PREDICTING HOT EVENTS IN THE EARLY PERIOD THROUGH BAYESIAN MODEL FOR SOCIAL NETWORKS

Abstract

The rapid expansion of social networking platforms has transformed the way information is generated, shared, and consumed across the world. Millions of users continuously interact on digital platforms such as Twitter, Facebook, Instagram, Reddit, YouTube, and online discussion forums, producing an enormous volume of data every second. Among the different types of information exchanged on these platforms, certain events attract unusually high attention within a short period of time. These events, commonly referred to as “hot events,” may include political incidents, sports updates, natural disasters, entertainment news, product launches, social movements, public emergencies, or viral trends. Identifying such events at an early stage has become an important research area because early prediction enables organizations, governments, businesses, and researchers to take timely actions and make informed decisions