Project Abstract

This project explores the feasibility and efficacy of utilizing electroencephalogram (EEG) signals for biometric person recognition, with a focus on machine learning methodologies. The objective is Conduct comprehensive research to understand the characteristics and fine distinction of EEG signals as potential biometric identifiers. My research focuses on exploring various machine learning methods, including CNNs, RNNs, and other deep learning approaches. These methods are carefully examined to see how well they can classify EEG signals and match them with individual identities. That is, these models are trained on EEG data to learn the unique brainwave patterns associated with individual identities. Performance evaluation of the proposed system is conducted rigorously, assessing accuracy, precision, recall, and other metrics under various conditions. Challenges and opportunities associated with EEG-based biometric recognition, such as signal variability and noise interference, are identified, along with strategies for further research and refinement.The methodology involves data collection from the real world or existing data source, model development using machine learning techniques, evaluation using cross-validation methods, and identify the challenges, limitations, pros, and cons associated with utilizing EEG signals for biometric recognition.Overall, this research aims to provide insights into the potential of EEG signals for biometric recognition, leveraging machine learning approaches.

Keywords: Algorithms, Machine Learning, Statistical Analysis

 Conference Details

Session: Presentation Stream 32 at Presentation Slot 3

Location: GH037 at Wednesday 8th 13:30 – 17:00

Markers: Thomas Reitmaier, Raziyeh Moghaddas (GTA)

Course: MSc Computer Science, Masters PG

Future Plans: I’m looking for work