Volume 16, Number 3
Cyberguard: Fake Profile Detection using Machine Learning
Authors
M.Vijaya Lakshmi, Aishwarya Rao, Kalyani Boddulah, Ojaswi Cheekati and Varsha Vadla, G. Narayanamma Institute of Technology and Science (For Women), India
Abstract
Social media platforms that foster high levels of user engagement, such as Facebook, Instagram, and Twitter, have huge impact on people's lives everywhere. They still have issues with false profiles, which may be created by automated systems, computer programs, or individuals. These false accounts facilitate illicit activities like phishing and identity theft as well as the dissemination of rumors. To solve this issue, our study employs many machine learning techniques to differentiate between authentic and fraudulent Twitter profiles. Analysis is done using key data including the quantity of friends and followers, activity trends, and more. The algorithms such as Random Forest, XG Boost, LSTM, and neural networks emphasizes how crucial itis to choose important criteria while evaluating social media accounts. After training, the models generatea value of 1 for bogus profiles and 0 for real ones, allowing for the identification and the removal of bogus profiles to lessen cybersecurity risks.
Keywords
Random forest, LSTM, XG Boost, Twitter, social media, fake profiles, machine learning, and neural networks.