Block Chain Technology: Anomaly Detection in Bitcoin Using RFMLPAlgorithm

Abstract

Block chain technology has been used in many different fields, including banking, logistics, healthcare, and government. However, billions have been lost to cyber attacks on block chain apps in the last several years. An efficient approach for identifying malicious behavior in block chain networks is urgently required. Anomaly detection is a well-studied issue with a lengthy history of research. Anomalies are, in a nutshell, unusual or improbable occurrences. Theft and other illicit activity in financial networks are often outliers. Participants in the network want to spot anomalies as soon as possible to safeguard the overall safety and security of the system. However, fraud and anomaly detection techniques are constantly developing along with the financial industry. Furthermore, the most secure approach being brought into money is block chain technology. The number of scams, however, increases every year alongside these innovative technology. This is why we suggested a safe Anomaly detection technique that combines Machine Learning with Deep Learning. For the purpose of classifying Anomaly transactions, we used a unique hybrid RFMLP approach. For Bitcoin transaction anomaly detection, the RFMLP combines Random Forest (RF) with multilayer perceptron (MLP). We found that our RFMLP was more accurate than competing algorithms.