Machine Learning (ML) for Waste Classification: A Comparative Study of Different Algorithms

Abstract

Effective waste classification is crucial for efficient waste management and sustainable environmental practices. Machine learning (ML) algorithms have been increasingly used to classify waste due to their ability to learn from large datasets and improve classification accuracy. This study compares the performance of various ML algorithms in waste classification, including Support Vector Machines (SVM), Random Forest (RF), Decision Trees (DT), K-Nearest Neighbors (KNN), and Convolutional Neural Networks (CNN). The algorithms were trained and tested on a dataset of 1,500 waste samples, each with 12 features representing physical and chemical properties. The results show that the RF algorithm achieved the highest accuracy (93.3%) and F1-score (0.93), followed closely by the SVM algorithm (92.7% accuracy and 0.92 F1-score). The DT and KNN algorithms performed moderately well, while the CNN algorithm struggled to classify waste effectively. The study highlights the potential of ML algorithms in improving waste classification accuracy and suggests that RF and SVM algorithms are promising choices for waste management applications.