Abstract

Implementing machine learning (ML) techniques in E-commerce has increased as a decisive strategy for mitigating the growing surge of cybercrime. Online retailers can significantly improve their cybersecurity procedures by deploying ML potential to investigate large datasets and detect anomalies from fraudulent transactions. Despite its competency, ML adoption witnesses several obstacles, including technical difficulties, the requirement for significant investment in infrastructure, and regulatory compliance. By investigating these obstacles, organizations can build strategies to tackle them, allowing ML to safeguard online retail settings against cybercrimes. To better understand this phenomenon, we conducted semi-structured interviews with cybersecurity professionals in online retail. The findings reveal significant challenges, such as more quality data, updated systems, and an easy target for adversarial attacks. In addition, financial constraints, skills shortages, privacy and trust matters, and different organizational and technical barriers add more complications to ML adoption for online retailers.

Download

Abstract Only

Share

COinS