Intelligent Detection of Fake Social Media Accounts Using Bagged SVM Classifier

The surge of fraudulent accounts on social media platforms presents a considerable risk to user privacy, online security, and the integrity of digital communication. This research introduces an advanced detection approach for recognizing fraudulent social media accounts via a Bagged Support Vector Machine (SVM) classifier. The suggested method improves the resilience and accuracy of conventional SVM classifiers by utilizing the ensemble learning technique known as bootstrap aggregation (bagging). A thorough dataset containing authentic and fraudulent profiles is employed, integrating attributes such as account activity, content patterns, network interactions, and behavioral characteristics. The Bagged SVM model attains an accuracy of <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{9 5. 4 \%}$</tex>, a precision of <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{9 6. 3 \%}$</tex>, a recall of <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{9 4. 7 \%}$</tex>, and an F1-score of <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{9 5. 0 \%}$</tex>. Experimental results indicate enhanced performance relative to individual classifiers such as Random Forest (92.1 %), AdaBoost (89.3%), and Gradient Boosting (87.5%). The results confirm the efficacy of the proposed strategy in diminishing false positives and enhancing detection reliability. This sophisticated algorithm provides a scalable and effective solution for social media companies to address the increasing issue of false account proliferation.