A Novel Method for Community Detection using Multi-Objective Particle Swarm Optimization Algorithm

In the last decade, social media has exploded in popularity, with some platforms now having tens of millions of users. Many sorts of data, such as audio, video, and text, may be found in these networks and must be investigated individually. Because of the wide variety of uses for these networks and the impact of past events and societal norms, their designers confront a number of obstacles. The existence of communities inside social networks is one of their most distinguishing characteristics. Tools for “community extraction” make it easier to investigate all components of a network and classify activities that are typical of certain groups. Although it may seem like a simple task, recognizing communities on social media is computationally difficult owing to the unknown number of groups and the varied internal density of communities. The features of a network and the goal of studying it might guide the selection or development of a particular approach for finding communities within that network. There is a distinct grouping of current community detection techniques based on their individual characteristics. Selecting the right algorithm requires knowledge of the proper categorization and features of each category. In this research, we offer an approach to social media community discovery that makes use of a multi-objective particle swarm optimization algorithm. Evaluation findings confirm the superiority of this approach.