Analysis of Data Using Machine Learning Approaches in Social Networks

Abstract

The amount of data circulating on the Internet is increasing day by day. With the increasing use of social media in particular, the importance of analyzing these data is increasing. The use of machine learning approaches to analyze large amounts of data is still popular today. Today, the social network Facebook is the most popular social networking sites. In this study, some data taken on Facebook were analyzed by machine learning approaches and compared with performance metrics. Logistic Regression (LR), Random Forest (RF) and Adaboost (AB) were used for machine learning approaches. Performance metrics used for comparison are precision, recall and F1 score. Confusion matrix values and Receiver Operating Characteristic (ROC) curves for the results are also presented. It was observed that the results of the RF and LR studies were close to each other and gave better results than the study done with the AB.