless communication. It is understandable that the above-mentioned modem processor may not be integrated into the processor 170. In some embodiments, the processor and the memory may be implemented on a single chip, and in some embodiments, they may also be implemented separately on separate chips. The processor 170 may also be used to generate corresponding operation control signals, send them to corresponding components of the computing and processing device, read and process data in the software, especially read and process data and programs in the memory 120, so that each functional module therein performs corresponding functions, thereby controlling the corresponding components to act according to the requirements of the instructions.

[0081] Among them, the memory 120 can be used to store software codes related to executing the loan repayment ability prediction method, the processor 170 can execute the steps of the loan repayment ability prediction method, and can also schedule other units (such as the above-mentioned input unit 130 and display unit 140) to achieve corresponding functions.

[0082] The radio frequency unit 110 (optional) can be used for receiving and sending information or receiving and sending signals during a call, for example, after receiving the downlink information of the base station, it is sent to the processor 170 for processing; in addition, the designed uplink data is sent to the base station. Generally, the RF circuit includes but is not limited to an antenna, at least one amplifier, a transceiver, a coupler, a low noise amplifier (Low Noise Amplifier, LNA), a duplexer, etc. In addition, the radio frequency unit 110 can also communicate with network devices and other devices through wireless communication. The wireless communication can use any communication standard or protocol, including but not limited to Global System of Mobile communication (Global System of Mobile communication, GSM), General Packet Radio Service (General Packet Radio Service, GPRS), Code Division Multiple Access (Code Division Multiple Access, CDMA), Wideband Code Division Multiple Access (Wideband Code Division Multiple Access, WCDMA), Long Term Evolution (Long Term Evolution, LTE), email, Short Messaging Service (SMS), etc.

[0083] In this embodiment of the present application, the RF unit 110 can send data to the server 200 and receive processing results sent by the server 200.

[0084] It should be understood that the radio frequency unit 110 is optional and can be replaced by other communication interfaces, such as a network port.

[0085] The terminal 100 also includes a power supply 190 (such as a battery) for supplying power to various components. Preferably, the power supply can be logically connected to the processor 170 through a power management system, so that the power management system can manage functions such as charging, discharging, and power consumption.

[0086] The terminal 100 further includes an external interface 180 , which may be a standard Micro USB interface or a multi-pin connector, and may be used to connect the terminal 100 to communicate with other devices, or to connect a charger to charge the terminal 100 .

[0087] Although not shown, the terminal 100 may also include a flashlight, a wireless fidelity (WiFi) module, a Bluetooth module, sensors with different functions, etc., which are not described in detail here. Some or all of the methods described below may be applied in the following embodiments. Figure 2 In the terminal 100 shown.

[0088] Next describe Figure 1 The product form of the server 200;

[0089] Figure 3 A structural diagram of a server 200 is provided, such as Figure 3 As shown, the server 200 includes a bus 201, a processor 202, a communication interface 203, and a memory 204. The processor 202, the memory 204, and the communication interface 203 communicate with each other via the bus 201.

[0090] The bus 201 may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 3 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0091] The processor 202 may be any one or more of a central processing unit (CPU), a graphics processing unit (GPU), a microprocessor (MP), or a digital signal processor (DSP).

[0092] The memory 204 may include a volatile memory, such as a random access memory (RAM). The memory 204 may also include a non-volatile memory, such as a read-only memory (ROM), a flash memory, a hard disk drive (HDD), or a solid state drive (SSD).

[0093] The memory 204 may be used to store software codes related to the loan repayment capacity prediction method, and the processor 202 may execute the steps of the loan repayment capacity prediction method of the chip, and may also schedule other units to implement corresponding functions.

[0094] It should be understood that the above-mentioned terminal 100 and server 200 can be centralized or distributed devices, and the processors in the above-mentioned terminal 100 and server 200 (such as processor 170 and processor 202) can be hardware circuits (such as application specific integrated circuit (ASIC), field-programmable gate array (FPGA), general-purpose processor, digital signal processor (DSP), microprocessor or microcontroller, etc.), or a combination of these hardware circuits. For example, the processor can be a hardware system with an instruction execution function, such as a CPU, DSP, etc., or a hardware system without an instruction execution function, such as an ASIC, FPGA, etc., or a combination of the above-mentioned hardware systems without an instruction execution function and hardware systems with an instruction execution function.

[0095] In order to solve the above problems, the embodiment of the present application provides a loan repayment ability prediction method. The loan repayment ability prediction method of the embodiment of the present application is described in detail below with reference to the accompanying drawings.

[0096] Figure 4 A flowchart of a loan repayment capacity prediction method provided in an embodiment of the present application is shown as follows: Figure 4 As shown, Figure 5 The figure is a schematic diagram of the construction process of the prior knowledge matrix. A loan repayment capacity prediction method provided in an embodiment of the present application may include steps 401 to 406. These steps are described in detail below.

[0097] See also Figure 4 The loan repayment capacity prediction method disclosed in the embodiment of the present application may include:

[0098] Step S401: Obtain personal loan information, which includes loan application conditions and loan repayment status.

[0099] First of all, it should be noted that the user information (including but not limited to user personal loan information) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant laws, regulations and standards of relevant countries and regions.

[0100] In this solution, the first step is to collect partial labeling ("partial labeling" is a key term, which comes from a branch of machine learning - weakly supervised learning) personal loan information. The personal loan information may include loan application conditions and loan repayment status, where the loan application conditions may include work unit, education, health status, loan amount, loan purpose and other loan application conditions. The loan repayment status includes: the repayment status of the loan project that the lender has generated, and the repayment status includes repayment on time, repayment after overdue, overdue and unpaid loans, and inability to repay overdue loans. This solution is based on the partial labeling learning method to collect personal loan information, so the environment for collecting personal information is more relaxed and simple.

[0101] After collecting personal loan information, it is necessary to organize the personal loan information, unify the attributes of the collected personal loan information, merge the data with similar loan application conditions, and conduct quantitative statistics on the corresponding loan repayments.

[0102] Step S402: Generate and store a feature space matrix matching the loan application conditions and a label space matrix matching the loan repayment situation.

[0103] See also Figure 5 In this step, after obtaining personal loan information, the personal loan information is converted into a logical structure and stored in the personal loan information database, and the collected personal loan information is used as a training set to train the prediction model. Specifically, in this process, the loan application conditions are used as feature domains and converted into feature space matrix X, and the loan repayment situation is used as a label domain and converted into a label space matrix Y.

[0104] The feature space matrix X and the label space matrix Y are taken as examples in the training set. The candidate labels of each instance in the training set belong to one or more of an element in the label space matrix (loan repayment status, such as repaying the loan on time, repaying the loan after the due date, not repaying the loan after the due date, being unable to repay the overdue loan, etc.), but only one of these space labels is a true label.

[0105] Step S403: determining k neighbors of the personal loan information based on the distance value of the feature space matrix, and screening the k neighbors based on the Jaccard similarity coefficient.

[0106] See also Figure 5 , the first step in building the prior knowledge matrix: building a directed weighted graph model on the training set where V is the vertex set consisting of instance loan applications, ε is the edge set representing the top k nearest neighbor relationships between instances, is the prior knowledge matrix. In this weighted graph model, the prior knowledge matrix is ​​an unknown quantity and needs to be constructed according to subsequent steps.

[0107] See also Figure 5 , the second step of constructing the prior knowledge matrix: determine the first k neighboring instances according to the distance in the instance feature space, and screen the k neighbors according to the Jaccard similarity coefficient in the label space (instances with completely different candidate labels must have different real labels), and then further construct the prior knowledge matrix S based on the screening results in step S404.

[0108] When processing loan application data, existing solutions usually rely on loan application conditions in feature space for classification. Loan application conditions may include work unit, education, health status, loan amount, loan purpose, etc. These conditions or attributes constitute the feature vector of loan application conditions. By using these feature vectors, the k neighbors that are most similar to the loan application conditions to be classified are found by calculating the distance (such as Euclidean distance, Manhattan distance, etc.), that is, k nearest neighbors. However, after research, the applicant found that relying solely on the features in the loan application conditions to select k nearest neighbors may not be sufficient to ensure the accuracy of classification. Because in actual situations, there may be instances with similar feature vectors but completely different loan repayment situations, which may be caused by other factors that have not been considered (such as sudden events of the borrower, changes in the market environment, etc.).

[0109] In order to improve the accuracy of classification, the present application can also combine the loan repayment situation in the tag space to assist in classification. By combining the loan repayment situation in the tag space, the k nearest neighbors selected in the feature space can be further screened or verified, thereby improving the accuracy of classification.

[0110] Specifically, the Jaccard similarity coefficient can be used to filter k nearest neighbors in the label space. The Jaccard similarity coefficient is an indicator to measure the similarity between two sets. It is equal to the size of the intersection of two sets (each set includes loan application conditions and loan repayment status) divided by the size of the union, that is, In this scenario, the tag of each loan application (i.e., loan repayment status) can be regarded as a set (actually a multi-valued set, including one of multiple elements such as "repaying the loan on time, repaying the loan after the due date, not repaying the loan after the due date, and being unable to repay the overdue loan, because in this context each loan application will eventually have a clear repayment result). Then, the Jaccard similarity coefficient between the loan application conditions to be classified and the k nearest neighbors in the tag space is calculated, and those neighbors that are not similar to the loan application conditions to be classified in terms of tags (i.e., different repayment status) are filtered out.

[0111] This filtering method ensures that two loan applications that do not share any common candidate tags (i.e., different repayment situations) belong to different categories. In other words, if two loan applications have the same loan application conditions but different candidate tags (i.e., different loan repayment situations), then the predicted actual loan repayment situations of the two loan applications can be considered different. The method combining feature space and tag space information can more comprehensively consider various factors of loan applications, thereby more effectively improving the accuracy of loan application classification.

[0112] Step S404: constructing a priori knowledge matrix based on the screening results, wherein each element in the priori knowledge matrix includes corresponding loan application conditions and loan repayment status marks.

[0113] See also Figure 5 , the third step of constructing the prior knowledge matrix: filter the results to construct the prior knowledge matrix

[0114] Step S405: Use the prior knowledge matrix to train the initial prediction model to obtain the target prediction model.

[0115] After determining the prior knowledge matrix, the constructed initial prediction model is iteratively trained using the prior knowledge matrix to obtain a target prediction model. The input of the target prediction model is the loan application conditions, and the output is the loan repayment situation with the highest confidence. The loan repayment situation is used as the repayment ability prediction result of the loan application, and whether the loan service can be provided to the loan application is confirmed based on the prediction result.

[0116] Step S406: using the target prediction model to predict the newly acquired loan application.

[0117] In this step, after obtaining the loan application submitted by the user, the user's loan application conditions are extracted from the loan application, and the loan application conditions are used as input data of the target prediction model. The model will use the objective function to process the input data to obtain the confidence levels corresponding to each loan repayment situation that matches the input data, and output the maximum confidence level and its corresponding loan repayment situation as the prediction result.

[0118] It can be seen from the above scheme that the present application proposes a method for predicting loan repayment ability. The method determines the k neighbors of personal loan information based on the distance value of the feature space matrix based on the loan application conditions and loan repayment status, and screens the k neighbors based on the Jaccard similarity coefficient. Then, a priori knowledge matrix is ​​constructed based on the screening results. Finally, the initial prediction model is trained using the priori knowledge matrix to obtain a target prediction model. The target model is used to predict the newly acquired loan application, so that the output result of the target prediction model is more reliable and has a higher accuracy rate.

[0119] In the technical solution disclosed in this embodiment, before using the prior knowledge matrix to train the initial prediction model, it is necessary to pre-build the initial prediction model. The built-in function of the initial prediction model can be set according to user needs to build the initial prediction model. The objective function of the initial prediction model is:

[0120]

[0121] Said S is a matrix revealing the similarity of loan applications;

[0122] The W is a parameter of the prediction model;

[0123] The P is a label confidence matrix that reveals the loan repayment ability;

[0124] The p ij represents the confidence that the i-th loan application belongs to the j-th label;

[0125] The Φ is a feature mapping function that maps the original feature space X to a higher dimensional space;

[0126] The L s is the semi-positive definite graph Laplacian matrix based on the similarity matrix S;

[0127] The a, b, c, and λ are all weighted parameters;

[0128] In the above formula, the subscript F is a symbol mark and does not participate in the calculation process.

[0129] The process of training the initial prediction model using the prior knowledge matrix and solving the objective function, specifically, the process of training the initial prediction model using the prior knowledge matrix to obtain the target prediction model, specifically includes: solving the objective function of the initial prediction model based on the prior knowledge matrix by the alternating direction multiplier method until the objective function converges or reaches the maximum number of iterations, and obtaining the target prediction model. That is, optimizing each variable in each objective function in turn, when optimizing a certain variable, it is necessary to treat other variables as constants until the objective function converges or reaches the maximum number of iterations, and finally obtaining the target prediction model,

[0130] Specifically, taking the above objective function For example, the optimization process of variables S, W, and P may include:

[0131] The values ​​of the fixed variables W and P are based on the formula Solve the value of S. The process of solving S is a standard quadratic programming problem. Therefore, the Lagrange multiplier method can be used to solve it.

[0132] The values ​​of the fixed variables S and W are based on the formula Solving the value of P, wherein the process of solving P is a nonlinear programming problem and can be solved by the augmented Lagrangian method;

[0133] The values ​​of the fixed variables S and P are based on the formula Solve the value of W. The process of solving W is a standard quadratic programming problem. Therefore, the Lagrange multiplier method can be used to solve it.

[0134] Substitute the final solved W, S and P into the objective function The target prediction model can be obtained.

[0135] This embodiment discloses a loan repayment capacity prediction device. For the specific working contents of each unit in the device, please refer to the contents of the above method embodiment.

[0136] The following is a description of a loan repayment capacity prediction device provided by an embodiment of the present invention. The loan repayment capacity prediction device described below and the loan repayment capacity prediction method described above can be referenced to each other.

[0137] See also Figure 6 , the loan repayment capacity prediction device may include:

[0138] The information collection unit 10 is used to obtain personal loan information, wherein the personal loan information includes loan application conditions and loan repayment status;

[0139] An information processing unit 20, used to generate and store a feature space matrix matching the loan application conditions and a label space matrix matching the loan repayment conditions;

[0140] A screening unit 30, configured to determine k neighbors of the personal loan information based on the distance value of the feature space matrix, and screen the k neighbors based on the Jaccard similarity coefficient;

[0141] A priori knowledge construction unit 40, used to construct a priori knowledge matrix based on the screening results;

[0142] A training unit 50 is used to train the initial prediction model using a priori knowledge matrix to obtain a target prediction model;

[0143] The prediction unit 60 is used to use the target prediction model to predict the newly obtained loan application.

[0144] The present application also provides an electronic device in an embodiment. Figure 7As shown, it shows a schematic diagram of the structure of an electronic device suitable for implementing the embodiments of the present application. The electronic device in the embodiments of the present application may include but is not limited to fixed terminals such as mobile phones, notebook computers, PDAs (personal digital assistants), PADs (tablet computers), desktop computers, etc. Figure 7 The electronic device shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0145] like Figure 7 As shown, the electronic device may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 601, which can perform various appropriate actions and processes for implementing the loan repayment capacity prediction method according to a program stored in a read-only memory (ROM) 602 or a program loaded from a storage device 608 to a random access memory (RAM) 603. When the electronic device is powered on, various programs and data required for the operation of the electronic device are also stored in the RAM 603. The processing device 601, the ROM 602, and the RAM 603 are connected to each other via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0146] Typically, the following devices may be connected to the I / O interface 605: an input device 606 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 607 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 608 including, for example, a memory card, a hard disk, etc.; and a communication device 609. The communication device 609 may allow the electronic device to communicate with other devices wirelessly or by wire to exchange data. Although Figure 7 An electronic device having various devices is shown, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead.

[0147] An embodiment of the present application also provides a computer program product, including computer-readable instructions. When the computer-readable instructions are executed on an electronic device, the electronic device implements any one of the loan repayment capacity prediction methods provided in the embodiments of the present application.

[0148] A computer-readable storage medium is also provided in an embodiment of the present application. The storage medium carries one or more computer programs. When the one or more computer programs are executed by an electronic device, the electronic device can implement any one of the loan repayment capacity prediction methods provided in the embodiments of the present application.

[0149] For the convenience of description, the above system is described as being divided into various modules according to their functions. Of course, when implementing the present invention, the functions of each module can be implemented in the same or multiple software and / or hardware.

[0150] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can refer to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system or system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can refer to the partial description of the method embodiment. The system and system embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without creative work.

[0151] Professionals may further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in the above description according to function. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians may use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0152] The steps of the method or algorithm described in conjunction with the embodiments disclosed herein may be implemented directly using hardware, a software module executed by a processor, or a combination of the two. The software module may be placed in a random access memory (RAM), a memory, a read-only memory (ROM), an electrically programmable ROM, an electrically erasable programmable ROM, a register, a hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.

[0153] It should also be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0154] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for predicting loan repayment ability, characterized in that: include: Obtaining personal loan information, including loan application conditions and loan repayment status; Generate and store a feature space matrix matching the loan application conditions and a label space matrix matching the loan repayment conditions; Determine k neighbors of the personal loan information based on the distance value of the feature space matrix, and screen the k neighbors based on the Jaccard similarity coefficient; Construct a priori knowledge matrix based on the screening results; The initial prediction model is trained using the prior knowledge matrix to obtain the target prediction model; The target prediction model is used to predict the newly acquired loan applications.

2. The loan repayment ability prediction method according to claim 1, characterized in that: Before using the prior knowledge matrix to train the initial prediction model, it also includes: Construct an initial prediction model, the objective function of which is: Said S is a matrix revealing the similarity of loan applications; The W is a parameter of the prediction model; The P is a label confidence matrix that reveals the loan repayment ability; The p ij represents the confidence that the i-th loan application belongs to the j-th label; The Φ is a feature mapping function that maps the original feature space X to a higher dimensional space; The L S is the semi-positive definite graph Laplacian matrix based on the similarity matrix S; The a, b, c, and λ are all trade-off parameters.

3. The loan repayment ability prediction method according to claim 2, characterized in that: The initial prediction model is trained using the prior knowledge matrix to obtain the target prediction model, including: The objective function of the initial prediction model is solved based on the prior knowledge matrix by an alternating direction multiplier method until the objective function converges or reaches a maximum number of iterations, thereby obtaining a target prediction model.

4. The loan repayment ability prediction method according to claim 2, characterized in that: Solving the objective function of the initial prediction model based on the prior knowledge matrix by an alternating direction multiplier method includes: The values ​​of the fixed variables W and P are based on the formula Solve for the value of S; The values ​​of the fixed variables S and W are based on the formula Solve for the value of P; The values ​​of the fixed variables S and P are based on the formula Solve for the value of W.

5. The loan repayment ability prediction method according to claim 4, characterized in that: include: When solving for the value of S and the value of W, the Lagrange multiplier method is used for solution; When solving the value of W, the augmented Lagrangian method is used.

6. The loan repayment ability prediction method according to claim 1, characterized in that: The loan application conditions shall at least include: the applicant's work unit, education background, health status, loan amount and loan purpose; The loan repayment status includes: the repayment status of the loan projects that the lender has already generated, and the repayment status includes repaying the loan on time, repaying the loan after the due date, not repaying the loan after the due date, and being unable to repay the overdue loan.

7. A loan repayment ability prediction device, characterized in that: include: An information collection unit, used to obtain personal loan information, wherein the personal loan information includes loan application conditions and loan repayment status; An information processing unit, used to generate and store a feature space matrix matching the loan application conditions and a label space matrix matching the loan repayment conditions; A screening unit, used for determining k neighbors of the personal loan information based on the distance value of the feature space matrix, and screening the k neighbors based on the Jaccard similarity coefficient; A priori knowledge construction unit, used to construct a priori knowledge matrix based on the screening results; A training unit, used to train the initial prediction model using a priori knowledge matrix to obtain a target prediction model; The prediction unit is used to use the target prediction model to predict the newly obtained loan application.

8. A computer program product, characterized in that It includes computer-readable instructions, and when the computer-readable instructions are executed on an electronic device, the electronic device implements the loan repayment capacity prediction method as described in any one of claims 1 to 6.

9. An electronic device, characterized in that: The method comprises at least one processor and a memory connected to the processor, wherein: The memory is used to store computer programs; The processor is used to execute the computer program so that the electronic device can implement the loan repayment ability prediction method as described in any one of claims 1 to 6.

10. A computer storage medium, characterized in that: The storage medium carries one or more computer programs, and when the one or more computer programs are executed by an electronic device, the electronic device can implement the loan repayment capacity prediction method as described in any one of claims 1 to 6.

Images