E-Commerce Security through Social Media Integration and Machine Learning

E-commerce combined with social media creates global connectivity and user experience but also creates security challenges in the form of data breaches, scams, and phishing. This paper proposes a secure e-commerce scheme based on logistic regression (SESbLR) to improve cyber system security in e-commerce platforms that incorporate social media. The objective of the paper is to develop a secure e-commerce SESbLR framework that integrates RSA encryption and logistic regression to detect and prevent fraudulent transactions using social media data and logistic Regression for fraud detection. Experimental evaluation demonstrates that the SESbLR model achieves high performance with 99.65% accuracy, 99.45% precision, 99.20% recall, and 99.00% f1-score. This presents a secure and defect-free e-commerce environment.