Abstract:
"In the rapidly evolving digital era, digital devices including smartphones, smart gadgets, tablets, and laptops, have become daily necessities for humans to learn, entertain and engage in social interactions. However, there is growing concern that the problematic use of digital devices can lead to the emergence of a noteworthy phenomenon of Screen Dependency Disorder (SDD). This research will elaborate on the significance of SDD in the context of the digital age, emphasising the implications for mental health and well-being. It also highlights the potential of Electroencephalogram (EEG) screening as a diagnostic tool and the significance of the deep learning approach to detect the severity scale of SDD in individuals objectively and accurately.
Developing a deep learning model for early detection of Screen Dependency Disorder (SDD) presents a critical challenge in addressing the growing prevalence of digital addiction. This research proposes a novel approach utilizing LSTM (Long Short-Term Memory) model with inputs from both questionnaire responses and Electroencephalogram (EEG) signal data provided in CSV format. By leveraging deep learning techniques, the aim is to automate the manual detection process, saving time, effort, and reducing human errors. The model is trained on a dataset containing labelled instances of SDD, aiming for binary classification. Hyperparameter tuning, architecture layer modifications, and pre processing techniques are applied to enhance model performance. "
