Prediction of Facebook Addiction Using Machine Learning

Abstract

In this modern age, social media is a part and parcel of our everyday life. Our daily life is undergoing multiple changes that directly have positive and negative effects. Social media has a positive and negative side too. Sometimes it is helpful but sometimes harmful. Facebook is one of the social media which uses mostly comparing to others. There are some changes in our daily life by using it. We are not concern about our timetable and we are scrolling Facebook without any need. The main reason behind Facebook is communication. Now it is not only communication but also wasting some valuable time. It is called addiction and most people eat, sleep and repeat Facebook. It is quite hard to predict Facebook addiction without collecting and testing data. Prediction is much easier when there is “Machine Learning”. The main purpose of the current study is to predict the addiction of people on Facebook and awareness of their daily routine with high accuracy. For the prediction of Facebook addiction, we use seven ‘Machine Learning’ classification algorithms such as Support Vector Machine (SVM), Decision Tree (DT), Naïve Bayes (NB), Linear Regression (LR), Logistic Regression (LR), Linear Discriminant Analysis (LDA), k-Nearest Neighbors (kNN). After using algorithms, we use Principal Component Analysis (PCA) for reducing data mathematically. Our finding demonstrates that kNN with a greater accuracy rate (93.53%) outperforms the k-Nearest Neighbors.