A novel simplicial complex representation of social media texts: The case of Twitter

Abstract

With the rise in technological advances in the last decades, there has been an increase in the complexity of the many systems that we are members of. The emergence and development of web-based platforms have particularly revolutionized the way we express our opinions and feelings, communicate with one another, and in general, the manner we gain and share information. This constant rise in the scale and variety of information and opinions and the overall complexity of online social networks have motivated the search for alternative methods that can capture the complex and higher-order characteristics of the interactions within these systems.For this purpose, this paper has attempted to showcase and introduce a novel method for studying the various dimensions of the characteristics and interactions present in online social networks. Using the concept of simplicial complexes, a novel framework is defined and presented to analyze the text contexts of opinions shared by users of these complex systems. The spread and propagation of information is then researched in the representations using epidemic models and then applied to real-life datasets obtained through Twitter.