Digital economic transaction processing and anomaly detection method based on machine learning

Abstract

The invention relates to the technical field of digital information processing, in particular to a digital economic transaction processing and anomaly detection method based on machine learning. The method comprises the following steps: carrying out digital transaction data acquisition and weighted feature analysis processing on a digital economic transaction platform to generate weighted digital transaction feature data; analyzing digital transaction strategy prediction data based on the weighted digital transaction feature data; performing abnormal transaction confrontation game feature analysis according to the digital transaction strategy prediction data to generate abnormal transaction confrontation game feature data; establishing a digital transaction fraud tracing tree model based on the abnormal transaction confrontation game feature data; and performing digital transaction abnormal feedback data analysis processing on the weighted digital transaction feature data based on the digital transaction fraud tracing tree model to generate digital transaction abnormal feedback data. According to the invention, by predicting variable and complex fraudulent transaction behaviors in digital transaction behaviors, efficient digital economic transaction anomaly detection is realized.

Description

Technical Field

[0001] The present invention relates to the field of digital information processing technology, and in particular to a digital economic transaction processing and anomaly detection method based on machine learning. Background Art

[0002] With the rapid development of the digital economy, transaction activities in the fields of e-commerce, online payment, digital finance, etc. have increased significantly. Accompanying this is the surge and complexity of transaction data, which often contains various forms of information, such as transaction amount, time, location, participant information, product description, user behavior, etc. Due to the diversity and complexity of transaction behaviors, the fraud methods of the digital economy are constantly changing. In order to avoid the anomaly detection or tracking of fraud methods in the digital economy, it is necessary to continuously detect the anomalies of digital economic transactions through intelligent technical means to achieve a good development of the digital economic transaction environment. However, traditional digital economic transaction processing and anomaly detection methods rely on predefined rules or thresholds to execute abnormal digital transaction detection methods. Abnormal digital transaction detection methods are based on past experience and known fraud patterns. With the continuous escalation of fraud methods, fraudsters can design more and more complex and hidden abnormal behaviors to avoid the tracking of traditional abnormal digital transaction detection methods, making it difficult to adjust in time to adapt to the abnormal changes in digital transactions. Summary of the invention

[0003] Based on this, the present invention provides a digital economic transaction processing and anomaly detection method based on machine learning to solve at least one of the above-mentioned technical problems.

[0004] To achieve the above purpose, a digital economic transaction processing and anomaly detection method based on machine learning includes the following steps:

[0005] Step S1: collecting digital transaction multimodal data from a digital economic transaction platform according to a preset digital economic transaction monitoring engine to obtain digital transaction multimodal data; performing digital transaction feature analysis and processing on the digital transaction multimodal data to generate digital transaction feature data;

[0006] Step S2: Designing digital transaction attention weight information based on digital transaction multimodal data; performing attention weight weighting processing on digital transaction feature data according to the digital transaction attention weight information to generate weighted digital transaction feature data;

[0007] Step S3: Acquire historical digital trading strategies; establish a digital trading strategy reinforcement learning model based on the historical digital trading strategies; transmit the weighted digital trading feature data to the digital trading strategy reinforcement learning model for digital trading strategy prediction processing to generate digital trading strategy prediction data;

[0008] Step S4: Performing feature derivative analysis and processing of multi-dimensional digital transaction fraud according to the digital transaction strategy prediction data to generate multi-dimensional digital transaction fraud feature derivative data; designing a confrontation game mathematical model of abnormal transactions based on the multi-dimensional digital transaction fraud feature derivative data to generate an abnormal transaction confrontation game model;

[0009] Step S5: performing abnormal transaction adversarial game feature analysis and processing according to the abnormal transaction adversarial game model to generate abnormal transaction adversarial game feature data; establishing a preliminary digital transaction fraud tracing tree model based on the abnormal transaction adversarial game feature data; optimizing the tree structure weight parameters of the preliminary digital transaction fraud tracing tree model according to the digital transaction attention weight information corresponding to the weighted digital transaction feature data to obtain the digital transaction fraud tracing tree model;

[0010] Step S6: Analyze and process the weighted digital transaction feature data based on the digital transaction fraud tracing tree model to generate digital transaction abnormality feedback data; transmit the digital transaction abnormality feedback data to the terminal to execute the digital transaction abnormality feedback operation.

[0011] The present invention collects digital transaction multimodal data through a preset digital economic transaction monitoring engine to ensure that the collected data covers all important aspects of the transaction platform, including multiple dimensions such as user behavior, transaction time, and transaction amount. This multimodal data collection improves the comprehensiveness of the data and helps to capture complex transaction patterns and potential abnormal behaviors. The collected multimodal data is analyzed and processed for digital transaction features to ensure that the key features in the data are effectively extracted, laying a solid foundation for subsequent analysis and processing, and helping to improve the accuracy of abnormal detection. Digital transaction attention weight information is designed through digital transaction multimodal data, and attention weight weighting processing is performed on digital transaction feature data based on the information, which can highlight the attention to specific important features in the transaction data, thereby improving the learning effect of subsequent models on these key features and enhancing the accuracy and reliability of prediction. The historical digital transaction strategy is obtained, and a digital transaction strategy reinforcement learning model is established based on this, effectively utilizing the lessons learned from historical data to optimize the current transaction strategy model. The weighted digital transaction feature data is transmitted to the reinforcement learning model for strategy prediction, and the behavior combination of digital transactions of the weighted digital transaction feature data can be predicted through the model to improve the effectiveness of transaction decisions. Based on the digital transaction strategy prediction data, multi-dimensional digital transaction fraud feature derivative analysis and processing are carried out to deeply explore potential fraud behavior patterns, provide a rich fraud feature foundation for subsequent anomaly detection, and ensure that the anomaly detection model can identify complex fraud behaviors. Designing an abnormal transaction adversarial game mathematical model based on multi-dimensional digital transaction fraud feature derivative data can improve the robustness of digital transaction anomaly detection, making it more adversarial to abnormal transaction fraud behaviors, thereby more effectively dealing with complex fraud behaviors. Through the abnormal transaction adversarial game model, abnormal transaction adversarial game feature analysis and processing are carried out, and the behavior patterns of different participants in the transaction process are dynamically analyzed. The generated abnormal transaction adversarial game feature data provides an important basis for tracing fraud behaviors. Based on the abnormal transaction adversarial game feature data and attention weight information, the preliminary digital transaction fraud tracing tree model is gradually optimized to obtain a more accurate and efficient fraud tracing tree model. This tree model can efficiently trace and identify abnormal transactions in complex transaction networks, providing strong support for fraud detection in the digital economic environment. Based on the optimized digital transaction fraud tracing tree model, the weighted digital transaction feature data is analyzed and processed to generate effective digital transaction anomaly feedback data. After being transmitted to the terminal, these feedback data can be used to implement feedback measures for abnormal transactions, thereby forming a closed-loop feedback system, continuously improving and optimizing the transaction anomaly detection mechanism, and improving the security and reliability of the entire system.

[0012] The beneficial effect of the present application is that the digital economic transaction processing and anomaly detection method based on machine learning of the present invention can adaptively adjust digital transaction strategies and detection methods by introducing machine learning models, especially reinforcement learning models. Unlike traditional methods that rely on predefined rules and thresholds, machine learning models extract features from historical data and predict future transaction behaviors through continuous learning and optimization. This dynamic adaptability enables the system to promptly discover and respond to emerging fraud patterns, no longer limited to past experience and known fraud patterns. By designing digital transaction attention weight information and adversarial game mathematical models, it can handle more complex and hidden abnormal behaviors. The weighted processing of attention weights enables the system to concentrate resources to analyze abnormal transaction features, thereby improving the accuracy and efficiency of anomaly detection. The introduction of adversarial game models enables the system to simulate and confront potential fraud strategies, identify complex fraud behaviors that circumvent traditional detection methods, and continuously optimize the model in the process of anomaly detection. By optimizing the weight parameters of the preliminary digital transaction fraud tracing tree model, the accuracy and effectiveness of the model are continuously improved. The ability to continuously learn and optimize keeps pace with the evolving fraud behaviors and gradually enhances the detection capabilities of new and complex digital transaction anomalies. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 A schematic diagram of the steps of a method for digital economic transaction processing and anomaly detection based on machine learning according to the present invention;

[0014] Figure 2 for Figure 1 Detailed implementation steps of step S3 in FIG.

[0015] Figure 3 for Figure 1 Detailed implementation steps of step S5;

[0016] 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

[0017] The technical method of the present invention is described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by technicians in this field without creative work are within the scope of protection of the present invention.

[0018] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0019] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0020] To achieve this, please refer to Figures 1 to 3 The present invention provides a digital economic transaction processing and anomaly detection method based on machine learning, comprising the following steps:

[0021] Step S1: collecting digital transaction multimodal data from a digital economic transaction platform according to a preset digital economic transaction monitoring engine to obtain digital transaction multimodal data; performing digital transaction feature analysis and processing on the digital transaction multimodal data to generate digital transaction feature data;

[0022] Step S2: designing digital transaction attention weight information based on digital transaction multimodal data; performing attention weight weighting processing on the digital transaction feature data according to the digital transaction attention weight information to generate weighted digital transaction feature data;

[0023] Step S3: Acquire historical digital trading strategies; establish a digital trading strategy reinforcement learning model based on the historical digital trading strategies; transmit the weighted digital trading feature data to the digital trading strategy reinforcement learning model for digital trading strategy prediction processing to generate digital trading strategy prediction data;

[0024] Step S4: Performing feature derivative analysis and processing of multi-dimensional digital transaction fraud according to the digital transaction strategy prediction data to generate multi-dimensional digital transaction fraud feature derivative data; designing a confrontation game mathematical model of abnormal transactions based on the multi-dimensional digital transaction fraud feature derivative data to generate an abnormal transaction confrontation game model;

[0025] Step S5: performing abnormal transaction adversarial game feature analysis and processing according to the abnormal transaction adversarial game model to generate abnormal transaction adversarial game feature data; establishing a preliminary digital transaction fraud tracing tree model based on the abnormal transaction adversarial game feature data; optimizing the tree structure weight parameters of the preliminary digital transaction fraud tracing tree model according to the digital transaction attention weight information corresponding to the weighted digital transaction feature data to obtain the digital transaction fraud tracing tree model;

[0026] Step S6: Analyze and process the weighted digital transaction feature data based on the digital transaction fraud tracing tree model to generate digital transaction abnormality feedback data; transmit the digital transaction abnormality feedback data to the terminal to execute the digital transaction abnormality feedback operation.

[0027] The present invention collects digital transaction multimodal data through a preset digital economic transaction monitoring engine to ensure that the collected data covers all important aspects of the transaction platform, including multiple dimensions such as user behavior, transaction time, and transaction amount. This multimodal data collection improves the comprehensiveness of the data and helps to capture complex transaction patterns and potential abnormal behaviors. The collected multimodal data is analyzed and processed for digital transaction features to ensure that the key features in the data are effectively extracted, laying a solid foundation for subsequent analysis and processing, and helping to improve the accuracy of abnormal detection. Digital transaction attention weight information is designed through digital transaction multimodal data, and attention weight weighting processing is performed on digital transaction feature data based on the information, which can highlight the attention to specific important features in the transaction data, thereby improving the learning effect of subsequent models on these key features and enhancing the accuracy and reliability of prediction. The historical digital transaction strategy is obtained, and a digital transaction strategy reinforcement learning model is established based on this, effectively utilizing the lessons learned from historical data to optimize the current transaction strategy model. The weighted digital transaction feature data is transmitted to the reinforcement learning model for strategy prediction, and the behavior combination of digital transactions of the weighted digital transaction feature data can be predicted through the model to improve the effectiveness of transaction decisions. Based on the digital transaction strategy prediction data, multi-dimensional digital transaction fraud feature derivative analysis and processing are carried out to deeply explore potential fraud behavior patterns, provide a rich fraud feature foundation for subsequent anomaly detection, and ensure that the anomaly detection model can identify complex fraud behaviors. Designing an abnormal transaction adversarial game mathematical model based on multi-dimensional digital transaction fraud feature derivative data can improve the robustness of digital transaction anomaly detection, making it more adversarial to abnormal transaction fraud behaviors, thereby more effectively dealing with complex fraud behaviors. Through the abnormal transaction adversarial game model, abnormal transaction adversarial game feature analysis and processing are carried out, and the behavior patterns of different participants in the transaction process are dynamically analyzed. The generated abnormal transaction adversarial game feature data provides an important basis for tracing fraud behaviors. Based on the abnormal transaction adversarial game feature data and attention weight information, the preliminary digital transaction fraud tracing tree model is gradually optimized to obtain a more accurate and efficient fraud tracing tree model. This tree model can efficiently trace and identify abnormal transactions in complex transaction networks, providing strong support for fraud detection in the digital economic environment. Based on the optimized digital transaction fraud tracing tree model, the weighted digital transaction feature data is analyzed and processed to generate effective digital transaction anomaly feedback data. After being transmitted to the terminal, these feedback data can be used to implement feedback measures for abnormal transactions, thereby forming a closed-loop feedback system, continuously improving and optimizing the transaction anomaly detection mechanism, and improving the security and reliability of the entire system.

[0028] As an embodiment of the present invention, refer to Figure 1The above is a schematic diagram of the steps of a method for processing and detecting anomalies of digital economic transactions based on machine learning in the present invention. In this embodiment, the method for processing and detecting anomalies of digital economic transactions based on machine learning includes the following steps:

[0029] Step S1: collecting digital transaction multimodal data from a digital economic transaction platform according to a preset digital economic transaction monitoring engine to obtain digital transaction multimodal data; performing digital transaction feature analysis and processing on the digital transaction multimodal data to generate digital transaction feature data;

[0030] In an embodiment of the present invention, a digital economic transaction monitoring engine is deployed on a digital economic transaction platform, such as an online e-commerce platform, an electronic transaction platform, etc. The engine can monitor and collect transaction information from multiple data sources in real time, including user behavior data (such as click streams, browsing history), transaction data (such as order amount, payment method), time and geographic location data, etc. Through the monitoring engine, multimodal data collection is performed on all transaction activities on the platform, and the collected data includes structured data (such as transaction amount, user ID) and unstructured data (such as product description, user comments). The collected data is uniformly stored in a data warehouse for subsequent analysis. The collected multimodal data is preprocessed, including data cleaning, deduplication and formatting, and feature extraction is performed on the cleaned data. According to different transaction behavior patterns, corresponding data feature methods are designed to extract features such as user purchase frequency, transaction time interval, volatility of transaction amount, etc., and these features are preliminarily analyzed using data clustering algorithms to identify typical transaction patterns and behavior features, and convert them into structured feature data to generate digital transaction feature data for subsequent steps.

[0031] Step S2: designing digital transaction attention weight information based on digital transaction multimodal data; performing attention weight weighting processing on the digital transaction feature data according to the digital transaction attention weight information to generate weighted digital transaction feature data;

[0032] In an embodiment of the present invention, the self-attention mechanism in deep learning is used to analyze the generated digital transaction feature data and design digital transaction attention weight information. The attention mechanism can identify the most influential features in the transaction data, such as the time period of high-frequency transactions, the abnormal consumption behavior of specific users, and the consumption frequency, and design corresponding attention weight information. The attention weight information is designed based on the relevant information that is most likely to cause digital transaction anomalies, and different weight values ​​are assigned to these features. According to the attention weight information, each feature data is weighted, and the weight coefficient corresponding to each feature data is identified to highlight the most important features. The weighted feature data will be able to provide more accurate results in subsequent model training and prediction, and generate weighted digital transaction feature data.

[0033] Step S3: Acquire historical digital trading strategies; establish a digital trading strategy reinforcement learning model based on the historical digital trading strategies; transmit the weighted digital trading feature data to the digital trading strategy reinforcement learning model for digital trading strategy prediction processing to generate digital trading strategy prediction data;

[0034] In an embodiment of the present invention, previous transaction strategies are extracted from historical transaction data, and these strategies are archived. Historical strategies include payment methods, purchasing behavior patterns, preferential usage, etc. adopted by users in different situations. A digital transaction strategy model is constructed using a reinforcement learning algorithm (such as deep Q learning). The state space of the model includes the user's behavioral characteristics, transaction environment characteristics, etc., and the action space includes transaction decisions that can be taken, such as recommending different products, adjusting payment methods, etc. The historical transaction strategies are input into the model for training. The model gradually learns and optimizes its strategy selection by simulating the execution effects of different strategies in different situations. The weighted digital transaction feature data is input into the trained reinforcement learning model, and the model predicts the strategy based on the current transaction status. The output of the model is a transaction prediction strategy for the current transaction scenario, such as recommending a certain payment method, triggering a certain preferential measure, etc. The generated strategy prediction data will be used as a basis and input into the next step for derivative analysis of fraud features.

[0035] Step S4: Performing feature derivative analysis and processing of multi-dimensional digital transaction fraud according to the digital transaction strategy prediction data to generate multi-dimensional digital transaction fraud feature derivative data; designing a confrontation game mathematical model of abnormal transactions based on the multi-dimensional digital transaction fraud feature derivative data to generate an abnormal transaction confrontation game model;

[0036] In the embodiment of the present invention, based on the digital transaction strategy prediction data, the Generative Adversarial Network (GAN) technology is used to perform multi-dimensional fraud feature derivative analysis, and the existing fraud pattern features are extracted by analyzing the user's transaction behavior and the predicted strategy data. For example, high-frequency small transactions, abnormal consumption within a specific time period, etc.

[0037] Using Generative Adversarial Network (GAN) technology, these features are combined and transformed to generate derivative data that can cover various potential abnormal transaction patterns. Based on the derivative data of multi-dimensional digital transaction fraud features, an abnormal transaction adversarial game model is constructed. The abnormal transaction adversarial game model simulates the fraud encountered in the transaction behavior and the corresponding response strategy. The game model continuously learns the derivative fraud behavior of the multi-dimensional digital transaction fraud feature derivative data and the corresponding fraud response method, optimizes the anomaly detection strategy, and enables the system to identify more complex and hidden fraud behaviors. The generated abnormal transaction adversarial game model includes the fraud behavior of digital transactions and the corresponding fraud behavior processing method, which will be further used for subsequent feature analysis and traceability tree model design.

[0038] Step S5: performing abnormal transaction adversarial game feature analysis and processing according to the abnormal transaction adversarial game model to generate abnormal transaction adversarial game feature data; establishing a preliminary digital transaction fraud tracing tree model based on the abnormal transaction adversarial game feature data; optimizing the tree structure weight parameters of the preliminary digital transaction fraud tracing tree model according to the digital transaction attention weight information corresponding to the weighted digital transaction feature data to obtain the digital transaction fraud tracing tree model;

[0039] In an embodiment of the present invention, the abnormal transaction behavior feature data generated in the game model is analyzed to extract key adversarial features, which represent the strategies adopted by fraudsters to evade detection, such as dispersing transaction amounts, operating in different locations, etc.

[0040] Based on the extracted adversarial features, a preliminary digital transaction fraud tracing tree model is designed. The nodes of the model represent different fraud behavior features, and the paths represent the correlation between the features. The digital transaction attention weight information corresponding to the weighted digital transaction feature data is combined with the adversarial feature data to optimize the weight parameters of the tracing tree model and adjust the weights of each node so that the model can more accurately reflect the evolution path of the actual fraud behavior. By gradually adjusting and optimizing the preliminary tracing tree model, a tuned digital transaction fraud tracing tree model is finally obtained. The model can dynamically adjust the priority of nodes and paths according to the weight information of different features to ensure that fraud can be quickly identified and traced when it occurs.

[0041] Step S6: Analyze and process the weighted digital transaction feature data based on the digital transaction fraud tracing tree model to generate digital transaction abnormality feedback data; transmit the digital transaction abnormality feedback data to the terminal to execute the digital transaction abnormality feedback operation.

[0042] In the embodiment of the present invention, the optimized digital transaction fraud tracing tree model is used to perform abnormal prediction processing on the weighted digital transaction feature data. The digital transaction fraud tracing tree model predicts the abnormal risk level of the current transaction according to the distribution and association of the weighted digital transaction feature data, such as analyzing the risks encountered in the current and future transaction decisions based on the current transaction feature data, and generating digital transaction abnormal prediction data. The digital transaction abnormal prediction data is compared with the preset abnormal evaluation standard to judge the accuracy of the abnormal prediction. During the evaluation process, the system considers a variety of factors, such as abnormal fluctuations in transaction amounts, abnormal changes in user behavior, etc., and converts them into abnormal quantitative indicators based on the evaluation standards to generate the final abnormal evaluation data. The abnormal evaluation data is mapped and integrated with the predicted data to generate comprehensive abnormal feedback data, summarize the analysis results of digital transaction anomalies and the corresponding quantitative indicators, and provide detailed explanations and suggestions for the predicted abnormal behavior. The generated abnormal feedback data is transmitted to the terminal system for further operation or automated response. The terminal can execute preset security measures according to the feedback information, such as freezing accounts, restricting transactions, sending alarms, etc., to ensure the security and stability of the transaction environment.

[0043] Preferably, step S1 comprises the following steps:

[0044] Step S11: collecting digital transaction multimodal data from the digital economic transaction platform according to a preset digital economic transaction monitoring engine to obtain digital transaction multimodal data;

[0045] Step S12: performing digital transaction behavior analysis based on the digital transaction multimodal data to generate digital transaction behavior data;

[0046] Step S13: performing cluster analysis on the digital transaction behavior data to generate clustered digital transaction behavior data;

[0047] Step S14: performing transaction pattern recognition processing based on the clustered digital transaction behavior data to generate digital transaction pattern data;

[0048] Step S15: Design a transaction pattern dimension reduction feature matrix based on the digital transaction pattern data to generate a transaction pattern dimension reduction feature matrix;

[0049] Step S16: Perform digital transaction feature mapping processing on the digital transaction multimodal data according to the transaction mode dimension reduction feature matrix to generate digital transaction feature data.

[0050] The present invention collects multimodal data on the digital economic transaction platform through a preset digital economic transaction monitoring engine, which can comprehensively and accurately obtain transaction data of different dimensions, provide a rich and diverse information source for subsequent data analysis, and help improve the comprehensiveness and reliability of anomaly detection. Behavioral analysis of digital transaction multimodal data helps to explore the deep-level characteristics of user transaction behavior. This analysis can identify implicit behavioral patterns, improve the ability to identify abnormal transaction behaviors, and more accurately capture abnormal signals. Cluster analysis of digital transaction behavior data can effectively group transaction behavior data and identify different transaction patterns. The clustering results provide more accurate input for subsequent pattern recognition and feature extraction, thereby improving the accuracy of anomaly detection. Transaction pattern recognition is performed based on clustering results to identify different behavior patterns in transaction data, help establish an understanding of different transaction behavior patterns, and thus be used to design dimensionality reduction features of data under each pattern. By designing a dimensionality reduction feature matrix for transaction pattern data, simplifying the data structure, reducing the computational complexity, and retaining key information, not only the processing speed of the system is improved, but also the sensitivity of the model to abnormal patterns is enhanced. The transaction pattern dimensionality reduction feature matrix is ​​mapped to the digital transaction multimodal data to accurately extract features related to anomaly detection in digital transactions. This mapping process enables the system to better capture abnormal features in transaction behavior and improve the effectiveness and accuracy of detection.

[0051] In an embodiment of the present invention, a preset digital economic transaction monitoring engine is deployed on a digital economic transaction platform. The engine is composed of multiple modules and can monitor and collect various types of transaction data from the platform in real time. The engine is configured to collect multimodal data, including but not limited to the following data types: transaction data (such as order amount, payment method), user behavior data (such as browsing history, click events), time data (such as transaction time, response time), geographic data (such as transaction location, IP address) and social data (such as user comments, interaction frequency). After the monitoring engine starts working, it continuously collects multimodal data for all transaction activities on the platform. The data collection process includes real-time capture of user operations on the platform, detailed information on transaction execution, and auxiliary information related to the transaction. The collected multimodal data is preprocessed, including data cleaning (removing noise and outliers), format standardization (unifying data format and units), data deduplication (eliminating duplicate data) and other steps. The preprocessed data ensures data quality and consistency, and provides a reliable basis for subsequent behavioral analysis to obtain digital transaction multimodal data. Use advanced data analysis techniques (such as time series analysis, frequency analysis, and trend analysis) to conduct in-depth behavioral analysis of multimodal data, focusing on analyzing users' transaction behavior patterns, such as users' purchase frequency, average transaction amount, purchase time period preference, and geographic location changes. By analyzing these behavioral data, the typical transaction behavior of each user or user group is identified, and the corresponding digital transaction behavior data is generated. Select a suitable clustering algorithm, such as K-means, hierarchical clustering, or density clustering (DBSCAN), to process the generated digital transaction behavior data, cluster the digital transaction behavior data, and classify similar transaction behaviors into the same group. For example, users with similar purchase frequencies and transaction amounts are clustered in the same group, or transactions with similar geographic locations and purchase time patterns are classified into one category. The clustering results reflect the differences and commonalities in the transaction behaviors of platform users, generate clustered digital transaction behavior data, and provide input for subsequent transaction pattern recognition. Based on the clustered behavioral data, a transaction pattern recognition model is constructed. The model can use machine learning algorithms such as support vector machines (SVM), decision trees, random forests or neural networks to identify transaction patterns. The model can accurately identify specific transaction patterns, such as high-frequency small transactions, large single transactions, and transactions with rapidly changing geographical locations, by training various types of clustered behavior data. The clustered digital transaction behavior data is input into the pattern recognition model, and the model analyzes the data to identify various types of transaction patterns. For each transaction pattern, the model generates corresponding digital transaction pattern data, which describes the typical characteristics of each type of transaction behavior. The model can also identify abnormal transaction patterns, providing an important basis for subsequent feature dimensionality reduction and mapping.On the basis of pattern recognition, various features in the transaction pattern data are selected and reduced in dimension. Dimension reduction techniques such as principal component analysis (PCA) and linear discriminant analysis (LDA) are used to reduce the high-dimensional transaction pattern feature data into a low-dimensional feature matrix. Through dimensionality reduction processing, the most representative features of transaction behavior, such as transaction time, amount, frequency, etc., are retained, while redundant or noise features are eliminated to ensure that the model can process more concise and information-intensive data in subsequent processing. The transaction pattern dimensionality reduction feature matrix is ​​designed, which contains the core features after dimensionality reduction processing. These features are arranged and combined in the matrix to facilitate subsequent feature mapping and data processing. The design of the matrix aims to highlight the features that best reflect the transaction pattern, thereby improving the computational efficiency and predictive ability of the model. The transaction mode dimensionality reduction feature matrix is ​​used to perform feature mapping on the original multimodal data, and the key features after dimensionality reduction are mapped back to the multimodal data set to generate new digital transaction feature data. Feature mapping ensures that the model can focus on analyzing those features identified as the most important in the dimensionality reduction process and ignore noise and minor features, thereby improving the processing efficiency of subsequent steps. The mapped data generates the final digital transaction feature data, which will have high information density and representativeness and is suitable for further analysis and processing, ultimately generating digital transaction feature data, including feature data with higher value under different transaction modes.

[0052] Preferably, step S11 includes the following steps:

[0053] Collecting initial digital transaction multimodal data on the digital economic transaction platform according to a preset digital economic transaction monitoring engine to obtain initial digital transaction multimodal data;

[0054] Performing data cleaning on the initial digital transaction multimodal data to obtain cleaned digital transaction multimodal data;

[0055] Data standardization is performed on the cleaned digital transaction multimodal data to obtain digital transaction multimodal data.

[0056] The present invention collects initial digital transaction multimodal data on the digital economic transaction platform through a preset digital economic transaction monitoring engine, and comprehensively obtains multimodal data on the transaction platform from multiple dimensions (such as user behavior, transaction type, transaction time, etc.), which can capture detailed information in the transaction process to the greatest extent, improve the diversity and coverage of data, and lay a solid foundation for subsequent analysis. The initial digital transaction multimodal data is cleaned to effectively remove invalid data such as noise, errors and outliers, improve the quality of the data set, and ensure that data related to digital transactions can be accurately obtained, thereby improving the reliability of subsequent data processing and analysis, and reducing analysis errors caused by data quality problems. The cleaned digital transaction multimodal data is standardized so that data from different sources and different types can be compared and analyzed on a unified scale, eliminating deviations caused by different data scales, enhancing data consistency and comparability, and providing more reliable input data for subsequent data feature extraction and model training.

[0057] In an embodiment of the present invention, a preset digital economic transaction monitoring engine is deployed on a digital economic transaction platform. The engine is composed of multiple modules and can monitor and collect various types of transaction data from the platform in real time. The engine is configured to collect multimodal data, including but not limited to the following data types: transaction data (such as order amount, payment method), user behavior data (such as browsing history, click events), time data (such as transaction time, response time), geographic data (such as transaction location, IP address) and social data (such as user comments, interaction frequency). After the monitoring engine starts working, it continuously collects multimodal data for all transaction activities on the platform. The data collection process includes real-time capture of user operations on the platform, detailed information on transaction execution, and auxiliary information related to the transaction, pre-processing the collected multimodal data, and checking and removing transaction data that is repeatedly recorded during the collection process. For example, due to the phenomenon of multiple records of the same transaction caused by network delays or errors, fill in or delete the missing key fields in the record. Use data interpolation, mean filling and other technologies to repair missing data, or choose to delete related records when there are too many missing data. Use statistical methods or machine learning algorithms to identify and process outliers, such as abnormally large transaction amounts or abnormal transaction frequencies. These data points may be caused by data collection errors, system failures, or potential fraud. Standardize the cleaned digital transaction multimodal data to ensure that all data dimensions are on the same scale and eliminate differences between different data types. Standardize all numerical features (such as transaction amount and transaction frequency). Standardize all numerical features (such as transaction amount and transaction frequency). Common methods include Z-score standardization (converting data to zero mean and unit standard deviation) or Min-Max standardization (scaling data to the range of [0,1]), converting categorical features (such as payment methods and commodity categories) into numerical form, using One-Hot encoding, Label encoding, or target encoding to enable categorical data to be processed by machine learning algorithms, and standardizing timestamp and date data, such as converting time to seconds or days from a certain reference point, or extracting periodic features of time (such as hours, days of the week). The standardized digital transaction multimodal data is stored in a standardized data set area to obtain digital transaction multimodal data.

[0058] Preferably, step S2 comprises the following steps:

[0059] Step S21: Perform digital transaction frequency feature analysis based on the digital transaction behavior data to generate digital transaction frequency feature data;

[0060] Step S22: performing digital transaction time series feature analysis based on the digital transaction multimodal data to generate digital transaction time series feature data;

[0061] Step S23: Designing digital transaction attention weights according to the digital transaction frequency feature data and the digital transaction time series feature data to obtain digital transaction attention weight information;

[0062] Step S24: Perform attention weighting processing on the digital transaction feature data according to the digital transaction attention weight information to generate weighted digital transaction feature data.

[0063] The present invention can identify the frequency pattern of transaction behavior, such as the high or low frequency occurrence of certain transaction behaviors, by performing frequency feature analysis on digital transaction behavior data. This analysis helps to improve the sensitivity of detecting abnormal behavior, such as identifying suspicious transactions with significantly different frequencies from normal transaction behaviors, thereby improving the recognition accuracy of potential abnormal behaviors. Performing time series feature analysis based on digital transaction multimodal data can capture the time series pattern of transaction behavior, which helps to identify the time dependency and trend in transaction activities, such as certain abnormal transactions concentrated in a specific time period. Through time series feature analysis, time-related abnormal behaviors can be detected more effectively. Combining frequency feature data with time series feature data for attention weight design can assign reasonable weights to different transaction features, highlight the focus on key features, especially those that show abnormal behavior in frequency and time dimensions, thereby improving the effect of subsequent anomaly detection models. Weighted processing of digital transaction feature data according to the designed attention weight information can optimize the expression of feature data, so that the model pays more attention to those important features in anomaly detection, such as placing the computing resources of anomaly detection on new transaction data or digital transaction data corresponding to unconventional transaction behaviors, thereby improving the accuracy and robustness of digital transaction anomaly detection.

[0064] In an embodiment of the present invention, a statistical analysis method is used to calculate the transaction frequency characteristics of users or user groups in digital transaction behavior data, such as the average number of transactions per day, weekly transaction fluctuations, and transaction peaks in a specific time period. The analysis also includes identifying abnormal transaction frequencies, such as a sudden increase in transaction frequency or a large number of low-amount transactions in a short period of time. The generated frequency characteristic data includes transaction frequency indicators for each user or user group, which will serve as the basic data for subsequent steps. Multimodal data is sorted to extract time-related features, such as the timestamp of the transaction, the duration of the transaction, the transaction interval, etc. These time series data can reveal the time distribution characteristics of transaction behavior and help identify the transaction patterns of users in different time periods. Use time series analysis methods, such as moving average, time series decomposition (Seasonal Decomposition), autoregression (AR) model, etc. to analyze the time series characteristics of transaction behavior. Focus on identifying the periodicity of transaction behavior (such as peak trading time every day or every week), trend changes (such as the increase or decrease of transaction frequency over time), and possible sudden abnormal events (such as large transactions at abnormal time points). The analysis also includes statistics on the time intervals of transaction behavior, identifying the correlation model between transaction frequency and time intervals, and the generated time series feature data will describe the time distribution pattern of each user or transaction group. Select a deep learning model based on the attention mechanism, such as the self-attention mechanism (Self-Attention) or the Transformer model, to design the attention weight of digital transactions. By jointly analyzing the frequency feature data and the time series feature data, automatically identify the features that contribute most to the prediction of transaction behavior and anomaly detection. Input the frequency feature data and the time series feature data into the pre-designed attention mechanism model, and the model dynamically assigns weights according to the importance of the input features. The importance of the features is automatically learned by the model based on the performance in the training data. The weight information includes the weight value of each feature, reflecting its relative importance in the overall transaction behavior analysis. For example, a high-frequency trading time period for a specific user is given a higher weight, while the trading frequency on regular working days is given a lower weight. The generated attention weight information will provide guidance for the weighted processing of subsequent feature data. According to the attention weight information, the digital transaction feature data is weighted. Each feature data is weighted according to its corresponding weight value. The specific steps of weighted processing include marking the corresponding weight value of each feature and generating weighted feature data. These weighted data can better reflect the key features in actual transaction behavior, highlight the most important features for anomaly detection, and generate weighted digital transaction feature data. This dataset contains transaction features after weight adjustment, which can effectively improve the performance of the model in subsequent prediction and detection.

[0065] Preferably, step S3 comprises the following steps:

[0066] Step S31: Obtain historical digital trading strategies;

[0067] Step S32: defining the digital trading strategy state space according to the historical digital trading strategy to obtain the digital trading strategy state space;

[0068] Step S33: Analyze the transaction strategy correlation characteristics of the historical digital transaction strategies to obtain transaction strategy correlation characteristic data;

[0069] Step S34: defining the digital trading strategy action space according to the trading strategy associated feature data and the digital trading strategy state space to obtain the digital trading strategy action space;

[0070] Step S35: Designing a digital trading strategy network according to the digital trading strategy action space to obtain a digital trading strategy network;

[0071] Step S36: Designing a digital transaction value network according to the digital transaction strategy network and the historical digital transaction strategy to obtain a digital transaction value network;

[0072] Step S37: Designing a reinforcement learning mathematical model of the digital transaction strategy based on the digital transaction strategy network and the digital transaction value network to establish a digital transaction strategy reinforcement learning model;

[0073] Step S38: transmitting the weighted digital transaction feature data to the digital transaction strategy reinforcement learning model for digital transaction strategy prediction processing to generate digital transaction strategy prediction data.

[0074] The present invention obtains historical digital trading strategies, and can use past trading data and experience accumulation to provide a reliable basis for model construction. The collection and analysis of historical data is used to identify previously effective strategies, and then provide references for current transaction processing and anomaly detection, ensuring that there is a certain historical basis when processing new data, which helps to improve the accuracy and robustness of predictions. By defining the state space of digital trading strategies, all trading states and their associations can be clearly depicted. State space definition is the core step of reinforcement learning, which helps the reinforcement learning model established later to clarify the various states of the current trading environment and associate different strategy choices for each state, ensuring that the model can effectively identify and adapt to different trading states when facing a complex trading environment, thereby optimizing the detection and decision-making process. By analyzing the associated features of historical digital trading strategies, the intrinsic connections between different strategies can be excavated, key features can be identified, and factors that affect the prediction of trading behavior can be extracted and summarized, providing data support for the subsequent action space definition, significantly improving the pertinence and effectiveness of strategy selection. By defining the action space of the strategy, it is possible to clearly define the strategies or actions that can be taken in different states. The definition of the action space enables the model to evaluate multiple strategy options in each trading state and optimize the execution effects of these options, cope with various complex trading scenarios and potential anomalies, and improve the overall detection effect. By designing a digital trading strategy network, it is possible to systematically represent and process the complex relationships in the strategy space. The establishment of the strategy network enables the system to efficiently select and optimize strategies, and capture the nonlinear relationships and mutual influences between strategies through the network structure, which not only improves the efficiency of strategy reasoning, but also enhances the intelligent decision-making ability of the reinforcement learning model established subsequently. By designing a value network, the effects (i.e., value) of different strategies in long-term execution are evaluated and quantified. The construction of the value network provides an evaluation criterion for the model, enabling the system to select strategies to optimize long-term benefits when faced with multiple strategies, improving the decision-making quality of the reinforcement learning model established subsequently in trading strategy prediction, so that anomaly detection not only focuses on short-term effects, but also takes into account long-term effects, ensuring the reliability and sustainability of detection results. By combining the strategy network and the value network and designing a reinforcement learning model, the system can adaptively optimize trading strategies, that is, predict subsequent trading strategies based on trading data. The reinforcement learning model optimizes strategies through continuous testing and feedback, ensuring that the model can continue to learn and improve when facing dynamic and complex trading environments, thereby improving the accuracy and efficiency of detection and reducing false positives and false negatives. By inputting weighted feature data into the reinforcement learning model, the system can use the optimized strategy for efficient prediction. The data generated by the prediction process provides an accurate source of information for subsequent anomaly detection and strategy adjustment, ensuring that the detection model can respond in real time and accurately, improving the performance of the system in complex trading environments, and providing strong support for the final anomaly detection.

[0075] As an embodiment of the present invention, refer to Figure 2 As shown, Figure 1 FIG. 1 is a flow chart of detailed implementation steps of step S3 in FIG. 1 . In this embodiment, step S3 includes:

[0076] Step S31: Obtain historical digital trading strategies;

[0077] In the embodiment of the present invention, past digital transaction strategy data are extracted from the historical records of the transaction platform. These strategy data include transaction behavior decisions taken by users in different time periods and different transaction scenarios, such as transaction strategies obtained by combining different transaction behaviors in historical digital transactions, including key information such as transaction time, amount, participants, payment methods, and geographical distribution, as well as how these factors affect the final transaction success rate or user satisfaction. These historical transaction strategy data are archived to form a structured database to obtain historical digital transaction strategies.

[0078] Step S32: defining the digital trading strategy state space according to the historical digital trading strategy to obtain the digital trading strategy state space;

[0079] In an embodiment of the present invention, historical digital transaction strategies are analyzed to determine the main factors affecting the selection of transaction strategies. When defining the state space, these factors are used as state variables. These variables include transaction amount ranges, user historical behavior characteristics, time characteristics, geographic locations, etc. The state space is composed of a combination of these variables. In order to simplify processing, the continuous state variables are discretized, and each state in the state space is encoded and stored to obtain the digital transaction strategy state space.

[0080] Step S33: Analyze the transaction strategy correlation characteristics of the historical digital transaction strategies to obtain transaction strategy correlation characteristic data;

[0081] In the embodiment of the present invention, the historical digital trading strategies under each state are subjected to correlation feature analysis to identify which trading strategies have a strong correlation with specific state variables. For example, the correlation between each trading state and the trading strategy in the historical trading strategy is analyzed, and the characteristic data with significant correlation is extracted using association rule analysis (such as Apriori algorithm) or correlation analysis technology. The trading strategy correlation feature data is generated, including the trading strategy information related to each trading state. These feature data are aggregated to form the basis for the subsequent action space definition.

[0082] Step S34: defining the digital trading strategy action space according to the trading strategy associated feature data and the digital trading strategy state space to obtain the digital trading strategy action space;

[0083] In the embodiment of the present invention, after the state space and associated features are clarified, the action space of the digital trading strategy is defined, and the action space includes all trading strategy options that can be selected in a specific state, that is, the trading action behavior route of a specific trading state in the digital trading strategy state space. The action space is stored in the form of a matrix or table, each row represents a state, each column represents an action in the state, and the stored information includes the historical execution effect of each action, so as to obtain the digital trading strategy action space.

[0084] Step S35: Designing a digital trading strategy network according to the digital trading strategy action space to obtain a digital trading strategy network;

[0085] In the embodiment of the present invention, a digital trading policy network (Policy Network) is designed based on the action space. The network architecture adopts a deep neural network. The input layer receives state space data, the hidden layer models the complex relationship between the state and the action, and the output layer generates corresponding trading policy action suggestions. The network design takes into account the nonlinear relationship between different states and the interaction between high-dimensional features. Efficient policy reasoning is achieved through structures such as multi-layer perceptron (MLP) or convolutional neural network (CNN), and the policy network is trained using state-action pairs in historical data. Through supervised learning, the network gradually optimizes its parameters so that the optimal trading policy suggestion can be output under a given state. During the training process, cross-validation technology is used to avoid overfitting and ensure the good performance of the network on unknown data to obtain a digital trading policy network.

[0086] Step S36: Designing a digital transaction value network according to the digital transaction strategy network and the historical digital transaction strategy to obtain a digital transaction value network;

[0087] In the embodiment of the present invention, a digital transaction value network (Value Network) is designed to evaluate the long-term effect of each transaction strategy. The value network inputs state and action data and outputs the value score of each strategy, that is, the expected return or effect of the strategy in long-term execution. The network structure is similar to the digital transaction strategy network, but the output layer has only one node, which represents the expected value under a given state and strategy. The value network is trained using historical strategy data, with the goal of predicting the true value of each strategy under a given state. During the training process, the error between the predicted value and the actual value is minimized by adjusting the network weights (using the mean square error as the loss function). The network training adopts an iterative optimization method (such as gradient descent), and regularization is used to prevent overfitting, ensuring the stability and generalization ability of the model, so as to obtain a digital transaction value network.

[0088] Step S37: Designing a reinforcement learning mathematical model of the digital transaction strategy based on the digital transaction strategy network and the digital transaction value network to establish a digital transaction strategy reinforcement learning model;

[0089] In the embodiment of the present invention, a digital trading strategy model based on reinforcement learning is established by combining the strategy network and the value network, and deep Q learning (DQN) or the policy gradient method (Policy Gradient) is used as the core algorithm. The goal of the reinforcement learning model is to optimize the digital trading strategy through continuous experimentation and feedback. The input of the model is the state space feature, and the output is the optimal strategy action. The model interacts with the environment and continuously updates the parameters of the strategy network and the value network, so that the strategy with the greatest long-term benefit can be selected under different states. The reinforcement learning model is trained in a simulation environment, which generates different trading scenarios based on historical trading data. The model tests different strategy actions in these scenarios and updates the network parameters based on the feedback obtained (such as user response, transaction success rate, etc.). The model gradually optimizes its strategy through multiple rounds of iterations, so that it can make the best decision when facing a complex and changing trading environment.

[0090] Step S38: transmitting the weighted digital transaction feature data to the digital transaction strategy reinforcement learning model for digital transaction strategy prediction processing to generate digital transaction strategy prediction data.

[0091] In the embodiment of the present invention, the weighted digital transaction feature data is input into the trained reinforcement learning model. The model calculates based on the input feature data and infers the optimal strategy action under the current transaction state. The reinforcement learning model will comprehensively consider factors such as current feature information, historical strategy effects, and long-term value to generate prediction data. The model outputs digital transaction strategy prediction data, which contains suggestions for the optimal strategy under the current transaction environment. The generated strategy prediction data will be transmitted to the next step or directly applied to transaction decisions to improve the success rate of transactions and user experience.

[0092] Preferably, step S36 includes the following steps:

[0093] Step S361: dividing the historical digital trading strategies into historical trading strategy action nodes according to the digital trading strategy network to obtain historical trading strategy action nodes;

[0094] Step S362: performing action node status evaluation on the historical transaction strategy action node to obtain historical transaction strategy action node status evaluation data;

[0095] Step S363: Design a digital transaction value network based on historical transaction strategy action node status evaluation data to obtain a digital transaction value network.

[0096] The present invention divides the action nodes of historical digital trading strategies, and can decompose complex trading strategies into smaller and more manageable action nodes. Each action node represents a specific trading decision or behavior unit. This division method helps to refine and clarify the components of different trading strategies, more accurately identify the key behavior nodes in the trading process, and improve the sensitivity and accuracy of anomaly detection. Perform a state evaluation on each action node to quantify the performance and role of each node in different historical trading strategies. The state evaluation gives each node a clear indicator based on the transaction status in the historical trading strategy, reflecting its effectiveness and risk in the strategy execution process, identifying which nodes are high-risk and which nodes are crucial to successfully detecting and preventing abnormal behaviors, and providing key data support for the design of subsequent value networks. By using the state evaluation data of action nodes to design a value network, the system can systematically evaluate the long-term effects of different trading strategies. The design of the value network not only takes into account the immediate benefits of each action node, but also takes into account its long-term impact. The establishment of the value network ensures that the system can quantify the transaction behavior states of the predicted transaction decisions in a dynamic and complex transaction environment, so as to predict the subsequent digital transaction decisions of each weighted digital transaction feature data, so as to improve the ability to detect predictive anomalies.

[0097] In the embodiment of the present invention, the constructed digital trading strategy network is used to analyze each decision point in the historical digital trading strategy. Each decision point corresponds to a specific trading behavior or strategy selection. By analyzing these decision points, each historical strategy is divided into multiple action nodes. Each node represents a specific decision or operation. These operations are connected together through different paths and nodes in the trading strategy network, reflecting the execution order and logical relationship of the strategy. In each strategy execution path, the strategy is divided into multiple continuous action nodes, and similar division operations are performed on all historical strategies to generate a series of historical trading strategy action nodes. These nodes represent all decision points that may occur in the transaction process. An action node state evaluation model is constructed to evaluate the performance and effect of each action node in the historical trading strategy. The evaluation model is based on a machine learning algorithm, such as a random forest or a support vector machine, to analyze the state performance of different nodes. The evaluation indicators include the execution frequency, success rate, execution time, user response, etc. of each node. These indicators can reflect the role and importance of each action node in the transaction process. Input historical transaction decision data into the evaluation model, and conduct detailed status evaluation for each action node. The evaluation model calculates the key status data of each node by analyzing historical data, such as the execution effect of the node under specific circumstances (such as peak period, promotion period). The output of the evaluation model includes the status evaluation data of each node, such as "efficient node", "inefficient node", "key node" and other classifications, to help identify the action nodes that play a key role in the transaction strategy. Summarize all evaluation results to generate historical transaction strategy action node status evaluation data. The evaluation data will be stored in a table or matrix form, including the status evaluation results of each action node. These data will be used for subsequent value network design to ensure that the value network can accurately reflect the actual effect of the transaction strategy. Based on the action node status evaluation data generated in the previous step, design a digital transaction value network. The goal of the value network is to evaluate and predict the overall effect (i.e. expected value) of different strategies in long-term execution, including the impact on transaction success rate, user satisfaction, long-term user retention rate, etc. Design a deep neural network architecture as the basis of the digital transaction value network. The input layer of the network accepts action node status evaluation data, including the status score of each node, execution effect and other characteristics. The hidden layer mines the deep connections between different nodes by nonlinearly combining these input data, and the output layer provides the overall value score of each trading strategy, which reflects the expected long-term return of the strategy under different states. The network uses architectures such as multi-layer perceptron (MLP) or recurrent neural network (RNN) to achieve value assessment of complex strategies. The value network is trained using historical trading data. The training goal is to minimize the error between the network's predicted value and the actual strategy effect. Through multiple iterations, the network's weights and biases are adjusted so that the network can accurately predict the long-term value of different strategies.Cross-validation and regularization techniques are used in the network training process to prevent overfitting of the model and ensure good generalization when processing new data. After training, the generated digital transaction value network will be used to evaluate and predict the performance of different strategies in digital transactions in real time, and applied to the reinforcement learning model to predict the choice of transaction strategy for real-time digital transactions.

[0098] Preferably, step S4 comprises the following steps:

[0099] Step S41: performing multi-dimensional digital transaction combination analysis processing according to the digital transaction strategy prediction data to obtain multi-dimensional digital transaction combination data;

[0100] Step S42: performing a multi-dimensional digital transaction fraud feature analysis based on the multi-dimensional digital transaction combination data to generate multi-dimensional digital transaction fraud feature data;

[0101] Step S43: using a preset generative adversarial network model to perform fraud feature derivation processing on the multi-dimensional digital transaction fraud feature data to generate multi-dimensional digital transaction fraud feature derived data;

[0102] Step S44: Designing a mathematical model of the adversarial game of abnormal transactions based on the multi-dimensional digital transaction fraud feature derived data to generate an adversarial game model of abnormal transactions.

[0103] The present invention performs multi-dimensional combined analysis on digital transaction strategy prediction data, and comprehensively analyzes transaction behaviors in different dimensions. It captures complex and diverse interaction patterns in transaction behaviors, and reveals abnormal features that cannot be discovered by single-dimensional analysis. It performs fraud feature analysis on multi-dimensional digital transaction combination data, analyzes potential fraud features in transaction behaviors, and identifies abnormal behaviors hidden in normal transaction patterns, which provides an important basis for further anomaly detection, enhances the detection capability of the system when dealing with complex and hidden fraud behaviors, and effectively reduces the false alarm rate and missed alarm rate. Through the adversarial generative network (GAN) to derive multi-dimensional fraud feature data, the system can simulate potential complex fraud behaviors and generate new fraud features. This derivative processing helps the system introduce more diverse fraud patterns in model training, improve the robustness and generalization ability of the model, better predict and prevent new and variant fraud behaviors, and ensure that the detection model always maintains high accuracy and adaptability. Design an adversarial game model based on multi-dimensional fraud feature derivative data to simulate and predict the game process between the fraudster and the detection system. The adversarial game model helps the system develop more effective detection strategies by analyzing the avoidance strategies that fraudsters may adopt, improves the system's automated detection capabilities for complex and abnormal transactions, and effectively responds to the dynamic changes and evolution of fraudulent behaviors, ensuring that the system can continue to optimize and maintain its advantages in a fierce adversarial environment.

[0104] In an embodiment of the present invention, relevant data is extracted from the digital transaction strategy prediction data, which includes the predicted transaction strategy and corresponding feature data, such as user behavior features, transaction environment features, etc. The extracted data is combined and analyzed using multidimensional data analysis technology. The combined analysis is to analyze the combination of multiple sent digital transaction strategies in the digital transaction strategy prediction data, and identify the interactions between different dimensions, as well as the impact of these interactions on transaction behavior and strategy selection. The result of the analysis generates multidimensional digital transaction combination data, which reflects the complex associations and potential patterns between the dimensions. The multidimensional digital transaction combination data is analyzed to identify fraud features. For example, it is analyzed whether there are frequent small transactions within a specific time period, or large-scale fund transfers between multiple geographical locations in a short period of time. The feature analysis not only focuses on the anomalies of a single dimension, but also considers how the interactions between multiple dimensions lead to the occurrence of abnormal transaction behaviors. The generated fraud feature data reflects these complex patterns and potential risks. The analysis results are integrated to generate multidimensional digital transaction fraud feature data, including the fraud behavior of each transaction behavior. A preset generative adversarial network (GAN) model is constructed, which includes two core parts: a generator and a discriminator. The generator is responsible for generating possible fraud feature data, while the discriminator is responsible for distinguishing these generated data from real fraud feature data. Through multiple rounds of adversarial training, the generator gradually learns to generate fraud features similar to real data, and the discriminator continuously improves its discrimination ability, so that the entire network is continuously optimized in the generation and detection of fraud features. The multi-dimensional digital transaction fraud feature data is input into the preset GAN model, and the generator generates new fraud feature data based on the input data. The generated data contains potential fraud patterns that have not been detected in the original data, helping the system to identify more concealed or new fraud behaviors. The generated multi-dimensional digital transaction fraud feature data contains the original feature data and new fraud features generated by the GAN model. When designing the adversarial game model, we first establish a theoretical foundation based on basic concepts in game theory, such as Nash equilibrium and evolutionary stable strategy (ESS). The model aims to simulate the dynamic confrontation between fraudsters and defense systems, help predict the evasion strategies adopted by fraudsters, and use the derived data of multi-dimensional digital transaction fraud characteristics to build an adversarial game model for abnormal transactions. The model contains two participants: the fraudster and the detection system. The goal of the fraudster is to maximize the success rate of its transaction, while the goal of the detection system is to maximize its detection accuracy. Through multiple rounds of iterations, the model simulates the performance of both parties under different strategies and analyzes the possible strategy combinations that both parties may adopt during the game. The results of each round of the game will be fed back to the model to adjust the strategy choices of both parties.The adversarial game model is trained through historical data and generated derivative data, so that it can accurately predict the game results under different conditions. During the model optimization process, the reinforcement learning algorithm is used to further enhance the model's decision-making ability in complex game environments. Through continuous optimization, the model can capture more complex adversarial strategies adopted by fraudsters and adjust the detection system's strategy to meet these challenges. After training is completed, the final abnormal transaction adversarial game model is generated to provide the system with accurate fraud response strategies to effectively respond to and detect complex fraud behaviors in actual trading environments.

[0105] Preferably, step S5 comprises the following steps:

[0106] Step S51: performing abnormal transaction adversarial game feature analysis and processing according to the abnormal transaction adversarial game model to generate abnormal transaction adversarial game feature data;

[0107] Step S52: Designing digital transaction fraud tracing tree nodes according to the abnormal transaction adversarial game feature data to obtain digital transaction fraud tracing tree node data;

[0108] Step S53: Designing the associated labels of the fraud tracing tree structure based on the weighted digital transaction feature data to generate the associated label data of the fraud tracing tree structure;

[0109] Step S54: setting fraud tracing tree structure weight parameters according to the digital transaction attention weight information corresponding to the weighted digital transaction feature data to obtain fraud tracing tree structure weight parameters;

[0110] Step S55: Designing a step-by-step fraud tracing tree structure model of digital transactions for the digital transaction fraud tracing tree node data based on the fraud tracing tree structure associated label data to establish a preliminary digital transaction fraud tracing tree model;

[0111] Step S56: Optimize the tree structure weight parameters of the preliminary digital transaction fraud tracing tree model according to the fraud tracing tree structure weight parameters to obtain the digital transaction fraud tracing tree model.

[0112] The present invention is based on the feature analysis of the abnormal transaction confrontation game model. The system can deeply explore the strategic confrontation relationship between the fraudster and the detection system, extract the key feature data reflecting the confrontation process between the two parties, and provide an important reference basis for subsequent detection and response strategies. The generated confrontation game feature data can not only improve the accuracy of abnormal detection, but also help the system predict the trend of fraud behavior changes, thereby improving the overall prevention ability. Designing the nodes of the fraud tracing tree based on the confrontation game feature data can effectively decompose complex abnormal transaction behaviors into multiple key nodes, each node represents a specific fraud behavior feature or transaction mode, so that the tracing tree can show the entire fraud behavior chain in a modular way, which helps the system to quickly locate and identify the source of abnormal behavior and improve the efficiency of tracing and analysis. The fraud tracing tree structure is designed with associated labels. The system can assign appropriate labels to each node and branch to clarify its role and relationship in the entire tracing structure, and closely associate different transaction features with specific fraud behavior patterns, thereby enhancing the readability and interpretability of the tracing tree when identifying complex fraud behaviors, and ensuring that the system can obtain a clearer and more comprehensive perspective when conducting tracing analysis. By combining the attention weight information of weighted digital transaction feature data to set the weight parameters of the traceability tree, the system can assign appropriate weights to different nodes and paths. The setting of weight parameters ensures that in the traceability analysis, the system can focus more accurately on key features and important paths, thereby improving the accuracy and efficiency of the overall model in fraud tracing, and improving the adaptability and decision-making ability of the traceability tree in complex transaction environments. By gradually designing and building a fraud tracing tree model, it is possible to start from the basic nodes, refine and expand the structure of the traceability tree layer by layer, ensure that the model can adapt to the needs of fraud tracing of different scales and complexities, and can flexibly adjust the distribution of nodes and paths to adapt to changing transaction patterns and fraud methods. The preliminary fraud tracing tree model provides the system with a dynamic and flexible tool that can be gradually evolved and improved in practical applications. By optimizing the weight parameters of the initially designed traceability tree model, the system can further improve the accuracy and reliability of the model. The optimized tree structure weight parameters ensure that the system can more effectively identify the key nodes of fraudulent behavior and give priority to paths with higher risks, such as focusing on higher transaction frequencies or recent digital transactions. The weight optimization process makes the actual performance of the traceability tree model in the application even better, and enhances the system's prevention capabilities and response speed in high-risk transaction environments.

[0113] As an embodiment of the present invention, refer to Figure 3 As shown, Figure 1 FIG. 5 is a flow chart of detailed implementation steps of step S5, in this embodiment, step S5 includes:

[0114] Step S51: performing abnormal transaction adversarial game feature analysis and processing according to the abnormal transaction adversarial game model to generate abnormal transaction adversarial game feature data;

[0115] In the embodiment of the present invention, an abnormal transaction confrontation game model is used to conduct an in-depth analysis of historical transaction data and generated fraud feature derivative data. By simulating the strategic confrontation between fraudsters and defense systems, confrontation features of various types of transaction behaviors are extracted, including identifying abnormal behavior features under different strategy combinations, such as the strategy of fraudsters repeatedly using a certain transaction mode in a specific environment, or how fraudsters adjust their strategies to evade detection. Through the analysis of the confrontation game model, abnormal transaction confrontation game feature data is generated, including detailed confrontation features, such as frequent changes in transaction paths, sudden large-amount transactions, etc. These features reflect the evolution of fraudsters' strategies in multiple confrontations, ensuring that subsequent retrospective analysis can accurately capture key abnormal behaviors.

[0116] Step S52: Designing digital transaction fraud tracing tree nodes according to the abnormal transaction adversarial game feature data to obtain digital transaction fraud tracing tree node data;

[0117] In the embodiment of the present invention, the nodes of the traceability tree are defined according to the abnormal transaction confrontation game feature data. Each node represents a specific abnormal behavior or transaction feature. These nodes are connected by paths to form a complete fraud behavior chain. The design takes into account the causal relationship between the nodes and the strategy conversion during the game process. All extracted confrontation game features are classified and sorted to generate digital transaction fraud traceability tree node data, which includes the feature description of each node, the associated transaction behavior, and the correlation with other nodes. The node data will be used for the structural design of the traceability tree to ensure that each node accurately reflects the fraudulent behavior.

[0118] Step S53: Designing the associated labels of the fraud tracing tree structure based on the weighted digital transaction feature data to generate the associated label data of the fraud tracing tree structure;

[0119] In the embodiment of the present invention, the weighted digital transaction feature data is used to design labels for each node in the traceability tree. The role of the label is the association of each transaction action combination and the predicted transaction action, which helps the system to quickly identify the key nodes of fraudulent actions in digital transactions during the traceability process. The label design is based on the corresponding relationship of each digital transaction combination in the weighted digital transaction feature data, and the feature data is mapped to the nodes of the traceability tree. The designed labels are attached to each node of the traceability tree to generate the associated label data of the fraud traceability tree structure, and the associated label data will be used to associate each tree node in the traceability tree.

[0120] Step S54: setting fraud tracing tree structure weight parameters according to the digital transaction attention weight information corresponding to the weighted digital transaction feature data to obtain fraud tracing tree structure weight parameters;

[0121] In an embodiment of the present invention, the weight parameters of each node and path in the traceability tree are set according to the weighted digital transaction feature data and the corresponding attention weight information. The weight parameters determine which paths or nodes the system should give priority to during the traceability process. The attention weight information of each feature is considered during the setting process, so that the nodes corresponding to high-weight features have a higher priority in the traceability tree. In this way, the system can quickly lock the key path during the traceability process, such as the key path in high-frequency trading or recent transactions, thereby improving efficiency. The set weight parameters are saved as the core attributes of the traceability tree structure. These parameters will be used in the preliminary design and optimization process of the traceability tree model to ensure that the model can accurately reflect the real transaction risks.

[0122] Step S55: Designing a step-by-step fraud tracing tree structure model of digital transactions for the digital transaction fraud tracing tree node data based on the fraud tracing tree structure associated label data to establish a preliminary digital transaction fraud tracing tree model;

[0123] In an embodiment of the present invention, a step-by-step fraud tracing tree structure model of digital transactions is designed based on the fraud tracing tree structure associating label data with the digital transaction fraud tracing tree node data. During the design process, each node is connected according to causality and correlation to form a complete tracing path tree structure. The structural model design takes into account multiple factors, including the direct association between nodes, the influence of weight parameters on path priority, and the guiding role of label data on model branches. When designing the preliminary model, a step-by-step construction and verification method is adopted to construct a simplified tracing tree structure, conduct preliminary fraud tracing analysis, and gradually expand and optimize the model according to the analysis results. The model expansion at each stage is verified by actual transaction data to ensure that the model has sufficient accuracy and breadth when identifying fraudulent behavior, so as to establish a preliminary digital transaction fraud tracing tree model.

[0124] Step S56: Optimize the tree structure weight parameters of the preliminary digital transaction fraud tracing tree model according to the fraud tracing tree structure weight parameters to obtain the digital transaction fraud tracing tree model.

[0125] In the embodiment of the present invention, the global weight parameters are optimized based on the tracing tree model of the preliminary design. The optimization goal is to adjust the weights of each node and path so that the model can preferentially detect high-risk paths during the tracing process and reduce the waste of resources on low-risk paths. The weight parameters are adjusted by an optimization algorithm (such as a genetic algorithm or a particle swarm optimization algorithm). The best weight allocation party is found by simulating multiple transaction scenarios. After the optimization is completed, the final digital transaction fraud tracing tree model is obtained. The model has an optimized weight parameter setting and can efficiently and accurately identify and trace complex fraudulent behaviors in practical applications.

[0126] Preferably, step S51 includes the following steps:

[0127] Step S511: performing abnormal transaction game evolution processing according to the abnormal transaction confrontation game model to generate abnormal transaction game evolution data;

[0128] Step S512: Performing game evolution replication dynamic processing according to the abnormal transaction game evolution data to generate abnormal transaction game evolution replication dynamic data;

[0129] Step S513: performing game evolution equilibrium point analysis and processing according to the abnormal transaction game evolution replication dynamic data to generate game evolution equilibrium point data;

[0130] Step S514: performing abnormal transaction adversarial game feature analysis on the abnormal transaction adversarial game model according to the game evolution equilibrium point data to generate abnormal transaction adversarial game feature data.

[0131] The present invention performs game evolution processing on abnormal transactions, can simulate and track the strategy evolution process between the fraudster and the detection system, reveal the strategy change trend of both parties in the continuous confrontation, help identify the evolution path and emerging patterns of fraudulent behavior, so that abnormal detection is not limited to static rules, but can adapt to the changes of fraud strategies in real time, thereby enhancing the system's foresight and early warning capabilities. Through the replication dynamic processing of game evolution, the propagation and reinforcement effects of different strategies in the game process are quantified and evaluated, and it is understood why certain fraud strategies are widely adopted or abandoned in specific environments. The generated replication dynamic data provides a quantitative basis for further analysis of the effectiveness and risk of the strategy, and can give priority to those strategies that have evolved into mainstream fraud methods, thereby improving the initiative and pertinence of defense. Through equilibrium point analysis, the stable strategy combination that the two parties in the game may reach in long-term confrontation can be determined, that is, the Nash equilibrium point. The equilibrium point data reveals the equilibrium state of the optimal strategies of the fraudster and the detection system under specific conditions, predicts the behavior pattern of the fraudster, and optimizes its own detection strategy, so that the system always maintains the initiative in a complex trading environment and avoids being passive in strategic confrontation. Through game feature analysis based on equilibrium point data, we can deeply explore the key features of fraudulent behavior and optimize the anomaly detection model accordingly. The generated game feature data reflects which features are most critical for identifying and preventing fraudulent behavior during the game process, greatly improving the accuracy and efficiency of anomaly detection, so that the system can accurately identify core risk points and respond quickly when faced with changing fraud methods, thereby enhancing overall defense capabilities.

[0132] In the embodiment of the present invention, based on the abnormal transaction confrontation game model, the strategy set of each party (such as the fraudster and the detection system) is initialized. Each party has a set of strategies, and these strategies will continue to evolve during the game. The initial strategy includes common fraudulent behaviors (such as small-amount multiple transactions, rapid cross-regional fund transfers) and corresponding detection strategies (such as transaction frequency analysis, cross-border transaction monitoring, etc.). By simulating multiple rounds of games, the performance of participants under different strategy combinations is recorded. After each round of games, both parties adjust their own strategies according to the strategies of the other party, thereby promoting the evolution of strategies. The replicator dynamics equation in evolutionary game theory is used to describe the changing trend of strategies. The replicator dynamics processing is intended to simulate the propagation and changes of strategies during the game. The equation can predict which strategies will be retained and strengthened, and which strategies will gradually be eliminated. During the evolution process, the results of each round of games are recorded, including key information such as strategy selection, success rate, and cost. In each round of dynamic replication processing, key information such as the change in the proportion of strategies and the increase or decrease in the success rate are recorded to generate dynamic data of abnormal transaction game evolution replication. The dynamic data reflects how the strategy evolves and adapts to changes in the external environment during the game process, providing a basis for subsequent equilibrium point analysis. Based on the dynamic replication data, a game evolution equilibrium point analysis model is constructed. The model uses game theory tools such as Nash equilibrium analysis and evolutionary stable strategy (ESS) to analyze the stability of different strategy combinations. The model aims to find the equilibrium state that the strategies of both parties may reach during the game process, that is, in this state, neither party has the motivation to unilaterally change its strategy. By simulating multiple games, the equilibrium point is gradually approached. The strategy distribution at the end of each game is analyzed to determine when the system enters a stable state. For each possible strategy combination, its stability in the equilibrium state is calculated. The key characteristics of each equilibrium point, such as strategy combination, expected return, risk exposure, etc., are recorded and saved to generate game evolution equilibrium point data. Using the data of the equilibrium point of the game evolution, we analyze the characteristics and impact of each strategy in the equilibrium state. The characteristic analysis focuses on how the strategy evolves to the equilibrium point during the game process and the correlation between different strategies in the equilibrium state. By comparing the data of different equilibrium points, we identify high-risk strategy combinations, which usually represent effective strategies that fraudsters may use repeatedly. The analysis results are converted into abnormal transaction counter-game characteristic data, which includes fraudulent behavior at each equilibrium point, the combination of strategies to deal with fraudulent behavior, and the key nodes that trigger abnormal behavior.

[0133] Preferably, step S6 comprises the following steps:

[0134] Step S61: transmitting the weighted digital transaction feature data to the digital transaction fraud tracing tree model for digital transaction anomaly prediction processing to generate digital transaction anomaly prediction data;

[0135] Step S62: performing digital transaction anomaly assessment processing on the digital transaction anomaly prediction data based on a preset digital transaction anomaly assessment decision to generate digital transaction anomaly assessment data;

[0136] Step S63: Mapping the digital transaction anomaly assessment data to the digital transaction anomaly prediction data to perform digital transaction anomaly feedback information integration processing to generate digital transaction anomaly feedback data;

[0137] Step S64: Transmitting the digital transaction abnormality feedback data to the terminal to execute the digital transaction abnormality feedback operation.

[0138] The present invention transmits weighted digital transaction feature data to the fraud tracing tree model, which can make full use of the feature weight information constructed previously to perform accurate abnormal transaction prediction processing, ensuring that the prediction model can be analyzed based on the most relevant feature data, thereby generating high-precision abnormal prediction results, effectively improving the accuracy and sensitivity of the prediction, and enabling the system to promptly discover and warn potential abnormal transaction behaviors. By processing the prediction data using the preset abnormal evaluation decision, the prediction results are further refined and evaluated, which helps to filter out false alarms, ensure that only alarms are issued for real abnormal transaction behaviors, reduce the false alarm rate, and provide quantitative information on fraudulent behaviors, thereby improving the detection reliability and accuracy of the overall system. Mapping and integrating the abnormal evaluation data with the prediction data can form comprehensive abnormal feedback information, ensure that the feedback data has integrity and consistency, enable the system to accurately locate abnormal behaviors, and provide detailed references for further preventive measures, which can not only help operators quickly understand the problem, but also provide valuable feedback information for further optimization of the model. By transmitting abnormal feedback data to the terminal, the system can trigger corresponding preventive or remedial measures in real time or in advance, quickly respond to potential risks, and ensure that the system can not only detect abnormal transactions, but also take effective measures to intervene or prevent them, thereby protecting the security of the digital transaction environment. The real-time feedback mechanism enables the system to have closed-loop control capabilities, and can continuously optimize its own detection and protection capabilities, forming an adaptive and self-optimizing security system.

[0139] In the embodiment of the present invention, the weighted digital transaction feature data is input into the constructed digital transaction fraud tracing tree model. The model has the ability to predict complex transaction behaviors based on the fraud features and weight data generated in the previous steps. When the model is initialized, the feature data is loaded to ensure that the model can make full use of these weight information for prediction. After receiving the input data, the tracing tree model analyzes the input feature data according to the weight parameters of each node to determine whether the input feature data conforms to the known fraud pattern or has abnormal behavior. The model predicts abnormal transaction behavior according to the association path between the nodes. The prediction result includes the abnormal behavior characteristics or codes of each transaction. The abnormal behavior code reflects its specific abnormal or fraudulent behavior to generate digital transaction abnormal prediction data. According to the digital transaction abnormal evaluation decision criteria preset by the system, the abnormal prediction data is further evaluated. The evaluation criteria may include common abnormal indicators such as abnormal transaction amount, frequent transaction behavior, and sudden change of geographical location. On the basis of the preset standards, the system formulates multi-level evaluation rules to ensure that there are accurate evaluation indicators for each abnormal situation, compares the abnormal prediction data with the evaluation criteria, and judges the abnormal degree of each transaction through the evaluation model. The system will assign different quantitative indicators such as risk coefficients according to different abnormal characteristics. The evaluation process takes into account a variety of factors, including historical transaction patterns, changes in user behavior, and transaction environments, to ensure that the evaluation results are highly accurate and comprehensive, in order to generate digital transaction anomaly evaluation data, which records the evaluation results and risk scores of each transaction. The digital transaction anomaly evaluation data is mapped and integrated with the previously generated anomaly prediction data. By comparing and integrating the predictions with the evaluation results, the system can generate more comprehensive anomaly feedback information. The integration process includes combining various risk indicators in the evaluation with the anomaly features in the prediction to generate digital transaction anomaly feedback data, including specific anomalies or fraudulent behaviors, anomaly behavior scores, and response strategies. The generated digital transaction anomaly feedback data is transmitted to the designated terminal system. The terminal may include an automated risk control system, a user notification platform, or a manual review system. The data transmission process ensures the accuracy and timeliness of the information, so that the terminal system can respond to the anomaly feedback information in the first place. In the terminal system, the received anomaly feedback data will trigger the corresponding execution operation. For example, the automated system freezes suspicious accounts, blocks risky transactions, or sends warning notifications to users based on feedback data. After the execution process is completed, the terminal system will feed back the execution results (such as successful operation, user response, etc.) to the main system to improve the closed-loop control of transactions and continuously optimize the anomaly detection process.

[0140] The beneficial effect of the present application is that the digital economic transaction processing and anomaly detection method based on machine learning of the present invention can adaptively adjust digital transaction strategies and detection methods by introducing machine learning models, especially reinforcement learning models. Unlike traditional methods that rely on predefined rules and thresholds, machine learning models extract features from historical data and predict future transaction behaviors through continuous learning and optimization. This dynamic adaptability enables the system to promptly discover and respond to emerging fraud patterns, no longer limited to past experience and known fraud patterns. By designing digital transaction attention weight information and adversarial game mathematical models, it can handle more complex and hidden abnormal behaviors. The weighted processing of attention weights enables the system to concentrate resources to analyze abnormal transaction features, thereby improving the accuracy and efficiency of anomaly detection. The introduction of adversarial game models enables the system to simulate and confront potential fraud strategies, identify complex fraud behaviors that circumvent traditional detection methods, and continuously optimize the model in the process of anomaly detection. By optimizing the weight parameters of the preliminary digital transaction fraud tracing tree model, the accuracy and effectiveness of the model are continuously improved. The ability to continuously learn and optimize keeps pace with the evolving fraud behaviors and gradually enhances the detection capabilities of new and complex digital transaction anomalies.

[0141] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0142] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A digital economic transaction processing and anomaly detection method based on machine learning, characterized in that: The following steps are involved: Step S1: collecting digital transaction multimodal data from a digital economic transaction platform according to a preset digital economic transaction monitoring engine to obtain digital transaction multimodal data; performing digital transaction feature analysis and processing on the digital transaction multimodal data to generate digital transaction feature data; Step S2: Designing digital transaction attention weight information based on digital transaction multimodal data; performing attention weight weighting processing on digital transaction feature data according to the digital transaction attention weight information to generate weighted digital transaction feature data; Step S3: Acquire historical digital trading strategies; establish a digital trading strategy reinforcement learning model based on the historical digital trading strategies; transmit the weighted digital trading feature data to the digital trading strategy reinforcement learning model for digital trading strategy prediction processing to generate digital trading strategy prediction data; Step S4: Performing feature derivative analysis and processing of multi-dimensional digital transaction fraud according to the digital transaction strategy prediction data to generate multi-dimensional digital transaction fraud feature derivative data; designing a confrontation game mathematical model of abnormal transactions based on the multi-dimensional digital transaction fraud feature derivative data to generate an abnormal transaction confrontation game model; Step S5: performing abnormal transaction adversarial game feature analysis and processing according to the abnormal transaction adversarial game model to generate abnormal transaction adversarial game feature data; establishing a preliminary digital transaction fraud tracing tree model based on the abnormal transaction adversarial game feature data; optimizing the tree structure weight parameters of the preliminary digital transaction fraud tracing tree model according to the digital transaction attention weight information corresponding to the weighted digital transaction feature data to obtain the digital transaction fraud tracing tree model; Step S6: Analyze and process the weighted digital transaction feature data based on the digital transaction fraud tracing tree model to generate digital transaction abnormality feedback data; transmit the digital transaction abnormality feedback data to the terminal to execute the digital transaction abnormality feedback operation.

2. The method for digital economic transaction processing and anomaly detection based on machine learning according to claim 1 is characterized in that: Step S1 includes the following steps: Step S11: collecting digital transaction multimodal data from the digital economic transaction platform according to a preset digital economic transaction monitoring engine to obtain digital transaction multimodal data; Step S12: performing digital transaction behavior analysis based on the digital transaction multimodal data to generate digital transaction behavior data; Step S13: performing cluster analysis on the digital transaction behavior data to generate clustered digital transaction behavior data; Step S14: performing transaction pattern recognition processing based on the clustered digital transaction behavior data to generate digital transaction pattern data; Step S15: Design a transaction pattern dimension reduction feature matrix based on the digital transaction pattern data to generate a transaction pattern dimension reduction feature matrix; Step S16: Perform digital transaction feature mapping processing on the digital transaction multimodal data according to the transaction mode dimension reduction feature matrix to generate digital transaction feature data.

3. The method for digital economic transaction processing and anomaly detection based on machine learning according to claim 2 is characterized in that: Step S11 includes the following steps: Collecting initial digital transaction multimodal data on the digital economic transaction platform according to a preset digital economic transaction monitoring engine to obtain initial digital transaction multimodal data; Performing data cleaning on the initial digital transaction multimodal data to obtain cleaned digital transaction multimodal data; Data standardization is performed on the cleaned digital transaction multimodal data to obtain digital transaction multimodal data.

4. The method for digital economic transaction processing and anomaly detection based on machine learning according to claim 2 is characterized in that: Step S2 includes the following steps: Step S21: Perform digital transaction frequency feature analysis based on the digital transaction behavior data to generate digital transaction frequency feature data; Step S22: performing digital transaction time series feature analysis based on the digital transaction multimodal data to generate digital transaction time series feature data; Step S23: Designing digital transaction attention weights according to the digital transaction frequency feature data and the digital transaction time series feature data to obtain digital transaction attention weight information; Step S24: Perform attention weighting processing on the digital transaction feature data according to the digital transaction attention weight information to generate weighted digital transaction feature data.

5. The method for digital economic transaction processing and anomaly detection based on machine learning according to claim 1, characterized in that: Step S3 includes the following steps: Step S31: Obtain historical digital trading strategies; Step S32: defining the digital trading strategy state space according to the historical digital trading strategy to obtain the digital trading strategy state space; Step S33: Analyze the transaction strategy correlation characteristics of the historical digital transaction strategies to obtain transaction strategy correlation characteristic data; Step S34: defining the digital trading strategy action space according to the trading strategy associated feature data and the digital trading strategy state space to obtain the digital trading strategy action space; Step S35: Designing a digital trading strategy network according to the digital trading strategy action space to obtain a digital trading strategy network; Step S36: Designing a digital transaction value network according to the digital transaction strategy network and the historical digital transaction strategy to obtain a digital transaction value network; Step S37: Designing a reinforcement learning mathematical model of the digital transaction strategy based on the digital transaction strategy network and the digital transaction value network to establish a digital transaction strategy reinforcement learning model; Step S38: transmitting the weighted digital transaction feature data to the digital transaction strategy reinforcement learning model for digital transaction strategy prediction processing to generate digital transaction strategy prediction data.

6. The method for digital economic transaction processing and anomaly detection based on machine learning according to claim 5 is characterized in that: Step S36 includes the following steps: Step S361: dividing the historical digital trading strategies into historical trading strategy action nodes according to the digital trading strategy network to obtain historical trading strategy action nodes; Step S362: performing action node status evaluation on the historical transaction strategy action node to obtain historical transaction strategy action node status evaluation data; Step S363: Design a digital transaction value network based on historical transaction strategy action node status evaluation data to obtain a digital transaction value network.

7. The method for digital economic transaction processing and anomaly detection based on machine learning according to claim 1, characterized in that: Step S4 includes the following steps: Step S41: performing multi-dimensional digital transaction combination analysis processing according to the digital transaction strategy prediction data to obtain multi-dimensional digital transaction combination data; Step S42: performing a multi-dimensional digital transaction fraud feature analysis based on the multi-dimensional digital transaction combination data to generate multi-dimensional digital transaction fraud feature data; Step S43: using a preset generative adversarial network model to perform fraud feature derivation processing on the multi-dimensional digital transaction fraud feature data to generate multi-dimensional digital transaction fraud feature derived data; Step S44: Designing a mathematical model of the adversarial game of abnormal transactions based on the multi-dimensional digital transaction fraud feature derived data to generate an adversarial game model of abnormal transactions.

8. The method for digital economic transaction processing and anomaly detection based on machine learning according to claim 1, characterized in that: Step S5 includes the following steps: Step S51: performing abnormal transaction adversarial game feature analysis and processing according to the abnormal transaction adversarial game model to generate abnormal transaction adversarial game feature data; Step S52: Designing digital transaction fraud tracing tree nodes according to the abnormal transaction adversarial game feature data to obtain digital transaction fraud tracing tree node data; Step S53: Designing the associated labels of the fraud tracing tree structure based on the weighted digital transaction feature data to generate the associated label data of the fraud tracing tree structure; Step S54: setting fraud tracing tree structure weight parameters according to the digital transaction attention weight information corresponding to the weighted digital transaction feature data to obtain fraud tracing tree structure weight parameters; Step S55: Designing a step-by-step fraud tracing tree structure model of digital transactions for the digital transaction fraud tracing tree node data based on the fraud tracing tree structure associated label data to establish a preliminary digital transaction fraud tracing tree model; Step S56: Optimize the tree structure weight parameters of the preliminary digital transaction fraud tracing tree model according to the fraud tracing tree structure weight parameters to obtain the digital transaction fraud tracing tree model.

9. The method for digital economic transaction processing and anomaly detection based on machine learning according to claim 8, characterized in that: Step S51 includes the following steps: Step S511: performing abnormal transaction game evolution processing according to the abnormal transaction confrontation game model to generate abnormal transaction game evolution data; Step S512: Performing game evolution replication dynamic processing according to the abnormal transaction game evolution data to generate abnormal transaction game evolution replication dynamic data; Step S513: performing game evolution equilibrium point analysis and processing according to the abnormal transaction game evolution replication dynamic data to generate game evolution equilibrium point data; Step S514: performing abnormal transaction adversarial game feature analysis on the abnormal transaction adversarial game model according to the game evolution equilibrium point data to generate abnormal transaction adversarial game feature data.

10. The method for digital economic transaction processing and anomaly detection based on machine learning according to claim 1, characterized in that: Step S6 includes the following steps: Step S61: transmitting the weighted digital transaction feature data to the digital transaction fraud tracing tree model for digital transaction anomaly prediction processing to generate digital transaction anomaly prediction data; Step S62: performing digital transaction anomaly assessment processing on the digital transaction anomaly prediction data based on a preset digital transaction anomaly assessment decision to generate digital transaction anomaly assessment data; Step S63: Mapping the digital transaction anomaly assessment data to the digital transaction anomaly prediction data to perform digital transaction anomaly feedback information integration processing to generate digital transaction anomaly feedback data; Step S64: Transmitting the digital transaction abnormality feedback data to the terminal to execute the digital transaction abnormality feedback operation.

Images