ONCS: Experimental Design to Predict Students Electronic Gadget Addiction Using Optimized Neural Classification Scheme

In the last few years, artificial intelligence has significantly influenced children's education and behavioral activities, with one of the most prevalent applications being the prediction of students' digital device addiction based on activity data, including device usage duration, gaming habits, reactions to device deprivation, and social standing. This study introduces and develops an optimized neural classification protocol (ONC-M) model that can tell if a student is hooked on a certain gadget activity by using data from an expert questionnaire that could put students' privacy at risk. We divided the gadget addiction symptoms into two distinct categories. In order to get a sample of training and test data that is representative of the whole, a total of 512 questionnaires, each measuring 8 question sets x 64 target students, were distributed to parents, who were the primary responders. In order to gain valuable insights into the prevalence of these symptoms in children and the severity of their manifestations, we gathered responses that supplied us with relevant information. The collection of these data will make it possible to conduct additional research on effective intervention strategies and support networks for families that are affected. 98.40% accuracy, 96.50% precision, 91.25% recall, and 93.42% F1-score are the attributes that were chosen for the model from among those that were available. In light of these findings, the student authorities will be able to take measures to prevent students from becoming addicted to digital gadgets.