A student from the Department of Mathematics Publishes a Scientific Research Paper on Presenting a Rapid Innovative Model for Analyzing Surrounding Components – a Deep Neural Network to Enhance Driver Drowsiness Detection
A student from the Department of Mathematics Publishes a Scientific Research Paper on Presenting a Rapid Innovative Model for Analyzing Surrounding Components – a Deep Neural Network to Enhance Driver Drowsiness Detection
Student Nawar Alaa Al-Samak, a third-year student in the Department of Mathematics, published a scientific research paper titled “Introducing a Rapid Innovative Model for Analyzing Surrounding Components – a Deep Neural Network to Enhance Driver Drowsiness Detection.” The paper was published in the Swiss journal Big Data and Cognitive Computing (BDCC), ranked in the first quartile (Q1) of the Clarivate WOS database and also ranked in the first quartile (Q1) of the Scopus database. The research addresses the problem of drowsiness while driving, as one of the most significant causes of traffic accidents worldwide. The research team presented an innovative hybrid model that combines fast neighbor component analysis (FNCA) and deep neural networks (DNN) to detect drowsiness using electroencephalogram (EEG) signals. The model improves feature recognition and dimensionality reduction using FNCA, then accurately classifies mental states (alert, moderate, and sleepy) using a DNN. The model was tested on the SEED-VIG database and achieved an accuracy of 94.29%, outperforming state-of-the-art models such as TSception and CNN+LSTM. The results confirm the model’s effectiveness in instantly detecting drowsiness, making it a promising tool for enhancing intelligent vehicle safety systems. We wish the student all the best and success in his academic career.
