EEG-Based Person Identification Using Multi-Verse Optimizer as Unsupervised Clustering Techniques

Zaid Abdi Alkareem Alyasseri,João P Papa,Salwani Abdullah,Mohammed Azmi Al-Betar,Sharif Naser Makhadmeh,Ammar Kamal Abasi,Ahamad Tajudin Khader

doi:10.1007/978-981-33-4191-3_4

Zaid Abdi Alkareem Alyasseri, João P Papa + Show 5 more

https://doi.org/10.1007/978-981-33-4191-3_4

Copy DOI

Export

Save

Cite

Abstract
Full-Text
Similar Papers

Abstract

Listen

Recently, electroencephalogram (EEG) signal provides great potential for identification systems. Many studies have shown that the EEG introduces unique, universal features and natural robustness for spoofing attacks. The EEG represents a graphic recording of the electrical activity of the brain that can be measured by placing sensors (electrodes) at different locations on the scalp. This chapter proposes a new technique using unsupervised clustering and optimization techniques for user identification-based EEG signals. The proposed method employs four algorithms which are Genetic Algorithm (GA), Multi-verse Optimizer (MVO), Particle Swarm Optimization (PSO) and the k-means algorithm. A standard EEG motor imagery dataset is used to evaluate the proposed method’s performance, and its results are evaluated using four criteria: (i) Precision, (ii) Recall, (iii) F-Score and (v) Purity. It is worth mentioning that this work is one of the first to employ optimization methods with unsupervised clustering methods for person identification using EEG. As a conclusion, the MVO algorithm achieved the best results compared with GA, PSO and k-means. Finally, the proposed method can draw future directions to apply to different research areas.

Full Text