Predicting Smartphone Vision Syndrome: A Feasible Approach using Machine Learning Algorithms

Abstract

Smartphone Vision Syndrome (SVS) is an evitable problem for people who spend a great deal of time watching digital screens. It is a major concern for rapid growth in technology where the burden is significantly greater due to factors such as limited access to and use of personal protective equipment, as well as lesser break time. The objective of the model is to achieve a feasible and higher level of eye health for people who are working long hours with digital screens. The dataset is obtained through an online survey form containing metrics that contribute to the occurrence of SVS. After applying Machine Learning algorithms, namely Logistic Regression, Random Forest Classifier, Naïve Bayes and Support Vector Machine (SVM), the model’s overall performance is assessed using the test sample. Accuracies obtained by Random Forest, Support Vector Machine, Logistic Regression, Naïve Bayes, and Gaussian Naïve Bayes are 98.75%, 97.5%, 77.5%, 95% and 96.25%.