Work

Privacy and security preservation of bitcoin digital currency transactions on mixing services using enhanced heuristic-aided optimal weighted features and hybrid deep learning network

Year: 2026
Type: article
Abstract: Bitcoin (BTC) is a Peer-to-Peer (P2P) digital currency that facilitates transactions using pseudonymous identities. Mixing services improve the anonymity of transactions but are misused for illicit activities such as money laundering, raising security concerns in the BTC ecosystem. This research implements a hybridized deep learning approach for privacy preservation in BTC transactions. The required data are gathered from Elliptic Dataset and Ethereum Fraud Detection dataset, and these datasets are used for evaluation. The features are extracted from these datasets, followed by weighted attribute selection. The optimal feature selection and fine-tuning of the weights are implemented using the Improved Current Mussel Time-based Eurasian Oystercatcher Optimizer (ICMT-EOO) algorithm. The weighted attributes are classified using a Faster Regions-based Convolutional Neural Network with Dilated Bidirectional Long Short-Term Memory (FRCNN-DBi-LSTM). Experimental evaluation is performed using existing classifiers and optimization algorithms to demonstrate the efficiency of the approach. The proposed method achieves an accuracy of 95.2%, an F1-score of 92.98%, a recall of 95.21%, and a precision of 90.84%, respectively. The results confirm that the developed model achieves improved performance and enhanced security compared to existing methods.
Language: English
Open Access status: gold
Estimated APC paid: $2,690
OpenAlex ID: W7225992046
FWCI:
Cites: 1
Cited by:
Related to: 10