Classification of the Smartphone Addiction using the Artificial Neural Network

Abstract

Smartphones, as communication tools, have evolved and have a basic physical factor that makes them portable. Smartphones are fascinating devices that may quickly become addictive to their owners. Using an Artificial Neural Network (ANN) algorithm, this study aims to diagnose smartphone addiction based on self-control. The classification findings can be used to decide who should participate in a self-control campaign. The classification result could be used by the school to predict which pupils should be motivated to improve their self-control to avoid smartphone addiction. The study is carried out by creating and verifying the questionnaire, collecting the dataset, and categorizing the smartphone addiction based on the self-control. The findings of data collection yielded 168 participants who filled out the questionnaires that were distributed. These participants' responses were then collated and utilized as a dataset. The SAS response scores were then totaled and classified into low, medium, and high criteria. In this investigation, this criterion is chosen as the target class. The target class in this study was split into 54, 64, and 61 data points that were classified as low, medium, and high criteria, respectively. The accuracy of the algorithm on classifying the smartphone addiction is and 85.29% and 81.81 % for the training and testing, respectively. This result can be categorized as a good result.