Advanced Detection of Fake Social Media Accounts Using Machine Learning Algorithms

Abstract

The exponential growth of social media platforms has resulted in a surge of fake accounts, posing threats such as misinformation,
financial scams, and privacy breaches. This study proposes an efficient detection system for fake social media accounts using supervised machine learning algorithms. The dataset consists of various account features including profile characteristics, activity patterns, and interaction behavior. Multiple models, including Logistic Regression, Decision Trees, Random Forest, and Support Vector Machines (SVM), were evaluated to identify fake accounts accurately. Among the algorithms tested, Random Forest demonstrated the highest accuracy with an F1-Score of 0.89 and an AUC-ROC score of 0.90, surpassing other models. The proposed system effectively detects fake accounts by analyzing behavioral patterns and extracting significant account-level features. Additionally, the use of feature selection
techniques enhanced model performance and reduced computational complexity.To further validate the robustness of the proposed approach, cross-validation techniques were applied, ensuring reliable and unbiased results. Comparative analysis with existing detection methods demonstrated superior performance, highlighting the effectiveness of the implemented models. Moreover, the study explores the interpretability of the machine learning models, providing insights into key factors that distinguish fake accounts from genuine ones.This approach offers a scalable and reliable solution for social media platforms to mitigate the proliferation of fake accounts, ensuring a safer online environment. Future research can explore the integration of real-time detection systems and the application of deep learning for further improvements. The results underscore the importance of employing advanced machine learning techniques in enhancing cybersecurity in social media ecosystems. Additionally, collaborative efforts between social media companies, regulatory authorities, and researchers can further strengthen detection mechanisms, contributing to the reduction of malicious online activities.