Article: Classifying blockchain cybercriminal transactions using hyperparameter tuned supervised machine learning models Journal: International Journal of Computational Science and Engineering (IJCSE) 2023 Vol.26 No.6 pp.615 - 626 Abstract: Bitcoin is a crypto asset with transactions recorded on a decentralised, publicly accessible ledger. The real-world identity of the bitcoin blockchain users is masked behind a pseudonym, known as an address that provides a high level of anonymity, which is one of the reasons for its widespread use in criminal operations such as ransomware attacks, gambling, etc. As a result, the classification of diverse cybercriminal users' activities and addresses in the bitcoin blockchain is demanded. This research work presents a classification of user activities and addresses associated with illicit transactions using supervised machine learning (ML). The labelled dataset samples are trained using decision trees, ensemble, Bayesian, and instance-based learning. Extra Trees emerged as the best classification model, whereas Gaussian naïve Bayes as the worst. GridSearchCV is employed to optimise the CV accuracy of classification models with CV accuracy below 85% which led to an improvement in the CV accuracy. Inderscience Publishers - linking academia, business and industry through research

Title: Classifying blockchain cybercriminal transactions using hyperparameter tuned supervised machine learning models

Authors: Rohit Saxena; Deepak Arora; Vishal Nagar

Addresses: Department of Computer Science and Engineering, Amity School of Engineering and Technology, Amity University Uttar Pradesh, Lucknow Campus, India ' Department of Computer Science and Engineering, Amity School of Engineering and Technology, Amity University Uttar Pradesh, Lucknow Campus, India ' Department of Computer Science and Engineering, Pranveer Singh Institute of Technology, Bhauti, Kanpur, India

Abstract: Bitcoin is a crypto asset with transactions recorded on a decentralised, publicly accessible ledger. The real-world identity of the bitcoin blockchain users is masked behind a pseudonym, known as an address that provides a high level of anonymity, which is one of the reasons for its widespread use in criminal operations such as ransomware attacks, gambling, etc. As a result, the classification of diverse cybercriminal users' activities and addresses in the bitcoin blockchain is demanded. This research work presents a classification of user activities and addresses associated with illicit transactions using supervised machine learning (ML). The labelled dataset samples are trained using decision trees, ensemble, Bayesian, and instance-based learning. Extra Trees emerged as the best classification model, whereas Gaussian naïve Bayes as the worst. GridSearchCV is employed to optimise the CV accuracy of classification models with CV accuracy below 85% which led to an improvement in the CV accuracy.

Keywords: blockchain; bitcoin; supervised machine learning; classification; pseudo-anonymity; anonymity; GridSearchCV.

DOI: 10.1504/IJCSE.2023.135281

International Journal of Computational Science and Engineering, 2023 Vol.26 No.6, pp.615 - 626

Received: 01 Apr 2022
Received in revised form: 12 Jun 2022
Accepted: 25 Jul 2022
Published online: 04 Dec 2023 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article

Title: Classifying blockchain cybercriminal transactions using hyperparameter tuned supervised machine learning models

Keep up-to-date