Enhancing Social Media Influence with Cutting-Edge Machine Learning Approaches

Abstract

Homophily and community influence play the important roles in shaping user behavior towards others and towards content in social recommendation systems. This aspect has a considerable effect on how individuals engage with materials. Homophily is a core concept in social network analysis: it is the tendency of individuals to associate with others who have similar interests or characteristics. This research begins with datasets collected from Facebook, Instagram, and YouTube. Preprocessing occurred after the dataset was obtained from Kaggle sources. Following the completion of the initial dataset processing, the data is categorised using a fusion machine learning method. With accuracy rates of 97.11% for Facebook, 98.02% for Instagram, and 98.99% for YouTube, the suggested solution consistently outperforms existing approaches.