International Journal of Science, Engineering and TechnologyAuthors: Rohini Ashok Gamane, Vaibhav Dabhade
Abstract: The widespread use of social media has resulted in a surge of fake accounts, posing serious risks to individuals, organizations, and society at large. Identifying fake accounts effectively is essential to preserving the integrity and credibility of social media platforms. This study introduces a machine learning-based approach to detect fake social media accounts. We employed five machine learning algorithms—Support Vector Machines (SVM), K-Nearest Neighbors (KNN), Random Forest, Logistic Regression, and Artificial Neural Networks (ANN)—to classify accounts as fake or genuine. The dataset used in this study consisted of features extracted from social media profiles, such as user behavior, profile details, and network characteristics. Experimental results revealed that the ANN algorithm outperformed the others, achieving a high accuracy of 95.6% in detecting fake accounts. The proposed approach offers significant benefits for social media platforms by enabling more efficient detection and prevention of fake accounts. Furthermore, the findings of this study can guide the development of advanced fake account detection systems, contributing to a safer and more reliable online environment.
DOI: https://doi.org/10.5281/zenodo.17962377
