The proliferation of fake accounts on social media platforms poses significant challenges, including the dissemination of misinformation, manipulation of public opinion and cybersecurity threats. The fraudulent accounts evolve to mimic legitimate users, traditional detection methods struggle to keep pace. Leveraging various behavioural, content-based and network-specific features, we explore a range of supervised and unsupervised learning algorithms to identify anomalies indicative of fake profiles. The present research compares the models such as Random Forest, Support Vector Machines (SVM), and deep learning architectures. This examined their effectiveness in identifying fake accounts with high accuracy and low false- positive rates. Feature engineering plays a critical role in the success of these models, with key metrics consisting of account age, post frequency, linguistic patterns and network connectivity. Extensive experiments on data sets from popular platforms, comprising Twitter and Facebook, demonstrate the proposed system's potential in real-world applications. The findings suggest that machine learning techniques can significantly improve the speed and accuracy of fake account detection, offering social media companies a robust tool to combleelaat online fraud and safeguard user interactions. Finally, the present study presents a machine learning-based approach to enhance the detection of fake accounts on social media.
Download
Download XML
