Mitigation of Trust-Related Issues in Cryptocurrency Payments Using Machine Learning: A Review

Abstract

Cryptocurrency is a type of fiat currency in digital form, unlike the physical money that is commonly used for daily purposes. A blockchain is a base on which a cryptocurrency operates, i.e., it is a growing list of records of transactions happening in a particular cryptocurrency. Trust in a cryptocurrency comes into the picture when two stakeholders, virtually unknown to each other, are confident or not about each other’s reliability in the context of whether each one is getting the service they intended to get. Trust in cryptocurrency can exist between any two stakeholders, such as users, merchants, government agencies, and blockchain technology, who are a part of cryptocurrency transactions. Furthermore, direct or indirect involvement of different stakeholders in cryptocurrency transactions results in issues such as lack of transparency, ease of use, regulations of the government, privacy, security of users, etc. Traditional approaches to anomaly detection in blockchain primarily use machine learning methods because they can infer patterns from historical data to give decent accuracy on test data. This survey presents trust in a cryptocurrency payment and its issues. Furthermore, it also shows the mitigation approaches which use machine learning techniques to address these issues. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.