Scam Guardian

Abstract

Despite these successes, challenges persist due to evolving tactics employed by malicious actors. Nevertheless, ongoing researchendeavorsaim to develop ML algorithmscapable of adapting to emerging techniques. The project provides a timeline of key milestones, from the seminal 2009 study to recent developments, including the use of Graph Convolutional Networks (GCNs), the DeepFake Detection Challenge by Facebook in 2020, and Microsoft's announcement of detecting and removing over 60% of fake accounts on Twitter in 2021. Looking ahead, the project envisions a promising future for ML-based take account detection. Advancements in ML techniques are anticipated to enhance detection accuracy and extend towards proactive measures to prevent the creation of fake accounts. The report concludes by highlighting the dynamic nature of this field and the continuous evolution of ML algorithms to effectively combat the persistent challenge of fake account proliferatiorton social media platform.