Received a Certificate of Recognition from the British Computer Society (BCS) and the Hertfordshire Students’ Union for actively participating as a member of the University of Hertfordshire Computer Science Chapter during the academic year 2023–2024. This recognition was given for my involvement and contribution to student engagement activities, supported by BCS Student Engagement Coordinator Hannah Reay-Jones and Chapter President Aravind Jalajan. It was a proud moment to be acknowledged for my dedication to the student tech community and my efforts in promoting collaboration and professional growth within the chapter.

This study, presented at the Advance Computing and Ingenious Technology in Engineering Science (ACITES 2021) held in Greater Noida, India, focuses on the accurate detection and classification of epileptic seizures using electroencephalogram (EEG) signals. Epilepsy is a neurological disorder marked by recurrent seizures, and timely detection is critical for effective treatment. The aim of this research is to develop an automated system that improves diagnostic accuracy by analyzing EEG data. The dataset used originates from Universitätsklinikum Bonn’s open-access EEG recordings, consisting of 4096 time-series samples. Data preprocessing included noise reduction and transformation using the shearlet transform. Features were extracted from the transformed signals and classified using the k-nearest neighbors (KNN) algorithm. The performance of KNN was then compared with the Adaptive Network-Based Fuzzy Inference System (ANFIS), a previously used method. The evaluation showed that KNN achieved an impressive classification accuracy of 99.3%, significantly outperforming ANFIS, which recorded 85.9%. Additionally, the proposed KNN-based approach demonstrated superior sensitivity and specificity. The experimental setup considered statistical rigor with an 80% power and 0.5 maximum accepted error calculated using ClinCalc. This work highlights the efficiency of machine learning techniques in biomedical signal processing. It also underlines the importance of feature transformation for improved classification. The research contributes to the growing field of AI-assisted medical diagnostics. The results advocate for further exploration of lightweight, interpretable models like KNN in clinical applications. Overall, the study sets a promising direction for non-invasive, data-driven epilepsy diagnosis using EEG signals.