Data processing method, device and equipment

Abstract

The embodiment of the invention provides a data processing method, device and equipment, and relates to the field of artificial intelligence. The method comprises the following steps: acquiring multi-dimensional associated information of a target loan enterprise user, wherein the associated information of each dimension comprises information corresponding to a plurality of time periods of the dimension; constructing a multi-dimensional time sequence matrix based on the multi-dimensional association information, wherein matrix data corresponding to each time period in the multi-dimensional time sequence matrix comprises a matrix result of the multi-dimensional association information corresponding to the time period; determining the similarity between the target loan enterprise user and the historical loan enterprise user based on the multi-dimensional time sequence matrix and a historical multi-dimensional time sequence matrix of the historical loan enterprise user; wherein the historical loan enterprise user has a historical credit score; and determining the credit score of the target loan enterprise user according to the similarity and the historical credit score of the historical loan enterprise user. According to the method, the accuracy of the analysis conclusion of the credit condition of the target loan enterprise user is improved.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence, and in particular to a data processing method, device and equipment. Background Art

[0002] In the credit market, financial institutions measure the development and risk resistance of enterprises based on their credit status and formulate credit strategies to make the credit structure of financial institutions more reasonable.

[0003] Related technologies determine the credit status of an enterprise by analyzing a large amount of sound financial data of the enterprise. The data analysis model used can often only analyze and process data in a single dimension related to the enterprise's finances.

[0004] For many small, medium and micro enterprises, they have a short credit history and little accumulated financial data. Therefore, when using relevant technologies to analyze the credit status of such small, medium and micro enterprises, only a small amount of corporate financial data can be used to draw conclusions about the credit status of the enterprises. The relevant technologies for analyzing the credit status of enterprises based on single-dimensional data have the technical problem of inaccurate conclusions about the credit status of enterprises, especially small, medium and micro enterprises, which in turn affects the credit strategies of financial institutions and makes the credit structure of financial institutions unreasonable. Summary of the invention

[0005] The present application provides a data processing method, device and equipment to solve the above technical problems.

[0006] In a first aspect, the present application provides a data processing method, comprising:

[0007] Acquire multi-dimensional association information of target loan enterprise users, where each dimension of association information includes information corresponding to multiple time periods of the dimension;

[0008] Constructing a multi-dimensional time series matrix based on the multi-dimensional association information, wherein the matrix data corresponding to each time period in the multi-dimensional time series matrix includes the matrixed result of the multi-dimensional association information corresponding to the time period;

[0009] Determine the similarity between the target loan enterprise user and the historical loan enterprise user based on the multi-dimensional time series matrix and the historical multi-dimensional time series matrix of the historical loan enterprise user; wherein the historical loan enterprise user has a historical credit score;

[0010] The credit score of the target loan enterprise user is determined according to the similarity and the historical credit score of the historical loan enterprise user.

[0011] In a second aspect, the present application provides a data processing device, including:

[0012] An acquisition module is used to acquire multi-dimensional association information of target loan enterprise users, where each dimension of association information includes information corresponding to multiple time periods of the dimension;

[0013] A construction module, configured to construct a multi-dimensional time series matrix based on the multi-dimensional association information, wherein the matrix data corresponding to each time period in the multi-dimensional time series matrix includes a matrixed result of the multi-dimensional association information corresponding to the time period;

[0014] A first determination module is used to determine the similarity between the target loan enterprise user and the historical loan enterprise user based on the multi-dimensional time series matrix and the historical multi-dimensional time series matrix of the historical loan enterprise user; wherein the historical loan enterprise user has a historical credit score;

[0015] The second determination module is used to determine the credit score of the target loan enterprise user according to the similarity and the historical credit score of the historical loan enterprise user.

[0016] In a third aspect, the present application provides an electronic device, comprising: a processor, and a memory communicatively connected to the processor;

[0017] The memory stores computer-executable instructions;

[0018] The processor executes the computer-executable instructions stored in the memory to implement the first aspect and / or various possible implementations of the first aspect.

[0019] In a fourth aspect, the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer-executable instructions, and when the computer-executable instructions are executed by a processor, they are used to implement the first aspect and / or various possible implementations of the first aspect.

[0020] In a fifth aspect, the present application provides a computer program product, including a computer program, which, when executed by a processor, implements the first aspect and / or various possible implementations of the first aspect.

[0021] The data processing method, device and equipment provided by the present application construct a multi-dimensional time series matrix based on the multi-dimensional correlation information of multiple time periods of the target loan enterprise user, and determine the credit score of the target loan enterprise user by analyzing the similarity between the multi-dimensional time series matrix of the target loan enterprise user and the historical multi-dimensional time series matrix of the historical loan enterprise user, and the historical credit score of the historical loan enterprise user. The multi-dimensional time series matrix can be used to analyze the credit status of the target loan enterprise user from multiple dimensions, and the multi-dimensional time series matrix of the target loan enterprise user can be compared and analyzed with the historical multi-dimensional time series matrix of the historical loan enterprise user, thereby improving the accuracy of the analysis conclusion of the credit status of the target loan enterprise user. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0023] Figure 1 A flowchart of a data processing method provided in an embodiment of the present application;

[0024] Figure 2 A schematic diagram of the structure of a data processing device provided in an embodiment of the present application;

[0025] Figure 3 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application.

[0026] The above drawings have shown clear embodiments of the present application, which will be described in more detail later. These drawings and text descriptions are not intended to limit the scope of the present application in any way, but to illustrate the concept of the present application to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0027] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application. Instead, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.

[0028] First, the terms involved in this application are explained:

[0029] PEST model: P stands for Political, E stands for Economic, S stands for Social, and T stands for Technological. The PEST model is a theoretical framework for analyzing the macro environment. It mainly examines the four major external environmental factors that affect enterprises: politics, economy, society, and technology.

[0030] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) 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, storage, use, processing, transmission, provision, disclosure and application of the relevant data comply with the relevant laws, regulations and standards of the relevant countries and regions, take necessary confidentiality measures, do not violate public order and good morals, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0031] In addition, this application involves conducting big data analysis of user information (including but not limited to personal biometrics, identity data, consumption data, asset data, electronic terminal operation data, etc.), and using artificial intelligence technology to make automated decisions, and making technical solutions that have a significant impact on personal rights and interests based on the results of automated decisions. The application provides users with corresponding operation entrances for them to choose to agree or reject the results of automated decisions; if the user chooses to reject, the expert decision-making process will be entered.

[0032] It should be noted that the data processing methods, devices and equipment provided in this application can be used in the field of artificial intelligence, and can also be used in any field other than artificial intelligence. The application field of the data processing methods, devices and equipment in this application is not limited.

[0033] The specific application scenario of this application includes that before a financial institution lends money to an enterprise, it needs to analyze the credit status of the enterprise to guide the financial institution in formulating a lending strategy.

[0034] Related technologies determine the credit status of an enterprise by analyzing a large amount of sound financial data of the enterprise. The data analysis model used can often only process and analyze data in a single dimension related to finance. Therefore, under normal circumstances, the conclusions drawn about the credit status of an enterprise are accurate only when the financial data of the enterprise is sound.

[0035] Since most small and medium-sized enterprises have a short credit history and little accumulated financial data, there is little data available for credit analysis. As a result, the conclusions drawn from the analysis of the credit status of such small and medium-sized enterprises using relevant technologies are inaccurate. In addition, the conclusions drawn from the analysis of financial data in a single dimension cannot fully represent the credit status of the enterprise, and the impact of multi-dimensional data in the market on the credit of the enterprise should also be considered.

[0036] Based on the above analysis, it can be seen that the relevant technology can only analyze the single-dimensional data of the enterprise, and cannot simultaneously analyze the multi-dimensional data that affects the credit of the enterprise. There are technical problems such as inaccurate analysis of the credit situation of enterprises, especially small and medium-sized enterprises, which affects the credit strategy of financial institutions and makes the credit structure of financial institutions unreasonable.

[0037] The data processing method, device and equipment provided in this application are intended to solve the above technical problems in the prior art.

[0038] The technical solution of the present application and how the technical solution of the present application solves the above-mentioned technical problems are described in detail below with specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present application will be described below in conjunction with the accompanying drawings.

[0039] Figure 1 A flow chart of a data processing method provided in an embodiment of the present application is shown in FIG. Figure 1 As shown, the method includes:

[0040] S101. Acquire multi-dimensional association information of a target loan enterprise user, where each dimension of association information includes information corresponding to multiple time periods of the dimension.

[0041] It should be noted that the target loan enterprise users referred to in this step may be small, medium and micro enterprise users or large enterprise users. The multi-dimensional associated information obtained may be information within a preset period of time.

[0042] For example, the acquired multi-dimensional correlation information may be the multi-dimensional correlation information of the target loan enterprise user within two years before the current moment, and the multiple time periods may be multiple time periods within the two years.

[0043] S102: construct a multi-dimensional time series matrix based on the multi-dimensional correlation information, wherein the matrix data corresponding to each time period in the multi-dimensional time series matrix includes the matrixed result of the multi-dimensional correlation information corresponding to the time period.

[0044] It should be noted that in order to facilitate data statistics and analysis, multi-dimensional related information is matrixed, and other data representation forms that are convenient for data statistics and analysis are also applicable to this application, and this application does not limit this.

[0045] S103. Determine the similarity between the target loan enterprise user and the historical loan enterprise user based on the multi-dimensional time series matrix and the historical multi-dimensional time series matrix of the historical loan enterprise user; wherein the historical loan enterprise user has a historical credit score.

[0046] It should be noted that, since the historical loan enterprise user has obtained credit resources from the financial institution, the financial institution has saved the historical credit score of the historical loan enterprise user. In order to make the historical credit score of the historical loan enterprise user used for similarity comparison more accurate, the historical credit score can be the historical credit score result obtained by the financial institution analyzing the credit situation of the historical loan enterprise user using relevant technical solutions and adjusting it according to the actual credit situation of the historical loan enterprise user before and after the loan.

[0047] S104: Determine the credit score of the target loan enterprise user based on the similarity and the historical credit scores of the historical loan enterprise users.

[0048] In this step, the similarity obtained in step S103 can be used to determine the historical multi-dimensional time series matrix and historical credit score of the historical loan enterprise user used to calculate the credit score of the target loan enterprise user; the credit score of the target loan enterprise user is calculated by referring to the historical multi-dimensional time series matrix and historical credit score of the historical loan enterprise user. The credit score of the target loan enterprise user can be used to characterize the credit status of the target loan enterprise user.

[0049] The data processing method provided in this embodiment constructs a multi-dimensional time series matrix based on the multi-dimensional correlation information of multiple time periods of the target loan enterprise user, and determines the credit score of the target loan enterprise user by analyzing the similarity between the multi-dimensional time series matrix of the target loan enterprise user and the historical multi-dimensional time series matrix of the historical loan enterprise user, and the historical credit score of the historical loan enterprise user. Based on the multi-dimensional time series matrix, the credit status of the target loan enterprise user can be analyzed from multiple dimensions, and the multi-dimensional time series matrix of the target loan enterprise user and the historical multi-dimensional time series matrix of the historical loan enterprise user are compared and analyzed at the same time, thereby improving the accuracy of the analysis conclusion of the credit status of the target loan enterprise user.

[0050] There are a variety of data in the market that can be used to reflect the credit status of the target loan enterprise. Therefore, the information contained in the multi-dimensional correlation information can be determined by combining market analysis data, analysis data of the industry to which the enterprise belongs, and the enterprise's own data. The more accurate the determined multi-dimensional correlation information is, the more accurate the credit score of the target loan enterprise user will be. Specifically, the multi-dimensional correlation information can be obtained by combining the PEST model with the target loan enterprise user's own data. It should be noted that other methods for determining multi-dimensional correlation information are also applicable to this application, and this application does not limit this.

[0051] In some specific implementations, based on the PEST model and in combination with the target loan enterprise user's own data, the multi-dimensional correlation information obtained includes at least one of the following dimensional data: policy data, economic data, social data, technological data related to the industry of the target loan enterprise user, and financial data of the target loan enterprise user. The above multiple dimensional data can be obtained through various channels.

[0052] For example, policy data related to the industry of the target loan enterprise user can be obtained from public and authoritative policy conferences, publications, and news articles.

[0053] Economic data can be obtained from data publicly released by financial institutions that reflect the changing trends of the market economy.

[0054] Social data and scientific and technological data can be obtained from public social and scientific journals, articles published in the news, and corporate data disclosed by listed companies in the same industry.

[0055] The financial data of the target loan enterprise user may be obtained from public or unpublic financial data provided by the target loan enterprise user.

[0056] These data reflect the credit status of the target loan enterprise in multiple dimensions. The combination of multi-dimensional data forms a more accurate user portrait of the target loan enterprise users, so that financial institutions can use the combination of multi-dimensional data to conduct credit status analysis on the target loan enterprise users more accurately.

[0057] Since different dimensions have different statistical data information, different data information is recorded as different types of data. For the convenience of calculation, when constructing a multi-dimensional time series matrix in step S102, different types of data can be mapped into a unified standard data. In some embodiments of these embodiments, constructing a multi-dimensional time series matrix based on multi-dimensional association information may include the following sub-steps:

[0058] Step 1. For each dimension, obtain the unstructured data of the target loan enterprise users and the data related to the target industry within multiple time periods under this dimension, where the target industry is the industry to which the target loan enterprise users belong.

[0059] Among them, the unstructured data can include various types of data.

[0060] Exemplarily, the unstructured data can include text data, picture data, audio and video data, and report statistical data.

[0061] Step 2. For each time period, perform word segmentation on the unstructured data within this time period to obtain multiple word segmentation results.

[0062] Exemplarily, natural language processing technology can be used to perform word segmentation on the unstructured data. Using natural language processing technology to perform word segmentation on the unstructured data and obtaining multiple word segmentation results can include the following execution steps:

[0063] First, analyze the unstructured data, extract key information, convert this key information into text data, and segment the text data into multiple meaningful words according to preset rules.

[0064] Second, delete the meaningless words among the multiple words obtained by segmentation. The meaningless words can be stop words such as "de" (的) and "le" (了).

[0065] Then, analyze the word frequency and inter-word information of each word segment, and delete the word segments irrelevant to the application scenario, thereby obtaining multiple word segmentation results.

[0066] Step 3. Perform sentiment scoring on each word segmentation result to obtain the sentiment score results corresponding to each word segmentation result under this time period.

[0067] Exemplarily, the sentiment score results can include positive words, negative words, and neutral words. Each sentiment score result can be expressed as <word segmentation content, positive word>, or <word segmentation content, negative word>, or <word segmentation content, neutral word>.

[0068] Step 4. Construct a time series vector for this dimension from the sentiment score results corresponding to the word segmentation results of each time period; construct a multi-dimensional time series matrix from the time series vectors corresponding to each dimension.

[0069] In these implementations, a plurality of segmentation results are obtained by performing word segmentation processing on the non-structured data in the multi-dimensional associated information; sentiment scores are respectively performed on the plurality of segmentation results to obtain sentiment score results; and then a multi-dimensional time series matrix is ​​constructed using the sentiment score results corresponding to the plurality of dimensions in chronological order. The data of different dimensions are uniformly mapped into a plurality of sentiment score results, and the multi-dimensional time series matrix composed of the plurality of sentiment score results can be used for subsequent calculation processing.

[0070] In step S103, the multidimensional time series matrix used to determine the similarity between the target loan enterprise user and the historical loan enterprise users and the historical multidimensional time series matrix of the historical loan enterprise users have a large amount of data and the matrix data processing is complex. Therefore, these data can be processed with the help of artificial intelligence technology.

[0071] In some implementations of these implementations, determining the similarity between the target loan enterprise user and the historical loan enterprise user based on the multi-dimensional time series matrix and the historical multi-dimensional time series matrix of the historical loan enterprise user includes the following sub-steps:

[0072] Step 1: Obtain the historical multi-dimensional time series matrix and historical credit score of multiple historical loan enterprise users.

[0073] Step 2: Calculate the similarity between the target loan enterprise user and each historical loan enterprise user based on a time series classification model constructed by the historical multi-dimensional time series matrix and historical credit scores of multiple historical loan enterprise users.

[0074] It should be noted that the constructed time series classification model can be a machine learning model for processing time series data, and the model can classify unknown time series data based on known time series data. In step 2, the historical loan enterprise users with the closest similarity to the target loan enterprise users calculated by the model can be used to characterize the classification of the target loan enterprise users.

[0075] Specifically, in step 2, the time series classification model constructed by the historical multi-dimensional time series matrix and historical credit scores of multiple historical loan enterprise users calculates the similarity between the target loan enterprise user and each historical loan enterprise user, including:

[0076] First, the multi-dimensional time series matrix of the target loan enterprise user and the historical multi-dimensional time series matrices of multiple historical loan enterprise users are input into the time series classification model.

[0077] Second, based on the time series classification model, the similarity between the multi-dimensional time series matrix and each historical multi-dimensional time series matrix is ​​calculated to determine the similarity between the target loan enterprise user and each historical loan enterprise user.

[0078] It should be noted that the algorithm used to calculate the similarity between the multi-dimensional time series matrix and each historical multi-dimensional time series matrix may be a variety of algorithms, and this application does not limit this.

[0079] Exemplarily, the time series classification model can calculate the similarity between the multi-dimensional time series matrix and each historical multi-dimensional time series matrix based on the dynamic time series warping algorithm.

[0080] In these embodiments, a time series classification model is constructed by historical multi-dimensional time series matrices and historical credit scores of multiple historical loan enterprise users, and the similarity between the calculated multi-dimensional time series matrix and each historical multi-dimensional time series matrix can be used to subsequently determine the credit score of the target loan enterprise user.

[0081] Exemplarily, determining the credit score of the target loan enterprise user according to the similarity and the historical credit scores of the historical loan enterprise users in step S104 may include the following steps:

[0082] Step 1: The time series classification model determines the historical loan enterprise user that is closest to the target loan enterprise user based on the calculated similarity between the target loan enterprise user and each historical loan enterprise user.

[0083] Step 2: Analyze the relationship between the historical multi-dimensional time series matrix of the closest historical loan enterprise user and the multi-dimensional time series matrix of the target loan enterprise user to determine the adjustment factor for the historical loan score of the closest historical loan enterprise user.

[0084] Step 3: The credit score result of the target loan enterprise user is obtained by fine-tuning the historical credit score of the closest historical loan enterprise user based on the adjustment factor.

[0085] Since the constructed multi-dimensional time series matrix has a high dimension, the calculation process of the multi-dimensional time series matrix is ​​complicated. Therefore, the model calculation process can be optimized to make the model calculation more efficient.

[0086] Specifically, in some implementations, based on the time series classification model, the similarity between the multi-dimensional time series matrix and each historical multi-dimensional time series matrix is ​​calculated to determine the similarity between the target loan enterprise user and each historical loan enterprise user, including the following sub-steps:

[0087] Step 1: Use the principal component analysis method to reduce the dimensions of the multi-dimensional time series matrix and each historical multi-dimensional time series matrix to obtain the reduced-dimensional multi-dimensional time series matrix and each reduced-dimensional historical multi-dimensional time series matrix.

[0088] Exemplarily, multiple feature vectors of the same type in any time series vector in the multi-dimensional time series matrix may be combined to obtain a combined feature vector for calculating the similarity.

[0089] Step 2: Based on the time series classification model, calculate the similarity between the reduced multi-dimensional time series matrix and each reduced historical multi-dimensional time series matrix.

[0090] In these implementations, by performing dimensionality reduction processing on the multi-dimensional time series matrix, the complexity of the model operation is reduced, thereby improving the computing efficiency and saving computing resources.

[0091] Figure 2 A schematic diagram of the structure of a data processing device provided in an embodiment of the present application is shown in FIG. Figure 2 As shown, the enterprise data processing device 20 includes:

[0092] The acquisition module 201 is used to acquire multi-dimensional association information of the target loan enterprise user, where each dimension of association information includes information corresponding to multiple time periods of the dimension.

[0093] The construction module 202 is used to construct a multi-dimensional time series matrix based on the multi-dimensional correlation information, wherein the matrix data corresponding to each time period in the multi-dimensional time series matrix includes the matrix result of the multi-dimensional correlation information corresponding to the time period.

[0094] The first determination module 203 is used to determine the similarity between the target loan enterprise user and the historical loan enterprise user based on the multi-dimensional time series matrix and the historical multi-dimensional time series matrix of the historical loan enterprise user; wherein the historical loan enterprise user has a historical credit score.

[0095] The second determination module 204 is used to determine the credit score of the target loan enterprise user according to the similarity and the historical credit scores of the historical loan enterprise users.

[0096] In some specific implementations, the construction module 202 is also used to obtain, for each dimension, unstructured data related to the target loan enterprise users and the target industry in multiple time periods under the dimension, wherein the target industry is the industry to which the target loan enterprise users belong; for each time period, perform word segmentation on the unstructured data in the time period to obtain multiple word segmentation results; perform sentiment scoring on each word segmentation result to obtain the sentiment score results corresponding to each word segmentation result in the time period; construct a time series vector of the dimension from the sentiment score results corresponding to the word segmentation results of each of the multiple time periods; and form a multi-dimensional time series matrix from the time series vectors corresponding to each dimension.

[0097] In some of these embodiments, the first determination module 203 is also used to obtain the historical multi-dimensional time series matrix and historical credit score of each of the multiple historical loan enterprise users; and calculate the similarity between the target loan enterprise user and each historical loan enterprise user based on the time series classification model constructed by the historical multi-dimensional time series matrix and historical credit score of each of the multiple historical loan enterprise users.

[0098] In some of these embodiments, the first determination module 203 is also used to input the multi-dimensional time series matrix of the target loan enterprise user and the historical multi-dimensional time series matrices of multiple historical loan enterprise users into the time series classification model; based on the time series classification model, the similarity between the multi-dimensional time series matrix and each historical multi-dimensional time series matrix is ​​calculated to determine the similarity between the target loan enterprise user and each historical loan enterprise user.

[0099] In some of these embodiments, the first determination module 203 is also used to use the principal component analysis method to reduce the dimensions of the multidimensional time series matrix and each historical multidimensional time series matrix to obtain the reduced multidimensional time series matrix and the reduced historical multidimensional time series matrices; based on the time series classification model, calculate the similarity between the reduced multidimensional time series matrix and the reduced historical multidimensional time series matrices.

[0100] In some of these embodiments, the multi-dimensional correlation information obtained by the first determination module 203 includes at least one of the following dimensional data: policy data, economic data, social data, technological data related to the industry of the target loan enterprise user, and financial data of the target loan enterprise user.

[0101] Figure 3 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application, such as Figure 3 As shown, the electronic device 30 includes: a processor 301, and a memory 302 communicatively connected to the processor 301; the processor 301 can be communicatively connected to the memory 302 via a communication component 303; the processor 301, the memory 302, and the communication component 303 can be connected via a bus 304; the memory 302 stores computer-executable instructions; the processor 301 executes the computer-executable instructions stored in the memory 302 to execute the above method.

[0102] The specific implementation process of the processor 301 can be found in the above method embodiment, and its implementation principle and technical effect are similar, so this embodiment will not be repeated here.

[0103] The present application also provides a computer program product, including a computer program, which implements the above method when executed by a processor.

[0104] The present application also provides a computer-readable storage medium, in which computer-executable instructions are stored. When a processor executes the computer-executable instructions, the above method is implemented.

[0105] The above-mentioned readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk. The readable storage medium can be any available medium that can be accessed by a general or special-purpose computer.

[0106] An exemplary readable storage medium is coupled to a processor so that the processor can read information from the readable storage medium and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can be located in an application specific integrated circuit (Application Specific Integrated Circuits, referred to as: ASIC). Of course, the processor and the readable storage medium can also exist in the device as discrete components.

[0107] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all optional embodiments, and the actions and modules involved are not necessarily required by the present application.

[0108] It should be further noted that, although the various steps in the flowchart are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless there is a clear description in this article, the execution of these steps is not strictly limited in order, and these steps can be executed in other orders. Moreover, at least a portion of the steps in the flowchart may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these sub-steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a portion of the sub-steps or stages of other steps.

[0109] It should be understood that the above-mentioned device embodiments are only illustrative, and the device of the present application can also be implemented in other ways. For example, the division of units / modules in the above-mentioned embodiments is only a logical function division, and there may be other division methods in actual implementation. For example, multiple units, modules or components can be combined, or can be integrated into another system, or some features can be ignored or not executed.

[0110] In addition, unless otherwise specified, each functional unit / module in each embodiment of the present application may be integrated into one unit / module, each unit / module may exist physically separately, or two or more units / modules may be integrated together. The above-mentioned integrated unit / module may be implemented in the form of hardware or in the form of a software program module.

[0111] If the integrated unit / module is implemented in the form of hardware, the hardware may be a digital circuit, an analog circuit, etc. The physical implementation of the hardware structure includes but is not limited to transistors, memristors, etc. Unless otherwise specified, the processor may be any appropriate hardware processor, such as a CPU, a GPU, an FPGA, a DSP, an ASIC, etc. Unless otherwise specified, the storage unit may be any appropriate magnetic storage medium or magneto-optical storage medium, such as a resistive random access memory (RRAM), a dynamic random access memory (DRAM), a static random access memory (SRAM), an enhanced dynamic random access memory (EDRAM), a high-bandwidth memory (HBM), a hybrid memory cube (HMC), etc.

[0112] If the integrated unit / module is implemented in the form of a software program module and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a memory, including a number of instructions to enable a computer device (which can be a personal computer, server or network device, etc.) to execute all or part of the steps of the various embodiments of the present application. The aforementioned memory includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, disk or optical disk and other media that can store program codes.

[0113] In the above embodiments, the description of each embodiment has its own emphasis. For the part not described in detail in a certain embodiment, please refer to the relevant description of other embodiments. The technical features of the above embodiments can be combined arbitrarily. In order to make the description concise, all possible combinations of the technical features in the above embodiments are not described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0114] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the invention disclosed herein. The present application is intended to cover any modification, use or adaptation of the present application, which follows the general principles of the present application and includes common knowledge or customary techniques in the art that are not disclosed in the present application. The specification and examples are intended to be exemplary only, and the true scope and spirit of the present application are indicated by the following claims.

[0115] It should be understood that the present application is not limited to the precise structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present application is limited only by the appended claims.

Claims

1. A data processing method, comprising: Acquire multi-dimensional association information of target loan enterprise users, where each dimension of association information includes information corresponding to multiple time periods of the dimension; Constructing a multi-dimensional time series matrix based on the multi-dimensional association information, wherein the matrix data corresponding to each time period in the multi-dimensional time series matrix includes the matrixed result of the multi-dimensional association information corresponding to the time period; Determine the similarity between the target loan enterprise user and the historical loan enterprise user based on the multi-dimensional time series matrix and the historical multi-dimensional time series matrix of the historical loan enterprise user; wherein the historical loan enterprise user has a historical credit score; The credit score of the target loan enterprise user is determined according to the similarity and the historical credit score of the historical loan enterprise user.

2. The method according to claim 1, characterized in that The constructing a multi-dimensional time series matrix based on the multi-dimensional association information includes: For each dimension, obtain the target loan enterprise user and non-structured data related to the target industry in multiple time periods under the dimension, wherein the target industry is the industry to which the target loan enterprise user belongs; For each time period, the unstructured data in the time period is segmented to obtain multiple segmentation results; Perform sentiment scoring on each word segmentation result to obtain the sentiment score results corresponding to each word segmentation result in the time period; The time series vector of the dimension is constructed based on the sentiment score results corresponding to the word segmentation results of each of the multiple time periods; and the multi-dimensional time series matrix is ​​constructed based on the time series vectors corresponding to each dimension.

3. The method according to claim 1, characterized in that The determining the similarity between the target loan enterprise user and the historical loan enterprise user based on the multi-dimensional time series matrix and the historical multi-dimensional time series matrix of the historical loan enterprise user includes: Obtain the historical multi-dimensional time series matrix and historical credit scores of multiple historical loan enterprise users; The similarity between the target loan enterprise user and each historical loan enterprise user is calculated based on a time series classification model constructed by the historical multi-dimensional time series matrix and historical credit scores of multiple historical loan enterprise users.

4. The method according to claim 3, characterized in that The time series classification model constructed by the historical multi-dimensional time series matrix and historical credit scores of the plurality of historical loan enterprise users calculates the similarity between the target loan enterprise user and each historical loan enterprise user, including: Inputting the multi-dimensional time series matrix of the target loan enterprise user and the historical multi-dimensional time series matrices of each of the plurality of historical loan enterprise users into the time series classification model; Based on the time series classification model, the similarity between the multi-dimensional time series matrix and each of the historical multi-dimensional time series matrices is calculated to determine the similarity between the target loan enterprise user and each of the historical loan enterprise users.

5. The method according to claim 4, characterized in that The calculating the similarity between the multi-dimensional time series matrix and each of the historical multi-dimensional time series matrices based on the time series classification model, thereby determining the similarity between the target loan enterprise user and each of the historical loan enterprise users, includes: Use a principal component analysis method to reduce the dimensions of the multidimensional time series matrix and each of the historical multidimensional time series matrices to obtain a reduced-dimensional multidimensional time series matrix and reduced-dimensional historical multidimensional time series matrices; Based on the time series classification model, the similarity between the reduced multi-dimensional time series matrix and each of the reduced historical multi-dimensional time series matrices is calculated.

6. The method according to any one of claims 1 to 5, characterized in that: The multi-dimensional association information includes at least one of the following dimensional data: policy data, economic data, social data, technological data related to the industry of the target loan enterprise user, and financial data of the target loan enterprise user.

7. A device for processing enterprise data, comprising: An acquisition module is used to acquire multi-dimensional association information of target loan enterprise users, where each dimension of association information includes information corresponding to multiple time periods of the dimension; A construction module, configured to construct a multi-dimensional time series matrix based on the multi-dimensional association information, wherein the matrix data corresponding to each time period in the multi-dimensional time series matrix includes a matrixed result of the multi-dimensional association information corresponding to the time period; A first determination module is used to determine the similarity between the target loan enterprise user and the historical loan enterprise user based on the multi-dimensional time series matrix and the historical multi-dimensional time series matrix of the historical loan enterprise user; wherein the historical loan enterprise user has a historical credit score; The second determination module is used to determine the credit score of the target loan enterprise user according to the similarity and the historical credit score of the historical loan enterprise user.

8. An electronic device, characterized in that: include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory to implement the method according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, which are used to implement the method according to any one of claims 1 to 6 when executed by a processor.

10. A computer program product, characterized in that The invention comprises a computer program, which implements the method according to any one of claims 1 to 6 when being executed by a processor.

Images