Information processing method and system based on multi-structure digital financial service

Abstract

The invention relates to the technical field of financial service information processing, and discloses an information processing method and system based on multi-structure digital financial services, and the method comprises the steps: collecting customer financing information, and carrying out the preprocessing; carrying out financing behavior analysis on the preprocessed customer financing information; performing customer group identification on customers according to the extracted financing evaluation feature vectors, and generating financing evaluation feature cluster vectors of customer clusters; and constructing a financing evaluation model to carry out financing credit evaluation on all clients in the client cluster, and recommending financing products according to a financing credit evaluation result. According to the method, the periodic component and the trend component in the customer financing information are extracted to represent the time sequence characteristics of the customer financing information, the financing evaluation characteristic vector representing the time sequence change of the customer fund transaction demand is formed, and the financing evaluation characteristic cluster vectors of different customer clusters are mapped into the financing credit evaluation result. The demand degree of the customer for funds is represented, and customized financing product recommendation is realized.

Description

Technical Field

[0001] The present invention relates to the field of financial service information processing, and in particular to a method and system for processing financial service information based on multi-structure digital technology. Background Art

[0002] With the advent of the digital age, financing services are facing unprecedented challenges and opportunities. Digital financial services have become a trend in the development of the financial industry, and multi-structure digital financial service information processing methods, as an important part of digital financial services, are receiving more and more attention. Multi-structure digital financial service information processing methods refer to the use of artificial intelligence, big data, blockchain and other technologies to comprehensively analyze and process the information of multiple components (such as users, products, channels, etc.) involved in financial services to achieve more accurate, secure and efficient digital financial services. It can provide financing companies, financial institutions and government departments with fast and convenient service windows from multiple angles, which is of great significance to promoting the innovation and upgrading of digital financial services.

[0003] That is, the current information processing methods for multi-structure digital financial services have technical problems such as insufficient model generalization capabilities and insufficient support for personalized services. Summary of the invention

[0004] In view of this, the present invention proposes a multi-structure digital financial service information processing method, which integrates the information of multiple components such as financing enterprises, financing products, financing channels, government platforms, etc., to achieve accurate customer positioning, personalized service recommendations, etc., thereby improving service quality.

[0005] To achieve the above-mentioned purpose, the present invention provides a method for processing information of multi-structure digital financial services, comprising the following steps: S1: collecting and preprocessing customer financing information to obtain preprocessed customer financing information; the financing information includes at least 4 financing indicators, and the financing indicators include bank transaction records, income level, liabilities and credit rating information; S2: Analyze the financing behavior of the pre-processed customer financing information and extract the financing evaluation feature vector; S3: According to the financing evaluation feature vector, the customers are divided into different customer clusters based on the K-means algorithm and Gaussian mixture analysis, and the financing evaluation feature cluster vector of the customer cluster is generated; S4: constructing a financing assessment model, performing financing credit assessment on all customers in the customer cluster according to the financing assessment feature cluster vector, and recommending financing products according to the financing credit assessment results; The method of performing financing credit assessment on all customers in a customer cluster includes: receiving a financing assessment feature cluster vector of the customer cluster; performing residual convolution calculation on the financing assessment feature cluster vector to generate a convolution feature map of the financing assessment feature cluster vector; performing maximum pooling processing on the convolution feature map, and mapping the pooled convolution feature map into a financing credit assessment result.

[0006] As a further improvement method of the present invention: Optionally, the financing information collected from N customers is: ; in: Indicates the collected The financing information of the customers mentioned above, Respectively represent Financing data series of each client for each of the above mentioned financing indicators; Indicates Customers in The financing data sequence of each of the financing indicators; Represents the financing data sequence middle Financing data for the month, Represents the financing data sequence Middle Financing data for the month.

[0007] Optionally, in the step S1, the process of preprocessing the financing information is: S11: Construct filters for different financing indicators, where The filter of the financing indicator is , the filter The filter coefficient is: , using the filter to filter the corresponding financing data sequence: ; in: Represents the financing data sequence The filtering result of Indicates the financing data The filtering result of S12: Calculate the Fourier coefficients of the financing data sequence at H harmonic frequencies after different filtering processes: ; in: is the financing data sequence after filtering In the The Fourier coefficients at the harmonic frequencies, ; represents transpose; Represents the financing data sequence after filtering The Fourier coefficient matrix of ; is the harmonic matrix; For the Fourier coefficient matrix To solve, the solution formula is: ; S13: Using the Fourier coefficient matrix obtained by solving, the financing data sequence after filtering is subjected to harmonic filtering in a form-preserving manner, and the harmonic filtering result is used as the preprocessing result of the financing data sequence, wherein the financing data sequence after filtering is The harmonic filtering formula is: ; in: is the financing data sequence after filtering The harmonic filtering result is, for The harmonic filtering result; S14: Constructing the pre-processed customer financing information: ; in: No. the pre-processed customer financing information of the customers, Respectively represent The financing data sequence of the customers after preprocessing of each of the financing indicators.

[0008] Optionally, the S2 step includes: S21: Obtaining the pre-processed customer financing information The financing data sequence after preprocessing of each of the financing indicators; S22: using a time series decomposition method to obtain a time series correlation matrix of the financing data sequence after preprocessing; S23: Using the time series fusion module to iterate and fuse the time series state of the time series correlation matrix to obtain the pre-processed customer financing information The financing evaluation feature vector of the corresponding customer : ; in: represents the L2 norm; Representing the time series features in the time series correlation matrix; represents the trend feature convolution matrix, Represents the convolution operator.

[0009] Optionally, the step S22 includes: S221: Initialize and generate the financing data sequence after preprocessing The trend component , set the financing data sequence after preprocessing The current decomposition number is d, and the initial value of d is 0. After the dth time series decomposition, the financing data sequence The trend component of is: ,in Represents the trend component The mth component value in ; S222: Extracting the financing data sequence after preprocessing The periodic component : ; in: Represents the sequence smoothing function; represents an exponential function with a natural constant as base; The number of window sequence points representing the smoothing window; represents the difference parameter; S223: Decompose the trend component iteratively: ; in: Represents the periodic component The Mth component value in ; S224: Calculate the residual of the current d+1th decomposition : ; like If the decomposition value is less than the preset decomposition threshold, the iteration is terminated, and the trend component and the period component of the current d+1th decomposition are extracted as the financing data sequence after preprocessing. The timing characteristics of ,in , , represents the L2 norm; Otherwise, , return to step S222; S225: Construct the financing data sequence after preprocessing The time series correlation matrix: ; in: Represents the financing data sequence after preprocessing The time series correlation matrix.

[0010] Optionally, the S3 step includes: S31: Calculate the distance between the financing evaluation feature vectors of different customers: ; in: Indicates Customers and The distance between the financing evaluation feature vector of the customers, ; S32: Based on the distances between the financing evaluation feature vectors of different customers, the K-means algorithm is used to perform clustering processing on the financing evaluation feature vectors to obtain cluster centers, among which The cluster centers are , ; S33: generating Gaussian mixture parameters corresponding to each of the cluster centers; S34: Calculate and obtain different financing evaluation feature vectors to be assigned to The probability of the cluster center; The financing evaluation feature vector is assigned to the cluster center with the highest probability, forming customer clusters; Generate financing evaluation feature cluster vectors for customer clusters, where The financing evaluation feature cluster vector of customer clusters is .

[0011] Optionally, generating a financing assessment feature cluster vector of a customer cluster includes: Get the The belonging probability of each financing assessment feature vector in a customer cluster, wherein the belonging probability of the cluster center is 1; For All the belonging probabilities in the customer clusters are normalized so that the sum of all the belonging probabilities after the normalization is 1; The standardized attribution probability is used as the weight of the corresponding financing evaluation feature vector or the cluster center, and weighted processing is performed, and the weighted processing result is used as the first The financing assessment feature cluster vector of customer clusters.

[0012] Optionally, the financing assessment model includes an input layer, a residual convolution layer, a maximum pooling layer and a financing credit assessment layer; The input layer is used to receive the financing assessment feature cluster vector of the customer cluster; The residual convolution layer is a residual learning unit structure in ResNet, which is used to perform residual convolution calculation on the financing evaluation feature cluster vector to generate a convolution feature map of the financing evaluation feature cluster vector; The maximum pooling layer is a pooling layer structure in a convolutional neural network, which is used to perform maximum pooling processing on the convolution feature map to form a pooled convolution feature map; The financing credit assessment layer is a fully connected structure in a neural network, which is used to map the pooled convolutional feature map into a financing credit assessment result. The financing credit assessment result is displayed in the form of a score on a percentage basis. The higher the score, the greater the customer's capital demand and the higher the repayment ability, and thus recommends financing products with higher capital support.

[0013] Optionally, the step of performing financing credit assessment on all customers in the customer cluster includes: S41: The input layer receives The financing evaluation feature cluster vector of customer clusters ; S42: The residual convolution layer performs the calculation on the financing assessment feature cluster vector Perform residual convolution calculation to generate the financing evaluation feature cluster vector The convolutional feature map: ; in: Represents the financing evaluation feature cluster vector The convolution feature map of ; represents the residual information extraction matrix; S43: The maximum pooling layer performs pooling on the convolution feature map Perform maximum pooling processing to form the pooled convolution feature map ; ; in: Represents the maximum pooling operation; S44: The financing credit assessment layer converts the pooled convolution feature map Mapped to the financing credit assessment result: ; in: Indicates The financing credit evaluation results of all customers in a customer cluster; represents the collection of financing credit assessment results, , The financing credit assessment result set Any evaluation result in Indicates the evaluation result The corresponding mapping matrix is, Represents the financing credit evaluation result set Select the Achieve the maximum evaluation result As .

[0014] In order to solve the above problems, the present invention provides a multi-structure digital financial service information processing system, the system comprising: A data processing module is used to collect and pre-process customer financing information to obtain pre-processed customer financing information; The customer behavior identification module is used to analyze the financing behavior of the pre-processed customer financing information, extract the financing evaluation feature vector, identify the customer groups according to the extracted financing evaluation feature vector, divide the customers into different customer clusters, and generate the financing evaluation feature cluster vector of the customer cluster; The financing service device is used to build a financing evaluation model, use the financing evaluation feature cluster vector of the customer cluster as input to perform financing credit evaluation on all customers in the customer cluster, and recommend financing products based on the financing credit evaluation results.

[0015] In order to solve the above problem, the present invention further provides an electronic device, the electronic device comprising: A memory storing at least one instruction; Communication interface, enabling electronic equipment to communicate; and A processor executes instructions stored in the memory to implement the above-mentioned multi-structure digital financial service information processing method.

[0016] In order to solve the above problems, the present invention also provides a computer-readable storage medium, in which at least one instruction is stored. The at least one instruction is executed by a processor in an electronic device to implement the above-mentioned multi-structure digital financial service information processing method.

[0017] Compared with the prior art, the present invention proposes a method for processing information of multi-structure digital financial services, which has the following advantages: First, this scheme obtains customer financing information from multiple channels, uses filters to remove most of the noise data in the customer financing information, decomposes the information in the frequency domain, constructs Fourier coefficients at different harmonic frequencies, extracts the harmonic form that best represents the original information, and performs smoothing to obtain pre-processed customer financing information that represents key information. The time series characteristics of the customer financing information are characterized by extracting the periodic component and trend component in the customer financing information, and a financing evaluation feature vector that represents the changes in the customer transaction time series is constructed.

[0018] At the same time, this solution performs customer classification based on the financing evaluation feature vectors of different customers, and integrates Gaussian mixture parameters in the classification process to construct probability density functions of different cluster centers, realize the probability calculation of the non-cluster center financing evaluation feature vector belonging to different cluster centers, and realize customer classification by combining vector distance and belonging probability. By dividing customers into different customer clusters and building a financing evaluation model based on the financing evaluation feature cluster vector, this solution can adjust the model input and processing methods according to the characteristics of the customer group, improve the adaptability and generalization ability of the model, and is suitable for a variety of customer groups and scenarios.

[0019] The financing assessment feature cluster vectors of different customer clusters are mapped into financing credit assessment results to characterize the customer's demand for funds, realize customized financing product recommendations, and improve the level of personalization of financial services. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 A flowchart of a method for processing information of a multi-structure digital financial service provided by an embodiment of the present invention; Figure 2 A functional module diagram of a multi-structure digital financial service information processing system provided by an embodiment of the present invention; Figure 2 In: 100 multi-structure digital financial service information processing system, 101 data processing module, 102 customer behavior recognition module, 103 financing service device; Figure 3 A schematic diagram of the structure of an electronic device for implementing a method for processing information based on multi-configuration digital financial services provided by an embodiment of the present invention.

[0021] Figure 3 In: 1 electronic device, 10 processor, 11 memory, 12 program, 13 communication interface; The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0022] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0023] The embodiment of the present application provides a method for processing information based on multi-structure digital financial services. The execution subject of the method for processing information based on multi-structure digital financial services includes but is not limited to at least one of the electronic devices such as a server and a terminal that can be configured to execute the method provided by the embodiment of the present application. In other words, the method for processing information based on multi-structure digital financial services can be executed by software or hardware installed on a terminal device or a server device, and the software can be a blockchain platform. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc. Example

[0024] Embodiment 1:

[0025] S1: Collecting customer financing information and preprocessing it to obtain preprocessed customer financing information; the financing information includes at least 4 financing indicators, and the financing indicators include bank transaction records, income level, liabilities and credit rating information.

[0026] The financing information collected from N customers is: ; in: Indicates the collected The financing information of the customers mentioned above, Respectively represent Financing data series of each client for each of the above mentioned financing indicators; Indicates Customers in The financing data sequence of each of the financing indicators; Represents the financing data sequence middle Financing data for the month, Represents the financing data sequence Middle Financing data for the month.

[0027] In step S1, the process of preprocessing the financing information is as follows: S11: Construct filters for different financing indicators, where The filter of the financing indicator is , the filter The filter coefficient is: , using the filter to filter the corresponding financing data sequence: ; in: Represents the financing data sequence The filtering result of Indicates the financing data The filtering result of S12: Calculate the Fourier coefficients of the financing data sequence at H harmonic frequencies after different filtering processes: ; in: is the financing data sequence after filtering In the The Fourier coefficients at the harmonic frequencies, ; represents transpose; Represents the financing data sequence after filtering The Fourier coefficient matrix of ; is the harmonic matrix; For the Fourier coefficient matrix To solve, the solution formula is: ; S13: Using the Fourier coefficient matrix obtained by solving, the financing data sequence after filtering is subjected to harmonic filtering in a form-preserving manner, and the harmonic filtering result is used as the preprocessing result of the financing data sequence, wherein the financing data sequence after filtering is The harmonic filtering formula is: ; in: is the financing data sequence after filtering The harmonic filtering result is, for The harmonic filtering result; S14: Constructing the pre-processed customer financing information: ; in: No. the pre-processed customer financing information of the customers, Respectively represent The financing data sequence of the customers after preprocessing of each of the financing indicators.

[0028] S2: Performing financing behavior analysis on the pre-processed customer financing information to extract financing evaluation feature vectors.

[0029] The S2 step includes: S21: Obtaining the pre-processed customer financing information The financing data sequence after preprocessing of each of the financing indicators; S22: using a time series decomposition method to obtain a time series correlation matrix of the financing data sequence after preprocessing; S23: Using the time series fusion module to iterate and fuse the time series state of the time series correlation matrix to obtain the pre-processed customer financing information The financing evaluation feature vector of the corresponding customer : ; in: represents the L2 norm; Representing the time series features in the time series correlation matrix; represents the trend feature convolution matrix, Represents the convolution operator.

[0030] The step S22 includes: S221: Initialize and generate the financing data sequence after preprocessing The trend component , set the financing data sequence after preprocessing The current decomposition number is d, and the initial value of d is 0. After the dth time series decomposition, the financing data sequence The trend component of is: ,in Represents the trend component The mth component value in ; S222: Extracting the financing data sequence after preprocessing The periodic component : ; in: Represents the sequence smoothing function; represents an exponential function with a natural constant as base; The number of window sequence points representing the smoothing window; represents the difference parameter; S223: Decompose the trend component iteratively: ; in: Represents the periodic component The Mth component value in ; S224: Calculate the residual of the current d+1th decomposition : ; like If the decomposition value is less than the preset decomposition threshold, the iteration is terminated, and the trend component and the period component of the current d+1th decomposition are extracted as the financing data sequence after preprocessing. The timing characteristics of ,in , , represents the L2 norm; Otherwise, , return to step S222; S225: Construct the financing data sequence after preprocessing The time series correlation matrix: ; in: Represents the financing data sequence after preprocessing The time series correlation matrix.

[0031] S3: According to the financing evaluation feature vector, the customers are divided into different customer clusters based on the K-means algorithm and Gaussian mixture analysis, and the financing evaluation feature cluster vector of the customer cluster is generated.

[0032] The S3 step includes: S31: Calculate the distance between the financing evaluation feature vectors of different customers: ; in: Indicates Customers and The distance between the financing evaluation feature vector of the customers, ; S32: Based on the distances between the financing evaluation feature vectors of different customers, the K-means algorithm is used to perform clustering processing on the financing evaluation feature vectors to obtain cluster centers, among which The cluster centers are , ; S33: generating Gaussian mixture parameters corresponding to each of the cluster centers; S34: Calculate and obtain different financing evaluation feature vectors to be assigned to The probability of the cluster center; The financing evaluation feature vector is assigned to the cluster center with the highest probability, forming customer clusters; Generate financing evaluation feature cluster vectors for customer clusters, where The financing evaluation feature cluster vector of customer clusters is .

[0033] The generating of the financing assessment feature cluster vector of the customer cluster includes: Get the The belonging probability of each financing assessment feature vector in a customer cluster, wherein the belonging probability of the cluster center is 1; For All the belonging probabilities in the customer clusters are normalized so that the sum of all the belonging probabilities after the normalization is 1; The standardized attribution probability is used as the weight of the corresponding financing evaluation feature vector or the cluster center, and weighted processing is performed, and the weighted processing result is used as the first The financing assessment feature cluster vector of customer clusters.

[0034] S4: constructing a financing assessment model, performing financing credit assessment on all customers in the customer cluster according to the financing assessment feature cluster vector, and recommending financing products according to the financing credit assessment results; The method of performing financing credit assessment on all customers in a customer cluster includes: receiving a financing assessment feature cluster vector of the customer cluster; performing residual convolution calculation on the financing assessment feature cluster vector to generate a convolution feature map of the financing assessment feature cluster vector; performing maximum pooling processing on the convolution feature map, and mapping the pooled convolution feature map into a financing credit assessment result.

[0035] The financing assessment model includes an input layer, a residual convolution layer, a maximum pooling layer and a financing credit assessment layer; The input layer is used to receive the financing assessment feature cluster vector of the customer cluster; The residual convolution layer is a residual learning unit structure in ResNet, which is used to perform residual convolution calculation on the financing evaluation feature cluster vector to generate a convolution feature map of the financing evaluation feature cluster vector; The maximum pooling layer is a pooling layer structure in a convolutional neural network, which is used to perform maximum pooling processing on the convolution feature map to form a pooled convolution feature map; The financing credit assessment layer is a fully connected structure in a neural network, which is used to map the pooled convolutional feature map into a financing credit assessment result. The financing credit assessment result is displayed in the form of a score on a percentage basis. The higher the score, the greater the customer's capital demand and the higher the repayment ability, and thus recommends financing products with higher capital support.

[0036] The financing credit assessment of all customers in the customer cluster includes: S41: The input layer receives The financing evaluation feature cluster vector of customer clusters ; S42: The residual convolution layer performs the calculation on the financing assessment feature cluster vector Perform residual convolution calculation to generate the financing evaluation feature cluster vector The convolutional feature map: ; in: Represents the financing evaluation feature cluster vector The convolution feature map of ; represents the residual information extraction matrix; S43: The maximum pooling layer performs pooling on the convolution feature map Perform maximum pooling processing to form the pooled convolution feature map ; ; in: Represents the maximum pooling operation; S44: The financing credit assessment layer converts the pooled convolution feature map Mapped to the financing credit assessment result: ; in: Indicates The financing credit evaluation results of all customers in a customer cluster; represents the collection of financing credit assessment results, , The financing credit assessment result set Any evaluation result in Indicates the evaluation result The corresponding mapping matrix is, Represents the financing credit evaluation result set Select the Achieve the maximum evaluation result As .

[0037] Embodiment 2: like Figure 2 As shown, it is a functional module diagram of a multi-structure digital financial service information processing system provided by an embodiment of the present invention, which can implement the multi-structure digital financial service information processing method in Example 1.

[0038] The multi-configuration digital financial service information processing system 100 of the present invention can be installed in an electronic device. According to the functions to be implemented, the multi-configuration digital financial service information processing system can include a data processing module 101, a customer behavior recognition module 102 and a financing service device 103. The module of the present invention can also be referred to as a unit, which refers to a series of computer program segments that can be executed by an electronic device processor and can complete fixed functions, which are stored in the memory of the electronic device.

[0039] The data processing module 101 is used to collect and pre-process customer financing information to obtain pre-processed customer financing information; the financing information includes at least four financing indicators, including bank transaction records, income level, liabilities and credit rating information; The customer behavior identification module 102 is used to analyze the financing behavior of the pre-processed customer financing information, extract the financing evaluation feature vector, identify the customer groups according to the extracted financing evaluation feature vector, divide the customers into different customer clusters, and generate the financing evaluation feature cluster vector of the customer cluster; The financing service device 103 is used to construct a financing evaluation model, perform financing credit evaluation on all customers in the customer cluster using the financing evaluation feature cluster vector of the customer cluster as input, and recommend financing products based on the financing credit evaluation results.

[0040] In detail, each module in the multi-structure digital financial service information processing system 100 described in the embodiment of the present invention is used in the same manner as described above. Figure 1 The technical means described in the multi-structure digital financial service information processing method are the same and can produce the same technical effects, so they will not be repeated here.

[0041] Embodiment 3: like Figure 3 1 is a schematic diagram of the structure of an electronic device for implementing a method for processing information based on multi-structure digital financial services provided by an embodiment of the present invention.

[0042] The electronic device 1 may include a processor 10 , a memory 11 , a communication interface 13 and a bus, and may also include a computer program stored in the memory 11 and executable on the processor 10 , such as a program 12 .

[0043] The memory 11 includes at least one type of readable storage medium, and the readable storage medium includes a flash memory, a mobile hard disk, a multimedia card, a card-type memory (e.g., SD or DX memory, etc.), a magnetic memory, a disk, an optical disk, etc. In some embodiments, the memory 11 may be an internal storage unit of the electronic device 1, such as a mobile hard disk of the electronic device 1. In other embodiments, the memory 11 may also be an external storage device of the electronic device 1, such as a plug-in mobile hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (SecureDigital, SD) card, a flash card (Flash Card), etc. equipped on the electronic device 1. Further, the memory 11 may also include both an internal storage unit of the electronic device 1 and an external storage device. The memory 11 may not only be used to store application software and various types of data installed in the electronic device 1, such as the code of the program 12, etc., but also be used to temporarily store data that has been output or is to be output.

[0044] The processor 10 may be composed of an integrated circuit in some embodiments, for example, a single packaged integrated circuit, or a plurality of packaged integrated circuits with the same or different functions, including one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and combinations of various control chips, etc. The processor 10 is the control core (Control Unit) of the electronic device, and uses various interfaces and lines to connect various components of the entire electronic device, and executes or executes programs or modules stored in the memory 11 (such as programs 12 for implementing information processing based on multi-structured digital financial services), and calls data stored in the memory 11 to execute various functions of the electronic device 1 and process data.

[0045] The communication interface 13 may include a wired interface and / or a wireless interface (such as a WI-FI interface, a Bluetooth interface, etc.), which is generally used to establish a communication connection between the electronic device 1 and other electronic devices, and to achieve connection and communication between internal components of the electronic device.

[0046] The bus may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc. The bus is configured to realize connection and communication between the memory 11 and at least one processor 10, etc.

[0047] Figure 3 Only an electronic device with components is shown, and those skilled in the art will understand that Figure 3 The structure shown does not constitute a limitation on the electronic device 1, and may include fewer or more components than shown in the figure, or combine certain components, or arrange the components differently.

[0048] For example, although not shown, the electronic device 1 may also include a power source (such as a battery) for supplying power to various components. Preferably, the power source may be logically connected to the at least one processor 10 through a power management device, so that the power management device can realize functions such as charging management, discharging management, and power consumption management. The power source may also include one or more DC or AC power sources, recharging devices, power failure detection circuits, power converters or inverters, power status indicators, and other arbitrary components. The electronic device 1 may also include a variety of sensors, Bluetooth modules, Wi-Fi modules, etc., which will not be repeated here.

[0049] Optionally, the electronic device 1 may further include a user interface, which may be a display, an input unit (such as a keyboard), or a standard wired interface or a wireless interface. Optionally, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, and an OLED (Organic Light-Emitting Diode) touch device. The display may also be appropriately referred to as a display screen or a display unit, which is used to display information processed in the electronic device 1 and to display a visual user interface.

[0050] It should be understood that the embodiment is for illustration only and the scope of the patent application is not limited to this structure.

[0051] It should be noted that the serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages and disadvantages of the embodiments. And the terms "including", "comprising" or any other variants thereof in this article are intended to cover non-exclusive inclusion, so that a process, device, article or method including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, device, article or method. In the absence of further restrictions, an element defined by the sentence "including a ..." does not exclude the presence of other identical elements in the process, device, article or method including the element.

[0052] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes a number of instructions for a terminal device (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present invention.

[0053] The above are only preferred embodiments of the present invention, and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A method for processing information based on multi-structure digital financial services, characterized in that: The method comprises: S1: collecting and preprocessing customer financing information to obtain preprocessed customer financing information; the financing information includes at least 4 financing indicators, and the financing indicators include bank transaction records, income level, liabilities and credit rating information; S2: Analyze the financing behavior of the pre-processed customer financing information and extract the financing evaluation feature vector; S3: According to the financing evaluation feature vector, the customers are divided into different customer clusters based on the K-means algorithm and Gaussian mixture analysis, and the financing evaluation feature cluster vector of the customer cluster is generated; S4: Construct a financing assessment model, perform financing credit assessment on all customers in the customer cluster according to the financing assessment feature cluster vector, and recommend financing products according to the financing credit assessment result. The financing credit assessment on all customers in the customer cluster includes: receiving the financing assessment feature cluster vector of the customer cluster; performing residual convolution calculation on the financing assessment feature cluster vector to generate a convolution feature map of the financing assessment feature cluster vector; performing maximum pooling processing on the convolution feature map, and mapping the pooled convolution feature map to the financing credit assessment result.

2. A method for processing information based on multi-structure digital financial services as claimed in claim 1, characterized in that: The financing information collected from N customers is ； in: Indicates the collected The financing information of the customers mentioned above, Respectively represent Financing data series of each client for each of the above mentioned financing indicators; Indicates Customers in The financing data sequence of each of the financing indicators; Represents the financing data sequence middle Financing data for the month, Represents the financing data sequence Middle Financing data for the month.

3. A method for processing information based on multi-structure digital financial services as claimed in claim 2, characterized in that: In step S1, the process of preprocessing the financing information is as follows: S11: Construct filters for different financing indicators, where The filter of the financing indicator is , the filter The filter coefficient is: , using the filter to filter the corresponding financing data sequence: ; in: Represents the financing data sequence The filtering result of Indicates the financing data The filtering result of S12: Calculate the Fourier coefficients of the financing data sequence at H harmonic frequencies after different filtering processes: ； in: is the financing data sequence after filtering In the The Fourier coefficients at the harmonic frequencies, ; represents transpose; Represents the financing data sequence after filtering The Fourier coefficient matrix of ; is the harmonic matrix; For the Fourier coefficient matrix Solve it, the solution formula is: ; S13: Using the Fourier coefficient matrix obtained by solving, the financing data sequence after filtering is subjected to harmonic filtering in a form-preserving manner, and the harmonic filtering result is used as the preprocessing result of the financing data sequence, wherein the financing data sequence after filtering is The harmonic filtering formula is: ； in: is the financing data sequence after filtering The harmonic filtering result is, for The harmonic filtering result; S14: Constructing the pre-processed customer financing information: ； in: No. the pre-processed customer financing information of the customers, Respectively represent The financing data sequence of the customers after preprocessing of each of the financing indicators.

4. A method for processing information based on multi-structure digital financial services as claimed in claim 3, characterized in that: The S2 step includes: S21: Obtaining the pre-processed customer financing information The financing data sequence after preprocessing of each of the financing indicators; S22: using a time series decomposition method to obtain a time series correlation matrix of the financing data sequence after preprocessing; S23: Using the time series fusion module to iterate and fuse the time series state of the time series correlation matrix to obtain the pre-processed customer financing information The financing assessment feature vector of the corresponding customer : ； in: represents the L2 norm; Representing the time series features in the time series correlation matrix; represents the trend feature convolution matrix, Represents the convolution operator.

5. A method for processing information based on multi-structure digital financial services as claimed in claim 4, characterized in that: The step S22 includes: S221: Initialize and generate the financing data sequence after preprocessing The trend component , set the financing data sequence after preprocessing The current decomposition number is d, and the initial value of d is 0. After the dth time series decomposition, the financing data sequence The trend component of is: ,in Represents the trend component The mth component value in ; S222: Extracting the financing data sequence after preprocessing The periodic component : ； in: Represents the sequence smoothing function; represents an exponential function with a natural constant as base; The number of window sequence points representing the smoothing window; represents the difference parameter; S223: Decompose the trend component iteratively: ； in: Represents the periodic component The Mth component value in ; S224: Calculate the residual of the current d+1th decomposition : ； like If the decomposition value is less than the preset decomposition threshold, the iteration is terminated, and the trend component and the period component of the current d+1th decomposition are extracted as the financing data sequence after preprocessing. The timing characteristics of ,in , , represents the L2 norm; Otherwise, , return to step S222; S225: Construct the financing data sequence after preprocessing The time series correlation matrix: ； in: Represents the financing data sequence after preprocessing The time series correlation matrix.

6. A method for processing information based on multi-structure digital financial services as claimed in claim 4, characterized in that: The S3 step includes: S31: Calculate the distance between the financing evaluation feature vectors of different customers: ； in: Indicates Customers and The distance between the financing evaluation feature vector of the customers, ; S32: Based on the distances between the financing evaluation feature vectors of different customers, the K-means algorithm is used to perform clustering processing on the financing evaluation feature vectors to obtain cluster centers, among which The cluster centers are , ; S33: generating Gaussian mixture parameters corresponding to each of the cluster centers; S34: Calculate and obtain different financing evaluation feature vectors to be assigned to The probability of the cluster center; The financing evaluation feature vector is assigned to the cluster center with the highest probability, forming customer clusters; Generate financing evaluation feature cluster vectors for customer clusters, where The financing evaluation feature cluster vector of customer clusters is .

7. A method for processing information based on multi-structure digital financial services as claimed in claim 6, characterized in that: The generating of the financing assessment feature cluster vector of the customer cluster includes: Get the The belonging probability of each financing assessment feature vector in a customer cluster, wherein the belonging probability of the cluster center is 1; For All the belonging probabilities in the customer clusters are normalized so that the sum of all the belonging probabilities after the normalization is 1; The standardized attribution probability is used as the weight of the corresponding financing evaluation feature vector or the cluster center, and weighted processing is performed, and the weighted processing result is used as the first The financing assessment feature cluster vector of customer clusters.

8. A method for processing information based on multi-structure digital financial services as claimed in claim 1, characterized in that: The financing assessment model includes an input layer, a residual convolution layer, a maximum pooling layer and a financing credit assessment layer; The input layer is used to receive the financing assessment feature cluster vector of the customer cluster; The residual convolution layer is a residual learning unit structure in ResNet, which is used to perform residual convolution calculation on the financing evaluation feature cluster vector to generate a convolution feature map of the financing evaluation feature cluster vector; The maximum pooling layer is a pooling layer structure in a convolutional neural network, which is used to perform maximum pooling processing on the convolution feature map to form a pooled convolution feature map; The financing credit assessment layer is a fully connected structure in a neural network, which is used to map the pooled convolutional feature map into a financing credit assessment result. The financing credit assessment result is displayed in the form of a score on a percentage basis. The higher the score, the greater the customer's capital demand and the higher the repayment ability, and thus recommends financing products with higher capital support.

9. A method for processing information based on multi-structure digital financial services as claimed in claim 8, characterized in that: The financing credit assessment of all customers in the customer cluster includes: S41: The input layer receives The financing evaluation feature cluster vector of customer clusters ; S42: The residual convolution layer performs the calculation on the financing assessment feature cluster vector Perform residual convolution calculation to generate the financing evaluation feature cluster vector The convolutional feature map: ； in: Represents the financing evaluation feature cluster vector The convolution feature map of ; represents the residual information extraction matrix; S43: The maximum pooling layer performs pooling on the convolution feature map Perform maximum pooling processing to form the pooled convolution feature map ; ； in: Represents the maximum pooling operation; S44: The financing credit assessment layer converts the pooled convolution feature map Mapped to the financing credit assessment result: ； in: Indicates The financing credit evaluation results of all customers in a customer cluster; represents the collection of financing credit assessment results, , The financing credit assessment result set Any evaluation result in Indicates the evaluation result The corresponding mapping matrix is, Represents the financing credit evaluation result set Select the Achieve the maximum evaluation result As .

10. A multi-structure digital financial service information processing system, characterized in that: The system comprises: A data processing module, used to collect and pre-process customer financing information to obtain pre-processed customer financing information; the financing information includes at least four financing indicators, including bank transaction records, income level, liabilities and credit rating information; The customer behavior identification module is used to analyze the financing behavior of the pre-processed customer financing information, extract the financing evaluation feature vector, identify the customer groups according to the extracted financing evaluation feature vector, divide the customers into different customer clusters, and generate the financing evaluation feature cluster vector of the customer cluster; A financing service device is used to construct a financing evaluation model, perform financing credit evaluation on all customers in a customer cluster using the financing evaluation feature cluster vector of the customer cluster as input, and recommend financing products based on the financing credit evaluation results, so as to implement a multi-structure digital financial service information processing method as described in any one of claims 1 to 9.

Images