Sina Ahmadi, Rico Sennrich, Erfan Karami, Ako Marani, Parviz Fekrazad, Gholamreza Akbarzadeh Baghban, Hanah Hadi, Semko Heidari, Mahîr Dogan, Pedram Asadi, Dashne Bashir, Mohammad Amin Ghodrati, Kourosh Amini, Zeynab Ashourinezhad, Mana Baladi, Farshid Ezzati, Alireza Ghasemifar, Daryoush Hosseinpour, Behrooz Abbaszadeh, Amin Hassanpour, Bahaddin Jalal Hamaamin, Saya Kamal Hama, Ardeshir Mousavi, Sarko Nazir Hussein, Isar Nejadgholi, Mehmet Ölmez, Horam Osmanpour, Rashid Roshan Ramezani, Aryan Sediq Aziz, Ali Salehi, Mohammadreza Yadegari, Kewyar Yadegari, Sedighe Zamani Roodsari. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2025.

Epilepsy is a brain disorder that can significantly affect a patient's health. Therefore, seizure prediction techniques have gained a lot of attention to minimize the potential damages caused by epilepsy and improve the quality-of-life of epileptic patients. In this paper, an algorithm based on the linear Support Vector Machine (SVM) tool was proposed to classify intracranial electroencephalography (iEEG) signals as ictal or interical, in order to efficiently perform human seizure prediction. One of the most important parameters in predicting seizure is the size of the sliding window, whose optimization may significantly affect performance, as well as overlapping between windows. In this study, an optimum sliding window and overlapping rate are proposed for efficient seizure prediction. They allow accurate prediction of seizure events from a large set of EEG data. Applied to iEEG recordings of eight patients in the Freiburg EEG database, the proposed approach exhibits a sensitivity of 68% and specificity of 100% using 2-second-long window and 50% overlapping via 10 fold-cross validation.