Methods, apparatuses, devices, and media for processing data

Abstract

The application discloses a method, device, equipment and medium for processing data, relates to the field of computers, and particularly relates to the field of artificial intelligence. A specific implementation scheme is as follows: input data is acquired; a data processing result is determined according to the input data and a preset model set; in response to the determination that the data processing result meets a preset check condition, the data processing result is checked to obtain output data; and the output data is output. This process can improve the data processing effect.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the field of computer, in particular to the field of artificial intelligence, and more particularly to a method, apparatus, device and medium for processing data. BACKGROUND

[0002] At present, various models have been widely applied to various fields to solve corresponding technical problems. For example, in a financial risk control system, a risk control model can be used to score the credit of a user.

[0003] In practice, it is found that the accuracy of the output data obtained by using a single model to process data is closely related to the performance of the model itself. If the model itself has some problems such as unstable output and low accuracy, it will lead to poor data processing effect. SUMMARY

[0004] The present disclosure provides a method, apparatus, device and medium for processing data.

[0005] According to an aspect of the present disclosure, a method for processing data is provided, comprising: obtaining input data; determining a data processing result according to the input data and a preset model set; in response to determining that the data processing result meets a preset verification condition, verifying the data processing result to obtain output data; and outputting the output data.

[0006] According to another aspect of the present disclosure, an apparatus for processing data is provided, comprising: a data obtaining unit configured to obtain input data; a data processing unit configured to determine a data processing result according to the input data and a preset model set; a data verification unit configured to, in response to determining that the data processing result meets a preset verification condition, verify the data processing result to obtain output data; and a data output unit configured to output the output data.

[0007] According to another aspect of the present disclosure, an electronic device for processing data is provided, comprising: one or more computing units; a storage unit for storing one or more programs; when the one or more programs are executed by the one or more computing units, the one or more computing units implement any one of the above methods for processing data.

[0008] According to another aspect of the present disclosure, a non-transitory computer readable storage medium storing computer instructions is provided, wherein the computer instructions are used to make a computer execute any one of the above methods for processing data.

[0009] According to another aspect of the present disclosure, a computer program product is provided, comprising a computer program which, when executed by a processor, implements any one of the above methods for processing data.

[0010] According to the technology of the present application, a method for processing data is provided, which can determine a data processing result by using at least one model in a model set and input data, and improve the reliability of the data processing result. Furthermore, the data processing result can be further checked to obtain output data and output the output data when the data processing result meets a checking condition. Through the checking process, more accurate output data can be obtained, thereby improving the data processing effect.

[0011] It should be understood that the content described in this part is not intended to identify key or important features of the embodiments of the present disclosure, nor to limit the scope of the present disclosure. Other features of the present disclosure will become apparent from the following description. BRIEF DESCRIPTION OF DRAWINGS

[0012] The accompanying drawings are used to better understand the present scheme, and do not constitute a limitation on the present disclosure. Among them:

[0013] Figure 1 is an exemplary system architecture diagram according to the first embodiment of the present disclosure;

[0014] Figure 2 is a method for processing data according to the second embodiment of the present disclosure;

[0015] Figure 3 is a data processing scenario diagram that can implement the embodiments of the present disclosure;

[0016] Figure 4 is a method for processing data according to the third embodiment of the present disclosure;

[0017] Figure 5 is a device for processing data according to the fourth embodiment of the present disclosure;

[0018] Figure 6 is a block diagram of an electronic device for implementing the method for processing data according to the embodiments of the present disclosure. DETAILED DESCRIPTION

[0019] Exemplary embodiments of the present disclosure are described below with reference to the accompanying drawings, including various details of the embodiments of the present disclosure to help understanding. They should be considered only as exemplary. Therefore, those of ordinary skill in the art should recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Also, in order to be clear and concise, descriptions of well-known functions and structures are omitted in the following description.

[0020] It should be noted that the embodiments in the present application and the features in the embodiments can be combined with each other without conflict. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0021] Figure 1 is an exemplary system architecture schematic diagram according to the first embodiment of the present disclosure, which shows an exemplary system architecture 100 to which the embodiments of the method for processing data or the apparatus for processing data of the present application can be applied.

[0022] As shown in Figure 1 , the system architecture 100 can include terminal devices 101, 102, 103, a network 104 and a server 105. The network 104 is a medium for providing a communication link between the terminal devices 101, 102, 103 and the server 105. The network 104 can include various connection types, such as wired, wireless communication links or optical fiber cables, etc.

[0023] A user can use the terminal devices 101, 102, 103 to interact with the server 105 through the network 104 to receive or send messages, etc. The terminal devices 101, 102, 103 can be electronic devices such as mobile phones, computers and tablets, etc. Input data in a specific application scenario can be obtained in the terminal devices 101, 102, 103, such as user data for applying for a loan in a financial risk control scenario, or user face data obtained in a face recognition scenario. In these examples, the user data for applying for a loan and the user face data are the input data in the embodiments of the present application.

[0024] The terminal devices 101, 102, 103 can be hardware or software. When the terminal devices 101, 102, 103 are hardware, they can be various electronic devices, including but not limited to televisions, smartphones, tablet computers, e-book readers, vehicle-mounted computers, laptop computers and desktop computers, etc. When the terminal devices 101, 102, 103 are software, they can be installed in the above-mentioned electronic devices. They can be implemented as multiple software or software modules (for example, to provide distributed services), or as a single software or software module. No specific limitation is made herein.

[0025] The server 105 can be a server providing various services, for example, input data transmitted by the terminal devices 101, 102, and 103 can be acquired through the network 104, the server 105 can store a model set, models in the model set can have a serial connection relationship or a parallel connection relationship. After the input data is acquired, the input data can be processed by the models in the model set to obtain a data processing result. Further, if the data processing result satisfies a preset check condition, it means that the data processing result has a certain deviation, at this time, the data processing result can be checked to obtain output data in a reasonable range. The server 105 can transmit the output data in the reasonable range obtained finally to the terminal devices 101, 102, and 103 through the network 104, so that the terminal devices 101, 102, and 103 output the output data. For example, in a financial risk control scenario, a user inputs user data for applying for a loan, and the terminal devices 101, 102, and 103 can output a credit score corresponding to the user data. Optionally, an application result corresponding to the credit score can also be output, and the application result can include credit granting or loan rejection.

[0026] It should be noted that the server 105 can be hardware or software. When the server 105 is hardware, it can be implemented as a distributed server cluster composed of multiple servers, or as a single server. When the server 105 is software, it can be implemented as multiple software or software modules (for example, to provide distributed services), or as a single software or software module. This is not limited specifically.

[0027] It should be noted that the method for processing data provided by the embodiments of the present application can be executed by the server 105 or the terminal devices 101, 102, and 103. Accordingly, the device for processing data can be arranged in the server 105 or the terminal devices 101, 102, and 103.

[0028] It should be understood that Figure 1 The number of terminal devices, networks, and servers in

[0029] With reference to Figure 2 , Figure 2 is a schematic diagram of a method for processing data according to a second embodiment of the present disclosure, which shows a flow 200 of one embodiment of the method for processing data according to the present application. The method for processing data of the present embodiment includes the following steps:

[0030] Step 201, acquiring input data.

[0031] In this embodiment, the execution subject (such as the server 105 or the terminal device 101, 102, or 103 described above) can obtain input data that needs to be processed in different application scenarios. The application scenarios can include, but are not limited to, financial risk control, target identification, image classification, and the like, and the present embodiment does not limit the same. Different data needs to be processed in different application scenarios. For example, the financial risk control scenario needs to process the user identity data of the loan applicant, the target identification scenario needs to process the object or face data to be identified, the image classification scenario needs to process the image data, and the like. Further, the input data can be data stored in advance in the execution subject or real-time acquired data.

[0032] In step 202, the data processing result is determined according to the input data and the preset model set.

[0033] In this embodiment, the preset model set includes at least one model, and the at least one model can be combined in a certain combination manner to process the data. The combination manner can be serial combination, parallel combination, or both serial combination and parallel combination, and the present embodiment does not limit the same. For the serial combination, the input data can be input into the first model to obtain the data output by the first model, and then the data output by the first model can be input into the next model until the last model outputs the final data, and the final data is the data processing result. For the parallel combination, the input data can be input into at least one model in parallel, and the data corresponding to the input data can be output by each model, and then the data can be integrated to obtain the data processing result. For the combination manner that has both serial combination and parallel combination, the models in the serial combination can be processed according to the processing manner corresponding to the serial combination, and the models in the parallel combination can be processed according to the processing manner corresponding to the parallel combination.

[0034] Optionally, in the process of processing the input data by using each model in the model set, a logic calculation function can also be introduced, and auxiliary data processing such as logic judgment, logic operation, and function can be set according to the requirements of the application scenario. The logic judgment can include size judgment logic such as greater than, less than, or equal to, the logic operation can include and, or, and not, and the function can include weighted average, weighted summation, and the like. When the and logic operation is used, if the data output by a certain model in the model set does not meet the required data condition, the threads of other models are terminated, which can reduce the consumption of additional resources. When the or logic operation is used, if the data output by a certain model in the model set meets the required data condition, the threads of the other models are terminated. That is, the running of the models is managed based on the logic operation condition in the model set.

[0035] In step 203, in response to determining that the data processing result meets the preset check condition, the data processing result is checked to obtain output data.

[0036] In this embodiment, the execution subject can set a data range corresponding to a normal data processing result in advance. After obtaining the data processing result, the data processing result can be compared with the preset data range. If the comparison result indicates that the data in the data processing result is beyond the preset data range, it is determined that the data processing result meets the preset check condition. Alternatively, the preset check condition can also be whether a check instruction manually triggered by a staff based on the data processing result is received. If the check instruction is received, it indicates that the data processing result meets the preset check condition.

[0037] Further, the checking can include correcting or covering the data processing result. The correction refers to making some adjustments on the basis of the data processing result, and taking the adjusted data processing result as the output data. The covering refers to discarding the data processing result and determining the output data again according to the input data.

[0038] In step 204, the output data is output.

[0039] In this embodiment, the execution subject can directly output the output data, or can obtain a corresponding judgment conclusion based on the output data in combination with an application scenario, and output the judgment conclusion. This embodiment does not limit this.

[0040] Continuing to refer to Figure 3 , Figure 3 is a data processing scene diagram that can implement the embodiments of the present disclosure, which shows a schematic diagram of one application scene of the method for processing data according to the present application. In Figure 3 the application scene, the above-mentioned method for processing data can be applied to the scene of financial risk control. For example Figure 3As shown, first, input data, i.e., user data 301, is obtained. The user data 301 can be data of a user who needs to apply for a loan, such as the age, city, income record, credit record, and the like of the user who needs to apply for a loan. Further, the model combiner 302 includes the model set described above, which can include at least an artificial intelligence model for credit scoring. The filtered user data 301 is input into the artificial intelligence model to obtain data output by the model. In addition, the model set can also include other non-artificial intelligence models for scoring, and the user data 301 is input into these other models to obtain data output by the models. The output data is integrated to obtain a data processing result. At this time, it can be judged based on the model insurance module 303 whether the data processing result satisfies a verification condition. For example, the data processing result indicates that the score of user A is 30, and the model insurance module 303 can calculate the difference between the score 30 of user A and a preset score. If the difference exceeds a certain threshold (such as 20), the data processing result is verified to obtain a final scoring result 304.

[0041] The method for processing data provided by the above embodiments of the present application can determine a data processing result by using at least one model in the model set and input data, thereby improving the reliability of the data processing result. Further, the data processing result can be further verified in a case where the data processing result satisfies a verification condition, to obtain output data and output the output data. Through the verification process, more accurate output data can be obtained, thereby improving the data processing effect.

[0042] Referring back to Figure 4 , Figure 4 is a schematic diagram of a method for processing data according to a third embodiment of the present disclosure, which shows a flow 400 of another embodiment of the method for processing data according to the present application. As shown in Figure 4 , the method for processing data of the present embodiment can include the following steps:

[0043] Step 401, obtaining input data.

[0044] In the present embodiment, the detailed description of step 401 is referred to the detailed description of step 201, which is not repeated here.

[0045] Step 402, obtaining at least one data filtering condition.

[0046] In the present embodiment, the data filtering condition refers to a data judgment rule obtained by combining a conventional threshold judgment, a basic logical operation, and / or a custom rule judgment. Data that meets the data judgment rule is retained, and data that does not meet the data judgment rule is discarded. Each data filtering condition can be a different granularity screening condition.

[0047] In step 403, the input data is filtered by using at least one data filtering condition according to a preset filtering order, and filtered input data is obtained.

[0048] In this embodiment, the preset filtering order is an order of using the data filtering conditions, and the data filtering is performed by using at least one data filtering condition according to the filtering order. Alternatively, the preset filtering order can be an order from coarse granularity to fine granularity, and the data filtering is performed by first performing coarse-granularity data filtering and then performing fine-granularity data filtering. Each data filtering condition can be a code segment independent of each other, and before each data filtering condition is accessed, data format uniformity of each data filtering condition can be performed, and then each data filtering condition after format uniformity is introduced. Alternatively, each data filtering condition can be introduced by using a linked list structure.

[0049] In some optional implementation manners of this embodiment, the following steps can also be performed: information of data filtered by each data filtering condition is counted; and the counted information is outputted.

[0050] In this implementation manner, in the data filtering process, the filtering counter and the performance counter in the filtering component can be used to count information of data filtered by each data filtering condition, and the counted information is outputted, so as to facilitate the execution subject to adjust the data filtering condition according to the information. The filtering counter is used to count total data quantity, data interception quantity, etc. of each data filtering condition, and the performance counter is used to count filtering time corresponding to each data filtering condition. For example, if a data filtering condition can filter 100% of data, after the information is outputted, the execution subject can delete the filtering condition; or if the filtering granularity of a data filtering condition is relatively coarse, after the information is outputted, the execution subject can preposition the data filtering condition, and the data filtering condition is executed first, and then other data filtering conditions are executed.

[0051] In step 404, a data processing result is determined according to the filtered input data and the model set.

[0052] In this embodiment, for detailed description of step 404, please refer to the detailed description of step 202, and the input data in step 202 is replaced by the filtered input data, which will not be repeated here.

[0053] In some optional implementations of the embodiment, the data processing result is determined according to the input data and the preset model set, including: for each model in the model set, converting the input data into a data format corresponding to the model to obtain model input data of the model; determining a model processing result corresponding to the model according to the model input data and the model; performing format conversion on the model processing results to obtain model processing results conforming to the target format; and integrating the model processing results conforming to the target format to obtain the data processing result.

[0054] In the implementation, the model set can include at least one model, for each model, the input data can be converted into model input data of the model through an input mapping layer corresponding to the model set, and then a model processing result of the model can be determined according to the model and the model input data of the model. Then, the model processing results of the models are converted through an output mapping layer corresponding to the model set to obtain model processing results conforming to the target format, and then the model processing results are integrated to obtain the data processing result. The integration can include logical calculation, average calculation, weighted score calculation and the like. This process can unify the input and output formats of the models in the model set, facilitating the access of different models.

[0055] In step 405, the verification model is determined according to the execution logic of the model in the model set.

[0056] In the embodiment, the verification model can have the same execution logic as the model in the model set, for example, the model in the model set can be used to score the risk of a user, and the verification model can also score the risk of the user. Alternatively, the verification model can determine a corresponding correction logic operation mode according to the execution logic of the model in the model set, as the execution logic of the verification model, which is not limited in the embodiment.

[0057] In some optional implementations of the embodiment, the verification model is determined according to the execution logic of the model in the model set, including: in response to determining that the model set includes a black box model, determining a previous version model of the black box model as the verification model.

[0058] In the embodiment, if the model set includes a black box model, the previous version model of the black box model can be determined as the verification model, so that when the version of the black box model is updated, the historical version of the black box model is used to improve the stability of the output data.

[0059] In step 406, in response to determining that the data processing result meets the preset verification condition, the output data is determined based on the verification model and the data processing result, or the output data is determined according to the data output by the verification model.

[0060] In this embodiment, if the data in the data processing result meets the first data range, the data processing result can be corrected by using the verification model to obtain the output data; if the data in the data processing result meets the second data range, the input data or the filtered input data can be input into the verification model to obtain the data output by the verification model, and the data output by the verification model is taken as the output data. Optionally, the first data range is a data range with a smaller deviation from the standard data range, and the second data range is a data range with a larger deviation from the standard data range. Optionally, in the data verification process, the data verification record of this time can be recorded to enable the staff to optimize the verification model based on the data verification record.

[0061] In some optional implementation manners of this embodiment, the output data is determined according to the data output by the verification model, including: determining the data output by the verification model based on the verification model and the input data; and determining the output data according to the data output by the verification model.

[0062] In this implementation manner, if the verification model is a model of a previous version of a black box model, the input data or the filtered input data can be input into the previous version of the black box model, and the data output by the previous version of the black box model is taken as the output data.

[0063] In step 407, the output data is output.

[0064] In this embodiment, the detailed description of step 407 can refer to the detailed description of step 204, which is not repeated here.

[0065] From Figure 4 It can be seen that, compared with the embodiment corresponding to Figure 2 The flow 400 of the method for processing data in this embodiment can further filter the input data in sequence according to at least one data filtering condition and a corresponding filtering sequence, can filter out data that is not needed in the current application scenario, and improves the data effectiveness in the data processing process. Moreover, information of the filtered data obtained by statistics can be output for the user to refer to and adjust the data filtering condition. In addition, when data verification is performed, the data processing result can be adjusted based on the verification model to take the corrected data as the output data, or the data output by the verification model can be directly taken as the output data, to meet various verification requirements. In addition, when the model set includes a black box model, the previous version model of the black box model can be determined as the verification model, so that the output result of the previous version model can be used to maintain stability when the version of the black box model is updated, and the output data is more stable and reliable. Moreover, the input and output of each model in the model set can be format-converted to ensure that the model set can adapt to various different models, and the scalability is stronger.

[0066] Further referring to Figure 5 , Figure 5 is a schematic diagram of an apparatus for processing data according to the fourth embodiment of the present disclosure, which provides one embodiment of an apparatus for processing data, the apparatus embodiment corresponding to the method embodiment shown in Figure 2 , and the apparatus can be specifically applied to various terminal devices or servers.

[0067] As shown in Figure 5 , the apparatus for processing data 500 of the present embodiment comprises a data acquisition unit 501, a data processing unit 502, a data verification unit 503, and a data output unit 504.

[0068] The data acquisition unit 501 is configured to acquire input data.

[0069] The data processing unit 502 is configured to determine a data processing result according to the input data and a preset model set.

[0070] The data verification unit 503 is configured to verify the data processing result to obtain output data in response to determining that the data processing result satisfies a preset verification condition.

[0071] The data output unit 504 is configured to output the output data.

[0072] In some optional implementations of the present embodiment, the data processing unit 502 is further configured to: acquire at least one data filtering condition; sequentially filter the input data according to the preset filtering order using the at least one data filtering condition to obtain filtered input data; and determine the data processing result according to the filtered input data and the model set.

[0073] In some optional implementations of the present embodiment, the apparatus further comprises an information statistics unit configured to statistically obtain information of data filtered by each data filtering condition; and an information output unit configured to output the statistically obtained information.

[0074] In some optional implementations of the present embodiment, the apparatus further comprises a model determination unit configured to determine a verification model according to an execution logic of a model in the model set.

[0075] In some optional implementations of the present embodiment, the data verification unit 503 is further configured to: determine the output data based on the verification model and the data processing result; or determine the output data according to data output by the verification model.

[0076] In some optional implementations of the present embodiment, the model determination unit is further configured to: in response to determining that the model set comprises a black-box model, determine a previous version model of the black-box model as the verification model.

[0077] In some optional implementation forms of the embodiment, the data checking unit 503 is further configured to determine the data output by the checking model based on the checking model and the input data, and determine the output data according to the data output by the checking model.

[0078] In some optional implementation forms of the embodiment, the data processing unit 502 is further configured to, for each model in the model set, convert the input data into a data format corresponding to the model to obtain model input data, determine a model processing result corresponding to the model according to the model input data and the model, perform format conversion on the model processing results to obtain model processing results in a target format, and integrate the model processing results in the target format to obtain the data processing result.

[0079] It should be understood that the units 501 to 504 described in the apparatus 500 for processing data respectively correspond to the respective steps in the method described with reference to Figure 2 Thus, the operations and features described above for the method for processing data also apply to the apparatus 500 and the units contained therein, which will not be repeated here.

[0080] According to embodiments of the present disclosure, the present disclosure further provides an electronic device, a readable storage medium and a computer program product.

[0081] Figure 6 A block diagram of an electronic device 600 is shown for implementing the method for processing data according to embodiments of the present disclosure. The electronic device is intended to represent various forms of digital computers, such as laptops, desktops, tablets, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular telephones, smart phones, wearable devices, and other similar computing devices. The components shown here, their connections and relationships, and their functions, are meant to be examples only, and are not intended to limit the implementations of the present disclosure described and / or claimed in this document.

[0082] As shown in Figure 6 The device 600 includes a computing unit 601, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 602 or a computer program loaded from a storage unit 608 into a random access memory (RAM) 603. Various programs and data required for the operation of the device 600 can also be stored in the RAM 603. The computing unit 601, the ROM 602 and the RAM 603 are connected to each other through a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0083] A number of components in the device 600 are connected to the I / O interface 605, including: an input unit 606, such as a keyboard, a mouse, etc.; an output unit 607, such as various types of displays, speakers, etc.; a storage unit 608, such as a magnetic disk, an optical disk, etc.; and a communication unit 609, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 609 allows the device 600 to exchange information / data with other devices through computer networks, such as the Internet, and / or various telecommunication networks.

[0084] The computing unit 601 can be various general and / or special purpose processing components with processing and computing capabilities. Some examples of the computing unit 601 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The computing unit 601 performs various methods and processes described above, such as the method for processing data. For example, in some embodiments, the method for processing data can be implemented as a computer software program, which is tangibly embodied in a machine-readable medium, such as the storage unit 608. In some embodiments, part or all of the computer program can be loaded and / or installed onto the device 600 via the ROM 602 and / or the communication unit 609. When the computer program is loaded onto the RAM 603 and executed by the computing unit 601, one or more steps of the method for processing data described above can be performed. Alternatively, in other embodiments, the computing unit 601 can be configured to perform the method for processing data by any other appropriate means, such as by means of firmware.

[0085] The various implementations of the systems and techniques described above can be realized in digital electronic circuitry, integrated circuitry, a field programmable gate array (FPGA), an application specific integrated circuit (ASIC), a system on a chip (SOC), a programmable logic device (PLD), a computer hardware, firmware, software, and / or combinations thereof. These various implementations can include implementation in one or more computer programs that are executable and / or interpretable on a programmable system including at least one programmable processor, which can be special or general purpose, coupled to receive data and instructions from, and to transmit data and instructions to, a storage system, at least one input device, and at least one output device.

[0086] Program code for carrying out methods of the present disclosure can be written in any combination of one or more programming languages. The program code can be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the program code, when executed by the processor or controller, produces the functions / operations specified in the flowcharts and / or block diagrams. The program code can be executed entirely on a machine, partially on a machine, partially on a machine as a standalone software package, or entirely on a remote machine or server.

[0087] In the context of the present disclosure, a machine-readable medium can be a tangible medium that contains or stores a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can 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 the machine-readable storage medium will include one or more lines of electrical connections, portable computer disks, hard disk drives, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or Flash memory), optical fibers, portable compact disc read-only memories (CD-ROMs), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0088] 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 the user and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the computer. Other kinds of devices can be used to provide for interaction with a user as well; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form, including acoustic, speech, or tactile input.

[0089] The systems and techniques described here can be implemented in a computing system that includes a back end component (e.g., as a data server), or that includes a middleware component (e.g., an application server), 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 back end, 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 a local area network (LAN), a wide area network (WAN), and the Internet.

[0090] The computer system can include clients and servers. This relationship can be

[0091] It should be understood that the procedures shown above are merely examples of the various forms that the procedures can take and that the procedures can be re-ordered, added to, or deleted from without departing from the scope of the present disclosure. For example, the steps recited in the present disclosure can be performed in parallel, in series, or in different orders, as long as the desired results of the technology disclosed in the present disclosure are achieved, and are not limited herein.

[0092] The specific embodiments described above are not intended to be limiting, and persons skilled in the art will appreciate that various modifications, combinations, sub-combinations and alternatives can be made to the specific embodiments without departing from the spirit and principles of the disclosure. Accordingly, the disclosure is not limited to the specific embodiments described above.

Claims

1. A method for processing data, comprising: Obtain user data used to apply for loans, including the user's credit history; According to the preset filtering order, the user data is filtered sequentially using at least one data filtering condition to obtain filtered user data. The data filtering condition is a data judgment rule obtained by combining conventional threshold judgment and / or basic logical operation judgment. Data that meets the data judgment rule is retained, and data that does not meet the data judgment rule is discarded. Each data filtering condition is a screening condition with different granularity. The preset filtering order is the order of filtering granularity from coarse to fine. Based on the filtered user data and a preset model set, a credit score result is determined. The preset model set includes two or more models, and at least a credit score model. Based on the logical operation results between the models in the preset model set, the running state of the model is controlled to reduce resource consumption. In response to determining that the credit score result meets the preset verification conditions, a verification model is determined according to the execution logic of the models in the model set, and the credit score result is verified through the verification model to obtain the final credit score; The final credit score is output to the terminal device, and the corresponding loan application result is generated.

2. The method according to claim 1, wherein, The method further includes: Statistics are compiled on the data filtered by each data filtering condition. Output the statistical information.

3. The method according to claim 1, wherein, The process of verifying the credit score results through a verification model to obtain the final credit score includes: Based on the verification model and the credit scoring results, the final credit score is determined; or The final credit score is determined based on the data output by the verification model.

4. The method according to claim 1, wherein, The step of determining the verification model based on the execution logic of the models in the model set includes: In response to determining that the model set includes a black-box model, the previous version of the black-box model is identified as the verification model.

5. The method according to claim 4, wherein, The process of verifying the credit score results through a verification model to obtain the final credit score includes: Based on the verification model and the user data, determine the data output by the verification model; The final credit score is determined based on the data output by the verification model.

6. The method according to claim 1, wherein, The step of determining the credit score result based on the user data and a preset model set includes: For each model in the model set, the user data is converted into the data format corresponding to that model to obtain the model input data; Based on the model input data and the model itself, determine the corresponding model processing result; The processing results of each model are converted to a different format to obtain processing results that conform to the target format. The credit score result is obtained by integrating the processing results of various models that conform to the target format.

7. An apparatus for processing data, comprising: The data acquisition unit is configured to acquire user data used for loan applications, the user data including the user's credit records; The data processing unit is configured to filter the user data sequentially using at least one data filtering condition according to a preset filtering order to obtain filtered user data. The data filtering condition is a data judgment rule obtained by combining conventional threshold judgments and / or basic logical operation judgments. Data that meets the data judgment rule is retained, while data that does not meet the rule is discarded. Each data filtering condition is a screening condition with different granularities, and the preset filtering order is an order from coarse to fine granularity. Based on the user data and a preset model set, a credit scoring result is determined. The preset model set includes two or more models, and at least a credit scoring model. Based on the logical operation results between the models in the preset model set, the operating state of the models is controlled to reduce resource consumption. The data verification unit is configured to determine a verification model based on the execution logic of the models in the model set in response to determining that the credit score result meets the preset verification conditions, and to verify the credit score result through the verification model to obtain the final credit score. The data output unit is configured to output the final credit score to the terminal device and generate the corresponding loan application result.

8. The apparatus according to claim 7, wherein, The device further includes: The information statistics unit is configured to statistically analyze the information of the data filtered by each data filtering condition. The information output unit is configured to output statistically obtained information.

9. The apparatus according to claim 7, wherein, The data verification unit is further configured to: Based on the verification model and the credit scoring results, the final credit score is determined; or The final credit score is determined based on the data output by the verification model.

10. An electronic device for performing a method for processing data, comprising: At least one computing unit; as well as A storage unit communicatively connected to the at least one computing unit; wherein, The storage unit stores instructions that can be executed by the at least one computing unit, which, when executed by the at least one computing unit, enables the at least one computing unit to perform the method of any one of claims 1-6.

11. A non-transitory computer-readable storage medium storing computer instructions, wherein, The computer instructions are used to cause the computer to perform the method according to any one of claims 1-6.

12. A computer program product comprising a computer program that, when executed by a processor, implements the method according to any one of claims 1-6.

Images