Leveraging Social Media Data Fusion for Enhanced Student Evolution in Media Studies using Machine Learning

Najla M. Alnaqbi ¹ , Walaa Fouda ², Muhammad Eid Balbaa^3,*

¹Mohamed bin Zayed University for Humanities, UAE

²American University in the Emirates, UAE

³ Tashkent State University of Economics, Uzbekistan

Emails: Najla.alnaqbi@mbzuh.ac.ae; walaa.fouda@aue.ae; m.balbaa@tsue.uz

Abstract

 In the realm of media studies, understanding student evolution is a crucial aspect for educators and researchers. However, traditional research methods often struggle to capture the dynamic nature of media consumption and the intricate interactions between individuals and media content. To address this challenge, this paper focuses on leveraging social media data fusion and machine learning techniques to enhance the comprehension of student evolution. By integrating data from diverse social media sources and employing the CATBoost algorithm with the Greedy Target-based Statistics (Greedy TBS) technique, we aim to predict student outcomes based on a comprehensive set of attributes. The results showcase the superior performance of CATBoost in accurately capturing the complexities of student evolution, surpassing other machine learning algorithms. The findings hold immense significance for educators, empowering them with valuable insights into students' behaviors, preferences, and performance.

Keywords: social media data fusion; machine learning; CATBoost algorithm; student evolution; media studies.