<![CDATA[Blockchain technologies have gained a huge amount of importance in recent years, and the use of the blockchain concept in cryptocurrency transactions has always gained the faith of industrial standards. Ethereum is a blockchain platform that allows customers to conduct cryptocurrency transactions, which are then used to build and deploy the API using smart contracts. Blockchain can be used to change the value of money in crypto exchanges and banking systems. even though the blockchain system is consistent and reliable. Attackers still try to steal the money by executing well-known techniques like Ponzi scheme attacks or by using malware software. As the participants in the Ethereum platform are "anonymous," users can access multiple accounts under the same hash identity. As a result, it will be difficult to find the malicious users who are contributing to the fraudulent activities. Although activities such as Ponzi schemes are to be monitored by the authority in order to keep the API safe from scammers and the platform legitimate, In this paper, we detect malicious transaction nodes with the help of machine learning-based anomaly detection and also give the structural architecture for creating a secure API wallet, which solves the basic security protection from Ponzi-scheme multiple identity attacks by introducing KYC contracts in smart contracts of the Ethereum platform such that no duplicate users can misuse the cryptocurrencies. In this case, we use two different machine learning models that detect with a 95.24% accuracy and a 0.88% false positive ratio. and then we compare the capabilities of random forest and support vector machine classifiers to identify the anomaly-based accounts, which are on datasets of around 300 accounts. By introducing HD-Wallets for API, we show the rules for digital wallets and cryptocurrency transactions that protect against malware.]]>
