Data-Driven E-Waste Management: Leveraging Machine Learning Algorithms for Efficient Collection and Recycling Systems

Quick growth of e-waste poses a substantial risk environment and health concerns; thereby, there is an imminent need for efficient management solutions. This paper explains the utilization of ML-based algorithms in improving e-waste collection and recycling capabilities. It employs superior techniques like CNN and YOLOv8 along with image recognition to identify e-waste in real time. The CNN-VGG model uses deep convolutional layers to extract features in the E-Waste Dataset to classify e-wastes. YOLOv8 assists in the location and detection of e-waste objects. The new approach is developed in Python and demonstrates an accuracy of nearly 99%, better than other models such as FrHHGO-based ShCNN and SNN by a margin of 10.44%. This shows how CNN-VGG is good for the classification of e-waste. Adding these ML models into automation systems in recycling helps recycle more material and sustain the environment in e-waste management. The work published here advances more effective, ecologically sound e-waste recycling practice.