AI-Based E-Waste Classification Using Deep Learning

Abstract Electronic waste (e-waste) management remains a critical environmental challenge due to increasing volumes and inefficient manual sorting practices. This study develops and evaluates an AI-based e-waste classification system using deep learning with transfer learning on the MobileNetV2 architecture. The proposed model classifies images into 10 common e-waste categories and achieves an overall accuracy of 95%, with strong per-class performance (F1-scores ranging from 0.89 to 0.98). The system demonstrates high potential for practical deployment in recycling facilities, supporting automated sorting, resource recovery, and sustainable waste management. Keywords — E-waste classification, deep learning, MobileNetV2, transfer learning, image classification, electronic waste, automated sorting, lightweight neural network