Machine Learning Algorithms for Spam Detection in Social Media Ad Campaigns

Abstract

The growth of social media platforms has revolutionized advertising campaigns, enabling brands to reach vast audiences with tailored content. However, with this expansion, the prevalence of spam accounts has surged, potentially undermining the effectiveness of social media advertisements. This paper explores the application of machine learning (ML) algorithms in detecting spam within social media ad campaigns. We analyze the effectiveness of several common machine learning techniques, including decision trees, support vector machines, and neural networks, for identifying spam accounts and fraudulent engagement. A comparative study of these models, using a range of performance metrics such as accuracy, precision, recall, and F1-score, is presented. Our results demonstrate that while no single model outperforms others across all metrics, ensemble approaches and deep learning models show promise for achieving the highest detection accuracy. The research aims to provide insights into optimizing spam detection techniques, thereby improving the reliability and return on investment (ROI) for social media ad campaigns.