In recent years,cryptocurrencies represented by Bitcoin and Ethereum have been loved by the public for their decentralization,anonymity,and immutability.However,the anonymity they promote is pseudo-anonymity,and all transaction data is open and transparent.By analyzing the transaction data,some connections between transactions,addresses,and users can be found,so as to achieve deanonymization to a certain extent,resulting in the leakage of user privacy.The complete anonymity protection is easy to be used by criminals to carry out illegal activities such as money laundering and black market transactions,so it needs reasonable and effective supervision.Therefore,it is of great significance to design a method that combines transaction privacy protection and supervision mechanism,and it is also a research hotspot at home and abroad.Combining the advantages of address classification method and address clustering method,the thesis proposes a transaction data analysis method based on machine learning.Features are further extracted from historical transaction information,the existing feature set is expanded,and the required feature subset is selected through a fast feature screening method based on dichotomy.Using machine learning algorithms to build a classification model to identify address types,and select different types of addresses for cluster analysis to find out other addresses controlled by the same entity as an address,which proves that there is a problem of transaction privacy leakage in cryptocurrency.For the problem of privacy disclosure in cryptocurrency,this thesis summarizes and analyzes the existing privacy protection methods for cryptocurrency transactions.Aiming at the problems such as weak anonymity protection and lack of regulation over illegal transactions,this thesis proposes a regulatable approach to transactional privacy preservation,which achieves the goal of protecting legal transactions and regulating illegal transactions.By adding a third-party cryptocurrency mix service provider,the scheme uses blind signature to hide the relationship between the user’s transaction input address and the fund receiving address,so that the transaction input and output addresses can not be linked.By setting up a regulation group,a more reasonable and fair regulation mechanism is realized by using the idea of threshold secret sharing and voting to prevent illegal persons from using it to conduct illegal transactions.Through experiments and theoretical analysis,it is proved that the scheme can provide users with better anonymous protection services,and has good security. |