Toward characterizing blockchain-based cryptocurrencies for highly accurate predictions

Muhammad Saad, Jinchun Choi, Daehun Nyang, Joongheon Kim, Aziz Mohaisen

Research output: Contribution to journalArticlepeer-review

77 Scopus citations

Abstract

Recently, the Blockchain-based cryptocurrency market witnessed enormous growth. Bitcoin, the leading cryptocurrency, reached all-time highs many times over the year leading to speculations to explain the trend in its growth. In this article, we study Bitcoin and Ethereum and explore features in their network that explain their price hikes. We gather data and analyze user and network activity that highly impact the price of these cryptocurrencies. We monitor the change in the activities over time and relate them to economic theories. We identify key network features that help us to determine the demand and supply dynamics in a cryptocurrency. Finally, we use machine learning methods to construct models that predict Bitcoin price. Based on our experimental results using two large datasets for validation, we confirm that our approach provides an accuracy of up to 99% for Bitcoin and Ethereum price prediction in both instances.

Original languageEnglish
Article number8840919
Pages (from-to)321-332
Number of pages12
JournalIEEE Systems Journal
Volume14
Issue number1
DOIs
StatePublished - Mar 2020

Bibliographical note

Funding Information:
Manuscript received September 14, 2018; revised January 15, 2019 and April 16, 2019; accepted June 19, 2019. Date of publication September 17, 2019; date of current version March 2, 2020. This work was supported in part by National Research Foundation of Korea under Grants NRF-2016K1A1A2912757 and 2017R1A4A1015675 and in part by a Chung-Ang University Research Grant (2019). This article was presented in part at the Hot Topics in Pervasive Mobile and Online Social Networking (HotPOST 2018), Honolulu, HI, USA, April 2018. (Corresponding authors: Joongheon Kim; Aziz Mohaisen.) M. Saad and A. Mohaisen are with the University of Central Florida, Orlando, FL 32816 USA (e-mail: saad.ucf@knights.ucf.edu; mohaisen@cs.ucf.edu).

Publisher Copyright:
© 2007-2012 IEEE.

Keywords

  • Bitcoin
  • Blockchain
  • Ethereum
  • Prediction

Fingerprint

Dive into the research topics of 'Toward characterizing blockchain-based cryptocurrencies for highly accurate predictions'. Together they form a unique fingerprint.

Cite this