Spurious Social Network Profiles Identification Through Hybrid ML Techniques: Analysis and Observation

Due to the excessive growth of social networks, an unprecedented surge in the creation of fake accounts is seen. Such a provision poses security vulnerabilities and privacy threats. These accounts are often orchestrated to facilitate fraudulent schemes, propagate misinformation, and execute sophisticated malicious campaigns. This paper undertakes a comprehensive survey of the recent machine learning (ML) techniques and tools used for the detection of fake profiles on prominent platforms, notably "Instagram", "Twitter", and "Facebook". The emphasis is placed on advanced methodologies, including ensemble learning, deep neural networks, and hybridized ML approaches, with an in-depth evaluation of their algorithmic efficiency, computational trade-off, and domain-specific adaptability. The insights derived underscore the dominant trends, systemic limitations, and potential breakthroughs necessary to fortify the efficacy and scalability of fake account detection frameworks. This study helps cybersecurity practitioners and network researchers to detect fraudulent user accounts floating in social media and mitigate social engineering attacks.