Web Scraping Social Media Using Support Vector Machine (SVM)

Social Media Applications, especially Facebook, have revolutionized marketing. This paper presents a method using Facebook Scraper and Support Vector Machine (SVM) to classify pages based on likes and followers. The study, focused on Iraq, reveals preferences for clothing pages. Results demonstrate the effectiveness of SVM in understanding user interests. The Support Vector Machine was used to classify these pages. The aim of this work is to find out the pages or items preferred by people based on the number of likes and followers on these pages. Moreover, the popular interests were also classified for further analysis. In this research, Iraq was used as the case study. This work achieved an 85% accuracy in classifying user preferences on Facebook pages using SVM by splitting our dataset into two groups test and training sets, the results were presented as charts and found the pages that the Iraqi community preferred and had the most likes are the cloth and fashion pages.