Anti-counterfeiting decision-making method, device and equipment for credit enhancement component and medium

Abstract

The embodiment of the invention relates to the technical field of data processing, and discloses an anti-counterfeiting decision-making method, device and equipment for a credit increasing component, and a medium, and the method comprises the steps: obtaining features corresponding to the data of a credit increasing component to be processed, so as to obtain the features of the credit increasing component; and making a decision on the credit enhancement component features according to a credit enhancement analysis agent to obtain a target credit enhancement result. According to the anti-counterfeiting decision-making method of the credit enhancement component provided by the embodiment of the invention, the intelligence level and adaptability of the credit enhancement analysis intelligent agent are improved in an environment feedback manner by combining advanced technical means and experience of specialists in the professional field, so that continuously evolved counterfeiting threats are more effectively dealt with.

Description

Technical Field

[0001] The embodiments of the present invention relate to the field of data processing, and in particular to an anti-counterfeiting decision method, device, equipment and medium for a credit enhancement component. Background Art

[0002] In the field of financial counterfeiting and tampering detection, a variety of technologies and methods have emerged to ensure the security and credibility of financial information. The following are some of the main counterfeit detection technologies currently available:

[0003] Digital watermarking technology: Digital watermarking technology is a method of embedding information in documents, images or audio to verify their authenticity. By embedding a unique identifier, digital watermarking technology helps confirm the source and integrity of the file, providing a degree of protection against forgery.

[0004] Traditional machine learning methods: Traditional machine learning algorithms, such as support vector machines (SVM), decision trees, and random forests, are widely used to classify text and images. These methods can use historical data to train models to identify and classify different types of information.

[0005] Image and audio processing techniques: With image and audio processing techniques, operations such as spectrum analysis and Fourier transforms can be performed to detect unusual patterns in images and audio. These techniques perform well in dealing with some visual and auditory forgeries.

[0006] Rule-based detection systems: Some systems use a rule-based approach to detect by developing predetermined rules and thresholds. Such systems are adaptable to known counterfeit patterns and have a faster execution speed.

[0007] Deep learning methods: In recent years, deep learning techniques such as convolutional neural networks (CNN) and recurrent neural networks (RNN) have made significant progress in forgery detection. These deep learning methods can process complex image, audio, and text data, improving the ability to detect forgeries.

[0008] In summary, although there are some counterfeit detection technologies, existing technologies still have limitations when facing complex and advanced deep fake threats. Summary of the invention

[0009] The purpose of the present invention is to at least provide an anti-counterfeiting decision-making method, device, equipment and medium for a credit enhancement component, which can at least solve the problem that the existing technology still has limitations when facing complex and advanced deep counterfeiting threats, and at least achieve the beneficial effect of combining advanced technical means and the experience of experts in professional fields to improve the intelligence level and adaptability of the system, thereby more effectively responding to the evolving counterfeiting threats.

[0010] To solve the above technical problems, at least one embodiment of the present invention provides an anti-counterfeiting decision method for a credit enhancement component, the method comprising:

[0011] Obtaining features corresponding to the credit enhancement component data to be processed to obtain credit enhancement component features;

[0012] A decision is made on the features of the above-mentioned credit enhancement components according to the credit enhancement analysis agent to obtain a target credit enhancement result; wherein the above-mentioned credit enhancement analysis agent has a corresponding environmental feedback tool and adversarial optimization tool; the above-mentioned environmental feedback tool is used to assist the credit enhancement analysis agent in outputting the target credit enhancement result; the above-mentioned adversarial optimization tool is used to test the above-mentioned credit enhancement analysis agent.

[0013] At least one embodiment of the present invention further provides an anti-counterfeiting decision-making device for a credit enhancement component, the device comprising:

[0014] A feature acquisition module is used to acquire features corresponding to the credit enhancement component data to be processed, so as to obtain credit enhancement component features;

[0015] The result acquisition module is used to make decisions on the characteristics of the above-mentioned credit enhancement components according to the credit enhancement analysis agent to obtain the target credit enhancement result; wherein the above-mentioned credit enhancement analysis agent has corresponding environmental feedback tools and adversarial optimization tools; the above-mentioned environmental feedback tools are used to assist the credit enhancement analysis agent in outputting the target credit enhancement result; the above-mentioned adversarial optimization tools are used to test the above-mentioned credit enhancement analysis agent.

[0016] At least one embodiment of the present invention also provides an electronic device, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the anti-counterfeiting decision method of the credit enhancement component.

[0017] At least one embodiment of the present invention further provides a computer-readable storage medium storing a computer program, which implements the anti-counterfeiting decision method of the above-mentioned credit enhancement component when executed by a processor.

[0018] The anti-counterfeiting decision method of the credit enhancement component provided by the embodiment of the present invention improves the intelligence level and adaptability of the credit enhancement analysis agent by combining advanced technical means and the experience of experts in professional fields in a way of environmental feedback, thereby more effectively responding to the evolving counterfeit threats. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] One or more embodiments are exemplarily described by the pictures in the corresponding drawings, and these exemplary descriptions do not constitute limitations on the embodiments.

[0020] Figure 1 is a flow chart of an anti-counterfeiting decision method of a credit enhancement component provided by an embodiment of the present invention;

[0021] Figure 2 is a flow chart of an anti-counterfeiting decision method of a credit enhancement component provided by another embodiment of the present invention;

[0022] Figure 3 is a schematic diagram of an anti-counterfeiting decision-making device of a credit enhancement component provided by another embodiment of the present invention;

[0023] Figure 4 It is a schematic structural diagram of an electronic device provided by another embodiment of the present invention. DETAILED DESCRIPTION

[0024] In order to make the purpose, technical scheme and advantages of the embodiments of the present invention clearer, the embodiments of the present invention will be described in detail below in conjunction with the accompanying drawings. However, it can be understood by those skilled in the art that in the embodiments of the present invention, many technical details are proposed in order to enable the reader to better understand the present invention. However, even without these technical details and various changes and modifications based on the following embodiments, the technical scheme claimed in the present invention can also be implemented. The division of the following embodiments is for the convenience of description and should not constitute any limitation on the specific implementation of the present invention. The various embodiments can be combined and referenced with each other under the premise of no contradiction.

[0025] In order to facilitate understanding of the embodiments of the present invention, relevant contents about the anti-counterfeiting decision method of the credit enhancement component are first introduced here.

[0026] In the field of finance, a variety of technologies and methods have emerged to detect forgery and tampering to ensure the security and credibility of financial information. Forgery in the financial field is mainly deep fake tampering. Deep fake is a synthetic media production method based on deep learning technology. By using technologies such as generative adversarial networks (GANs) or adversarial attacks, realistic audio, video or images can be forged to make them look like real content. In the financial field, deep fakes may be used to forge screenshots of Alipay bills, WeChat bills, competitor credits, etc., thereby causing losses.

[0027] Here are some of the main current counterfeit detection technologies:

[0028] Digital watermarking technology: Digital watermarking technology is a method of embedding information in documents, images or audio to verify their authenticity. By embedding a unique identifier, digital watermarking technology helps confirm the source and integrity of the file, providing a degree of protection against forgery.

[0029] Traditional machine learning methods: Traditional machine learning algorithms, such as support vector machines (SVM), decision trees, and random forests, are widely used to classify text and images. These methods can use historical data to train models to identify and classify different types of information.

[0030] Image and audio processing techniques: With image and audio processing techniques, operations such as spectrum analysis and Fourier transforms can be performed to detect unusual patterns in images and audio. These techniques perform well in dealing with some visual and auditory forgeries.

[0031] Rule-based detection systems: Some systems use a rule-based approach to detect by developing predetermined rules and thresholds. Such systems are adaptable to known counterfeit patterns and have a faster execution speed.

[0032] Deep learning methods: In recent years, deep learning techniques such as convolutional neural networks (CNN) and recurrent neural networks (RNN) have made significant progress in forgery detection. These deep learning methods can process complex image, audio, and text data, improving the ability to detect forgeries.

[0033] The shortcomings of the prior art are as follows:

[0034] Digital watermarking technology: Some traditional forgery and tampering detection methods use digital watermarking technology to embed invisible identifiers in documents or images to prove their authenticity. However, these technologies may perform poorly in the face of complex deep forgeries.

[0035] Traditional machine learning methods: Traditional machine learning algorithms, such as support vector machines (SVM) or decision trees, are used to classify and identify multi-dimensional data such as text, images, and audio. However, the performance of these methods may be limited when dealing with large-scale, highly complex counterfeiting methods.

[0036] Image and audio processing techniques: Use image and audio processing techniques, such as Fourier transform, spectrum analysis, etc. to detect abnormal patterns in images and audio. These methods may be effective in some cases, but their ability to deal with deep fakes is limited.

[0037] Rule-based detection system: It uses predefined rules and thresholds for detection. It has certain adaptability to specific types of counterfeiting, but its recognition performance for new types of counterfeiting is weak.

[0038] Deep learning methods: In recent years, deep learning techniques have been gradually applied to the field of forgery detection, such as using convolutional neural networks (CNN) or recurrent neural networks (RNN) for image and audio processing. However, these methods may perform poorly in the absence of large-scale annotated data, and there is still room for improvement in the level of defense against deep forgeries.

[0039] In summary, although there are some counterfeit detection technologies, existing technologies still have limitations when facing complex and advanced deep fake threats.

[0040] In order to solve the above-mentioned technical problem that although there are some counterfeit detection technologies, the existing technologies still have limitations when facing complex and advanced deep counterfeit threats, the present invention proposes an anti-counterfeiting decision method for credit enhancement components, that is, deep counterfeit defense, wherein deep counterfeit defense refers to the use of a series of technologies and methods to resist and prevent the potential threats of deep counterfeit technology to information, identity and credit enhancement components. In the financial field, deep counterfeit defense aims to prevent false audio, video, images and other synthetic media from being used to forge financial information, identity verification documents or other key components of credit assessment. In this embodiment, the deep counterfeit defense method refers to the robustness of the credit enhancement analysis after being tested and updated by the adversarial optimization tool, and can defend against adversarial attacks and attacks based on GAN forged images.

[0041] The implementation details of the anti-counterfeiting decision method of the credit enhancement component of this embodiment are specifically described below. The following content is only the implementation details provided for the convenience of understanding and is not necessary for the implementation of this solution.

[0042] The anti-counterfeiting decision method of the credit enhancement component of this embodiment can be applied to electronic devices with communication, computing and data storage capabilities. The specific process can be as follows: Figure 1 As shown, including:

[0043] Step 101, obtaining features corresponding to the credit enhancement component data to be processed to obtain credit enhancement component features.

[0044] Specifically, the credit enhancement component to be processed is a financial credit enhancement component, wherein the financial credit enhancement component refers to the proof materials such as bank statements, Alipay bills, WeChat bills, competitive product credits, etc. that users actively provide to credit companies in order to increase the loan amount or reduce the loan interest rate in the field of financial credit. In this embodiment, the credit enhancement component to be processed mainly refers to screenshots of Alipay bills, WeChat bills, competitive product credits, etc. The features corresponding to the credit enhancement component data to be processed are extracted to obtain the credit enhancement component features.

[0045] It should be noted that the acquisition or use of the above data requires the consent of the user corresponding to the credit enhancement component to be processed, and the above data can only be obtained after the user's authorization.

[0046] Step 102, making decisions on the features of the above-mentioned credit enhancement components according to the credit enhancement analysis agent to obtain the target credit enhancement result; wherein the above-mentioned credit enhancement analysis agent has corresponding environmental feedback tools and adversarial optimization tools; the above-mentioned environmental feedback tools are used to assist the credit enhancement analysis agent in outputting the target credit enhancement result; the above-mentioned adversarial optimization tools are used to test the above-mentioned credit enhancement analysis agent.

[0047] Specifically, in this embodiment, the credit enhancement analysis agent is a deep neural network containing a multi-layer transformer network. The credit enhancement analysis agent deeply encodes the input credit enhancement components to be processed (including competitive product pictures, bank statements, WeChat bills, etc.), generates component features, and then makes decisions to obtain credit enhancement results (i.e., changes to the user's price and credit limit).

[0048] In addition, in this embodiment, the environmental feedback tool is formed according to the external environment in which the credit enhancement analysis agent is located, and is used to provide feedback on the behavior of the credit enhancement analysis agent.

[0049] This embodiment, by combining advanced technical means and the experience of experts in professional fields, improves the intelligence level and adaptability of the credit enhancement analysis agent in the form of environmental feedback, so as to more effectively respond to the evolving counterfeit threats.

[0050] In some embodiments, before obtaining the features corresponding to the credit enhancement component data to be processed to obtain the credit enhancement component features, the method further includes:

[0051] Step a1, obtain the initial credit enhancement component data.

[0052] Step a2, pre-processing the above-mentioned initial credit enhancement component data to obtain credit enhancement component data to be processed.

[0053] In this embodiment, after obtaining the initial credit enhancement component data, the initial credit enhancement component data is preprocessed, and the preprocessing includes: data cleaning, denoising and other operations to ensure data quality.

[0054] In some embodiments, the above step 101 includes:

[0055] Step 1011, obtaining the data type of the enhancement component data to be processed; wherein the above data type includes: text data, image data and audio data.

[0056] Step 1012: Obtain the credit enhancement component features corresponding to the credit enhancement component data to be processed of each data type according to the feature extraction method corresponding to each data type.

[0057] Specifically, the bag-of-words model and TF-IDF algorithm are used for feature extraction for text data; the convolutional neural network (CNN) is used for feature extraction for image data; and the Fourier transform is used for feature extraction for audio data.

[0058] In some embodiments, the above step 102 includes:

[0059] Step 1021, input the above-mentioned credit enhancement component characteristics into the credit enhancement analysis agent to obtain an initial credit enhancement result.

[0060] Specifically, the credit enhancement analysis agent makes decisions based on the characteristics of the credit enhancement components and obtains initial credit enhancement results (such as adjustments to the user's price or credit limit).

[0061] Step 1022: Update the initial enhancement result according to the environmental feedback corresponding to the environmental feedback tool to obtain an intermediate enhancement result.

[0062] Specifically, first, the above-mentioned environmental feedback tool is controlled to obtain the user information corresponding to the credit enhancement component data to be processed; wherein the above-mentioned user information includes: user basic information, user behavior information and user risk information. Secondly, the above-mentioned user information is input into the preset evaluation tool to obtain environmental feedback. The initial credit enhancement result is updated according to the obtained environmental feedback to obtain the intermediate credit enhancement result. As an example: based on user information and behavioral data, the environmental feedback tool assesses that the user's credit risk is high, and the credit enhancement analysis agent reduces the corresponding user's credit enhancement limit based on the feedback; conversely, if the user's credit is assessed to be good, the credit enhancement limit will be increased.

[0063] It should be noted that the acquisition or use of the above data (user basic information, user behavior information and user risk information) requires the consent of the user corresponding to the credit enhancement component to be processed, and the above data can only be obtained after the user's authorization.

[0064] Step 1023: Update the credit enhancement analysis agent according to the adversarial optimization tool to obtain the target credit enhancement result.

[0065] Specifically, the role of the adversarial optimization tool is mainly to simulate external factors that may have a negative impact on the credit enhancement analysis agent, such as malicious users or fraudulent behavior. Through a set of machine learning algorithms, the adversarial optimization tool generates various possible negative factors and verifies the robustness of the credit enhancement analysis agent, thereby helping the credit enhancement analysis agent optimize its learning strategy.

[0066] First, the adversarial optimization tool is controlled to generate several adversarial test conditions. Second, according to the several adversarial test conditions, the credit enhancement analysis agent is updated to obtain the target credit enhancement result. As an example: the adversarial optimization tool simulates some unconventional user behaviors or abnormal transaction patterns to test and challenge whether the strategy of the credit enhancement analysis agent can make the right decision in this situation.

[0067] In some embodiments, after controlling the adversarial optimization tool to generate a plurality of adversarial test conditions, the method further includes:

[0068] First, the above-mentioned environmental feedback is updated according to the above-mentioned adversarial test conditions; then, the above-mentioned initial enhancement result is updated according to the above-mentioned environmental feedback to obtain the target enhancement result.

[0069] Specifically, the adversarial optimization tool can also update the environmental feedback. As an example, the adversarial optimization tool simulates some unconventional user behaviors or abnormal transaction patterns to test whether the environmental feedback tool can accurately feedback the risk level of these behaviors.

[0070] The environmental analysis and adversarial simulation provided in this application refers to training and testing the decision model by simulating user behavior and market environment. The environmental module can provide feedback to help the decision model understand how it performs in the actual environment, while the adversarial module simulates possible risks and attacks to test the robustness of the decision model. This enables the credit enhancement analysis agent to self-learn and optimize according to the environmental feedback tool and adversarial optimization tool, continuously improve its decision results, and improve the accuracy of the decision.

[0071] The anti-counterfeiting decision method of the credit enhancement component provided by the present invention is described below through a specific embodiment: Figure 2 As shown, the score evaluation system (environmental feedback tool) and the adversarial generation system (adversarial optimization tool) jointly optimize the Agent (trust enhancement analysis agent) to make the decision result more accurate.

[0072] Another embodiment of the present invention relates to an anti-counterfeiting decision-making device for a credit enhancement component. The implementation details of the anti-counterfeiting decision-making of the credit enhancement component of this embodiment are specifically described. The following content is only for the convenience of understanding the implementation details provided, and is not necessary for the implementation of this solution. The schematic diagram of the anti-counterfeiting decision-making device of the credit enhancement component of this embodiment can be as follows: Figure 3 As shown, it includes a feature acquisition module 301 and a result acquisition module 302. Specifically,

[0073] The feature acquisition module 301 is used to acquire features corresponding to the credit enhancement component data to be processed to obtain credit enhancement component features.

[0074] The result acquisition module 302 is used to make decisions on the above-mentioned credit enhancement component characteristics according to the credit enhancement analysis agent to obtain the target credit enhancement result; wherein the above-mentioned credit enhancement analysis agent has corresponding environmental feedback tools and adversarial optimization tools; the above-mentioned environmental feedback tools are used to assist the credit enhancement analysis agent in outputting the target credit enhancement result; the above-mentioned adversarial optimization tools are used to test the above-mentioned credit enhancement analysis agent.

[0075] It is worth mentioning that all modules involved in this embodiment are logic modules. In practical applications, a logic unit can be a physical unit, a part of a physical unit, or a combination of multiple physical units. In addition, in order to highlight the innovative part of the present invention, this embodiment does not introduce units that are not closely related to solving the technical problem proposed by the present invention, but this does not mean that there are no other units in this embodiment.

[0076] Another embodiment of the present invention relates to an electronic device, such as Figure 4 As shown, it includes: at least one processor 401; and a memory 402 that is communicatively connected to the at least one processor 401; wherein the memory 402 stores instructions that can be executed by the at least one processor 401, and the instructions are executed by the at least one processor 401, so that the at least one processor 401 can execute the anti-counterfeiting decision method of the enhancement component in the above-mentioned embodiments.

[0077] Among them, the memory and the processor are connected in a bus manner, and the bus may include any number of interconnected buses and bridges, and the bus connects various circuits of one or more processors and memories together. The bus can also connect various other circuits such as peripherals, voltage regulators, and power management circuits, which are well known in the art and are therefore not further described herein. The bus interface provides an interface between the bus and the transceiver. The transceiver can be one element or multiple elements, such as multiple receivers and transmitters, providing a unit for communicating with various other devices on a transmission medium. The data processed by the processor is transmitted on a wireless medium via an antenna, and further, the antenna also receives data and transmits the data to the processor.

[0078] The processor is responsible for managing the bus and general processing, and can also provide various functions, including timing, peripheral interfaces, voltage regulation, power management, and other control functions. Memory can be used to store data used by the processor when performing operations.

[0079] Another embodiment of the present invention relates to a computer-readable storage medium storing a computer program, which implements the above method embodiment when executed by a processor.

[0080] That is, those skilled in the art can understand that all or part of the steps in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a program, and the program is stored in a storage medium, including a number of instructions for a device (which can be a single-chip microcomputer, chip, etc.) or a processor (processor) to perform all or part of the steps of the above-mentioned method of each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, referred to as: ROM), random access memory (Random Access Memory, referred to as: RAM), disk or optical disk and other media that can store program codes.

[0081] Those skilled in the art will appreciate that the above embodiments are specific embodiments for implementing the present invention, and that in actual applications, various changes may be made in form and detail without departing from the spirit and scope of the present invention.

Claims

1. A method for anti-counterfeiting decision-making of a credit enhancement component, characterized in that: The method comprises: Obtaining features corresponding to the credit enhancement component data to be processed to obtain credit enhancement component features; A decision is made on the characteristics of the credit enhancement component according to the credit enhancement analysis agent to obtain a target credit enhancement result; wherein the credit enhancement analysis agent has a corresponding environmental feedback tool and an adversarial optimization tool; the environmental feedback tool is used to assist the credit enhancement analysis agent in outputting the target credit enhancement result; and the adversarial optimization tool is used to test the credit enhancement analysis agent.

2. The anti-counterfeiting decision method of the credit enhancement component according to claim 1 is characterized in that: Before acquiring the features corresponding to the credit enhancement component data to be processed to obtain the credit enhancement component features, the method further includes: Obtain initial credit enhancement component data; The initial credit enhancement component data is preprocessed to obtain credit enhancement component data to be processed.

3. The anti-counterfeiting decision method of the credit enhancement component according to claim 1 is characterized in that: The obtaining of features corresponding to the credit enhancement component data to be processed to obtain credit enhancement component features includes: Obtaining the data type of the credit enhancement component data to be processed; wherein the data type includes: text data, image data and audio data; According to the feature extraction method corresponding to each data type, the credit enhancement component features corresponding to the credit enhancement component data to be processed of each data type are obtained.

4. The anti-counterfeiting decision method of the credit enhancement component according to claim 1, characterized in that: The step of obtaining a target credit enhancement result according to the credit enhancement component characteristics and the credit enhancement analysis agent includes: Inputting the credit enhancement component characteristics into the credit enhancement analysis agent to obtain an initial credit enhancement result; Updating the initial enhancement result according to the environmental feedback corresponding to the environmental feedback tool to obtain an intermediate enhancement result; The credit enhancement analysis agent is updated according to the adversarial optimization tool to obtain the target credit enhancement result.

5. The method for preventing counterfeiting of a credit enhancement component according to claim 4, characterized in that: The environmental feedback is generated according to the following steps: Controlling the environmental feedback tool to obtain user information corresponding to the credit enhancement component data to be processed; wherein the user information includes: basic user information, user behavior information and user risk information; The user information is input into a preset evaluation tool to obtain environmental feedback.

6. The method for preventing counterfeiting of a credit enhancement component according to claim 5, characterized in that: The updating of the credit enhancement analysis agent according to the adversarial optimization tool to obtain a target credit enhancement result includes: Controlling the adversarial optimization tool to generate a number of adversarial test conditions; According to the several adversarial test conditions, the confidence enhancement analysis agent is updated to obtain the target confidence enhancement result.

7. The method for preventing counterfeiting of a credit enhancement component according to claim 6, characterized in that: After controlling the adversarial optimization tool to generate a plurality of adversarial test conditions, the method further includes: updating the environmental feedback according to a number of the adversarial test conditions; The initial enhancement result is updated according to the environmental feedback to obtain a target enhancement result.

8. An anti-counterfeiting decision-making device for a credit enhancement component, characterized in that: The device comprises: A feature acquisition module is used to acquire features corresponding to the credit enhancement component data to be processed, so as to obtain credit enhancement component features; A result acquisition module is used to make decisions on the characteristics of the credit enhancement components according to the credit enhancement analysis agent to obtain a target credit enhancement result; wherein the credit enhancement analysis agent has corresponding environmental feedback tools and adversarial optimization tools; the environmental feedback tool is used to assist the credit enhancement analysis agent in outputting the target credit enhancement result; the adversarial optimization tool is used to test the credit enhancement analysis agent.

9. An electronic device, characterized in that: include: at least one processor; as well as, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the anti-counterfeiting decision method of the enhancement component as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the anti-counterfeiting decision method of the credit enhancement component described in any one of claims 1 to 7 is implemented.

Images