Techniques for Influencer Detection, Mapping, and Popularity Prediction in Social Media Networks

This paper recognizes the fact that in an age where consumption patterns are closely linked with social media platforms the measurement of key influential personalities and their effects on consumers is crucial. This paper offers an extensive analysis of the state-of-the-art methods for influencer identification and influence mapping. Approaches used in this research are based on Machine Learning and Deep Learning Algorithms which include Graph Attention Networks, Gated Recurrent Units, Graph-based Long Short-Term Memory Neural Networks, Recurrent Neural Networks, Stacked Autoencoders, Centrality Measures, and Frameworks such as Stimulus Organism Response. Comparing all of these approaches reveals their role in improving influencer marketing approaches, targeting, and the level of engagement, and profitability. The current study will benefit this area of knowledge by showing the outcomes, performance, and computational strengths of these approaches, paving the way for the enhancement of influencer marketing technology.