CLASSIFYING FAKE PROFILE IN FACEBOOK ACCOUNT USING SUPPORT VECTOR MACHINE

Abstract

The ability to connect people around the world and share videos, photos and communications has made online social media (OSM) popular. Despite much of literature available in this field, there is still lack of study focusing on the creation of fake profile in the OSM. In fact, there has been relatively little effort aimed at solving fake profile features using classification algorithm. For these reasons, Support Vector Machine (SVM) classifier was employed to classify the novel features that had been created in the fake profile particularly among Facebook users’ account. The study begins with the data collection process, data pre-processing, evaluation, testing and lastly obtaining the result. The findings have revealed that the SVM classifier able to predict the fake profile with high Classifying Accuracy (CA) and Area Under Curve (AUC). Ultimately, this finding will provide a new endeavour for countermeasure and protection of OSM users.