Block chain transaction address overseas and overseas identification system and method

Abstract

The invention discloses a blockchain transaction address overseas and overseas identification system and method, and the system comprises a data obtaining module which is used for obtaining blockchain transaction data in real time, and the transaction data comprises a transaction hash, a timestamp, a transaction amount and addresses of two transaction parties; a data processing module connected with the data acquisition module and used for preprocessing the acquired transaction data, including data cleaning, formatting and preliminary screening; the feature engineering module is connected with the data processing module and is used for extracting key features from the preprocessed transaction data and constructing a feature set for machine learning training; and high-precision identification: through comprehensive utilization of multi-dimensional features such as transaction time, amount and frequency, the accuracy of address identification is greatly improved, domestic and foreign user addresses can be accurately identified and distinguished, and the method has important significance for the public security department to rapidly position involved addresses, especially distinguishing domestic and foreign criminal activities.

Description

Technical Field

[0001] The present invention relates to the technical fields of blockchain technology, data analysis technology, and machine learning technology, and in particular, to a system and method for identifying and verifying the geographical location of exchange user addresses in a blockchain network. Background Art

[0002] With the wide application of blockchain technology globally, although its inherent anonymity and decentralization characteristics provide significant security advantages, they also bring challenges in supervision and law enforcement. Traditional geographical location identification methods based on official investigation reply letters not only consume a large amount of time and resources, but also have relatively limited benefits, and there are defects such as data privacy issues, high time costs, cooperation dependencies, and limited scope of application.

[0003] Therefore, there is an urgent need for a technical solution that can accurately and efficiently identify the geographical location of users without infringing on user privacy to solve the problems raised in the above background art. Summary of the Invention

[0004] The present invention provides a system and method for identifying the geographical location of blockchain exchange user addresses within the territory. The system deeply analyzes blockchain transaction data and combines advanced data mining and machine learning technologies to accurately identify the geographical location of user addresses. It is particularly applicable to the fields of financial supervision and legal compliance, and can effectively support the monitoring requirements of anti-money laundering and combating illegal activities such as virtual currency money laundering.

[0005] To achieve the above object, the present invention provides the following technical solutions: A system for identifying the geographical location of blockchain transaction addresses inside and outside the territory, comprising: A data acquisition module, configured to acquire blockchain transaction data in real time, where the transaction data includes transaction hash, timestamp, transaction amount, and addresses of both parties to the transaction; A data processing module, connected to the data acquisition module, configured to preprocess the acquired transaction data, including data cleaning, formatting, and preliminary screening; A feature engineering module, connected to the data processing module, configured to extract key features from the preprocessed transaction data and construct a feature set for machine learning training; A model training module, connected to the feature engineering module, using the XGBoost algorithm as the core classifier and optimizing hyperparameters through the CMA-ES algorithm, and evaluating the model performance using cross-validation techniques; A prediction module, connected to the model training module, configured to predict newly collected exchange user addresses and output a probability score indicating whether the address is within the territory of China; The application module is connected to the prediction module and assists financial regulatory agencies and exchanges in implementing location-related laws and regulations based on the prediction results.

[0006] As a further solution of the present invention: The data acquisition module ensures the integrity and accuracy of the acquired data by connecting to the blockchain network nodes.

[0007] As a further solution of the present invention: The data processing module uses a specific data cleaning algorithm to remove invalid or abnormal data records and formats the data.

[0008] As a further solution of the present invention: The key features extracted by the feature engineering module include transaction frequency, amount statistical characteristics, and transaction time series analysis, etc.

[0009] As a further solution of the present invention: When optimizing the hyperparameters, the model training module dynamically adjusts the optimization strategy according to the scale and feature distribution of the training dataset.

[0010] As a further solution of the present invention: The prediction module classifies and judges the output probability score according to a preset threshold.

[0011] As a further solution of the present invention: The application module generates location-related reports and warning information.

[0012] A method for identifying domestic and overseas blockchain transaction addresses, characterized by comprising the following steps: Step 1: Real-time acquire blockchain transaction data through the data acquisition module; Step 2: Preprocess the acquired transaction data through the data processing module; Step 3: Extract key features from the preprocessed transaction data through the feature engineering module to construct a feature set; Step 4: Train a model using the XGBoost algorithm through the model training module, optimize the hyperparameters, and evaluate the model performance; Step 5: Predict the newly collected exchange user addresses through the prediction module and output a probability score; Step 6: Assist in implementing relevant laws and regulations through the application module according to the prediction results.

[0013] As a further solution of the present invention: In step 1, acquire data by connecting to the blockchain network nodes.

[0014] As a further solution of the present invention: In step 4, dynamically adjust the hyperparameter optimization strategy according to the scale and feature distribution of the training dataset.

[0015] Compared with the prior art, the present invention has the following remarkable advantages: High-precision identification: By comprehensively utilizing multi-dimensional features such as transaction time, amount, and frequency, the accuracy of address identification has been significantly improved. It can accurately identify and distinguish domestic and foreign user addresses, which is of great significance for the public security department to quickly locate the addresses involved in cases, especially to distinguish domestic and foreign criminal activities.

[0016] Efficiency improvement: The automated data processing and prediction process significantly reduces the need for manual review and improves the processing speed. The system can analyze transaction activities, quickly generate investigation leads and evidence, and accelerate the progress of cases.

[0017] Privacy protection: The method does not rely on any personal sensitive information, ensuring the privacy and security of users. Geographical location identification is carried out without infringing on personal privacy, which complies with data protection regulations.

[0018] Optimized resource allocation: Through accurate geographical location prediction, the public security department can allocate investigation resources more effectively, prioritize high-risk or high-value cases, especially the investigation and evidence collection for domestic addresses involved in cases.

[0019] Reduced misjudgment rate: Advanced data analysis and machine learning models reduce misjudgments caused by human judgment biases, improving the accuracy of investigation and supervision. Description of the drawings

[0020] Figure 1 It is a schematic flow diagram of a system for identifying domestic and foreign blockchain transaction addresses.

[0021] Figure 2 It is a schematic flow diagram of a system for identifying domestic and foreign blockchain transaction addresses.

[0022] Figure 3 It is a schematic diagram of a system for identifying domestic and foreign blockchain transaction addresses.

[0023] Figure 4 It is a page of a system for identifying domestic and foreign blockchain transaction addresses where users input addresses, call models, and output results. Detailed implementation manners

[0024] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0025] Embodiment 1 Please refer to Figures 1 - 4, the domestic and overseas identification system for blockchain transaction addresses of the present invention includes a data acquisition module, a data processing module, a feature engineering module, a model training module, a prediction module, and an application module.

[0026] Data acquisition module: By connecting to the blockchain network nodes, it acquires blockchain transaction data in real time, including transaction hashes, timestamps, transaction amounts, and the addresses of both parties to the transaction, etc. This module ensures the integrity and accuracy of the acquired data, providing a reliable data basis for subsequent data processing and analysis. For example, in the Bitcoin blockchain network, the data acquisition module can connect to the API interface of Bitcoin nodes to acquire the latest transaction data in real time.

[0027] Data processing module: Connects to the data acquisition module and preprocesses the acquired transaction data, including data cleaning, formatting, and preliminary screening. This module uses specific data cleaning algorithms to remove invalid or abnormal data records, such as removing abnormal data with negative transaction amounts or unreasonable transaction times. At the same time, it formats the data to meet the requirements of subsequent feature engineering, for example, converting the transaction timestamp to a unified time format.

[0028] Feature engineering module: Connects to the data processing module, extracts key features from the preprocessed transaction data, and constructs a feature set for machine learning training. The key features extracted include transaction frequency, amount statistical characteristics, and transaction time series analysis, etc. For example, transaction frequency can reflect the trading activity of users within a certain period of time, amount statistical characteristics can include the mean, variance, etc. of transaction amounts, and transaction time series analysis can reveal the trading time patterns of users. By extracting and analyzing these features, rich information can be provided for model training, improving the recognition accuracy of the model.

[0029] Model training module: Connects to the feature engineering module, uses the XGBoost algorithm as the core classifier, optimizes the hyperparameters through the CMA-ES algorithm, and uses cross-validation technology to evaluate the model performance. When optimizing the hyperparameters, it dynamically adjusts the optimization strategy according to the scale and feature distribution of the training dataset to improve the training efficiency and performance of the model. For example, when the training dataset is large and the feature distribution is relatively complex, the number and depth of decision trees can be appropriately increased to improve the fitting ability of the model; when the training dataset is small and the feature distribution is relatively simple, the number and depth of decision trees can be appropriately reduced to avoid overfitting. Through cross-validation technology, the performance of the model can be comprehensively evaluated to ensure that the model has good generalization ability and high accuracy.

[0030] Prediction Module: Connected to the model training module, it is used to predict newly collected exchange user addresses and output the probability score of whether the address is within China. This module classifies and judges the output probability score according to a preset threshold. For example, when the probability score is greater than 0.5, it is judged that the address is within China; when the probability score is less than or equal to 0.5, it is judged that the address is outside China. By predicting newly collected user addresses, potential risk addresses can be discovered in a timely manner, providing decision-making support for financial regulatory agencies and exchanges.

[0031] Application Module: Connected to the prediction module, it assists financial regulatory agencies and exchanges in implementing geographical location-related laws and regulations according to the prediction results, and generates geographical location-related reports and warning information. For example, when it is found that a certain user address is within China and there are suspicious trading behaviors, the application module can generate corresponding reports and warning information to remind financial regulatory agencies and exchanges to conduct further investigations and handling. Through the application of the application module, the regulatory efficiency and accuracy can be effectively improved, supporting the monitoring needs of anti-money laundering and combating illegal activities such as virtual currency money laundering.

[0032] Method Steps The method for identifying domestic and overseas blockchain transaction addresses of the present invention includes the following steps: Step 1: Real-time obtain blockchain transaction data through the data acquisition module. As described above, the data acquisition module ensures the integrity and accuracy of the acquired data by connecting to blockchain network nodes.

[0033] Step 2: Preprocess the acquired transaction data through the data processing module. The data processing module uses a specific data cleaning algorithm to remove invalid or abnormal data records and formats the data to meet the requirements of subsequent feature engineering.

[0034] Step 3: Extract key features from the preprocessed transaction data through the feature engineering module to construct a feature set. The extracted key features include transaction frequency, amount statistical characteristics, and transaction time series analysis, etc. By extracting and analyzing these features, rich information is provided for model training.

[0035] Step 4: Train the model using the XGBoost algorithm through the model training module, optimize the hyperparameters, and evaluate the model performance. When optimizing the hyperparameters, dynamically adjust the optimization strategy according to the scale and feature distribution of the training data set to improve the training efficiency and performance of the model. Through cross-validation technology, comprehensively evaluate the performance of the model to ensure that the model has good generalization ability and high accuracy.

[0036] Step 5: Predict the newly collected exchange user addresses through the prediction module and output probability scores. The prediction module classifies and judges the output probability scores according to a preset threshold to timely discover potential risk addresses.

[0037] Step 6: Through the application module, assist in implementing relevant laws and regulations according to the prediction results, generate reports and warning information related to geographical locations, provide decision-making support for financial regulatory agencies and exchanges, and effectively improve the regulatory efficiency and accuracy.

[0038] In summary, the domestic and overseas identification system and method for blockchain transaction addresses of the present invention realize high-precision identification and efficient prediction of the geographical locations of user addresses through in-depth analysis of blockchain transaction data and application of machine learning technology, and have broad application prospects and important practical significance.

[0039] The above is only a preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention, according to the technical solution and inventive concept of the present invention, makes equivalent substitutions or changes, and should be covered by the protection scope of the present invention.

Claims

1. A domestic and overseas identification system for blockchain transaction addresses, characterized in that, including; a data acquisition module for real-time acquisition of blockchain transaction data, the transaction data including transaction hash, timestamp, transaction amount, and addresses of both parties to the transaction; a data processing module connected to the data acquisition module for preprocessing the acquired transaction data, including data cleaning, formatting, and preliminary screening; a feature engineering module connected to the data processing module for extracting key features from the preprocessed transaction data to construct a feature set for machine learning training; a model training module connected to the feature engineering module, using the XGBoost algorithm as the core classifier and optimizing hyperparameters through the CMA-ES algorithm, and evaluating model performance using cross-validation techniques; a prediction module connected to the model training module for predicting newly collected exchange user addresses and outputting a probability score indicating whether the address is within China; an application module connected to the prediction module for assisting financial regulatory agencies and exchanges in implementing laws and regulations related to geographical location based on the prediction results.

2. The domestic and overseas identification system for blockchain transaction addresses according to claim 1, wherein The data acquisition module ensures the integrity and accuracy of the acquired data through connection with blockchain network nodes.

3. The cross-border identification system for blockchain transaction addresses according to claim 1, characterized in that The data processing module uses a specific data cleaning algorithm to remove invalid or abnormal data records and performs formatting on the data.

4. The blockchain transaction address domestic and overseas identification system according to claim 1, characterized in that The key features extracted by the feature engineering module include transaction frequency, amount statistical characteristics, and transaction time series analysis, etc.

5. The domestic and overseas identification system for blockchain transaction addresses according to claim 1, characterized in that, When optimizing hyperparameters, the model training module dynamically adjusts the optimization strategy according to the scale and feature distribution of the training dataset.

6. The domestic and overseas identification system for blockchain transaction addresses according to claim 1, wherein The prediction module makes a classification judgment on the output probability score according to a preset threshold.

7. The onshore and offshore identification system for blockchain transaction addresses according to claim 1, characterized in that The application module generates reports and warning information related to geographical location.

8. A method for identifying domestic and overseas blockchain transaction addresses, characterized in that, including the following steps: Step 1: Real-time acquisition of blockchain transaction data through the data acquisition module; Step 2: Preprocessing the acquired transaction data through the data processing module; Step 3: Extracting key features from the preprocessed transaction data through the feature engineering module to construct a feature set; Step 4: Training a model using the XGBoost algorithm through the model training module, optimizing hyperparameters, and evaluating model performance; Step 5: Predicting newly collected exchange user addresses through the prediction module and outputting a probability score; Step 6: Assisting in implementing relevant laws and regulations according to the prediction results through the application module.

9. A method for identifying domestic and overseas blockchain transaction addresses according to claim 8, characterized in that, In Step 1, data is acquired through connection with blockchain network nodes.

10. The blockchain transaction address domestic and overseas identification system according to claim 8, wherein, In Step 4, the hyperparameter optimization strategy is dynamically adjusted according to the scale and feature distribution of the training dataset.

Images