MMS • RSS
Article originally posted on Data Science Central. Visit Data Science Central
I have just completed my Master of Science in Data Science program at SMU. One of the requirements of this program was to complete a research project called a Capstone project, sort of like a thesis but a bit different. Teaming up with two other students and two outside advisors, we completed the project. We chose to research trust in cryptocurrency. We came up with what we think can be the foundation for enhancing trust in the cryptocurrency marketplace. Creating a reputation score for each user can provide more trustworthy transactions while still keeping the anonymity of the users. Below is the abstract of our paper.
Trust is paramount for the effective operation of any monetary system. While the distributed architecture of blockchain technology on which cryptocurrencies operate has many benefits, the anonymity of users on the blockchain has provided criminal users an opportunity to hide both their identities and illicit activities. In this paper, we present a scoring mechanism for cryptocurrency users where the scores represent users’ trustworthiness as safe or risky transactors in the cryptocurrency community. In order to distinguish law-abiding users from potential threats in the Bitcoin marketplace, we analyze historical thefts to profile transactions, classify them into risky and non-risky categories using several machine learning techniques, and finally calculate a reputation score for every unique user based on their past association with any unlawful Bitcoin incident. The Support Vector Machine model based on two key attributes produces an accuracy of 86% and is considered the most applicable for our dataset. Our reputation score ranges from 0 to the total number of transactions by a given user where a higher score indicates greater trustworthiness in making Bitcoin transactions. This score helps to identify reputable users and, therefore, acts as a guideline for safe Bitcoin transactions. In the cryptocurrency marketplace, our self-attestation metric in the form of a reputation score offers a foundation for enhancing trust between transacting parties.
If you are interested in our work, below is the link to our paper. Also, please leave some comments and let us know your thoughts on our work!