Enhancing Machine Learning Innovative Model for Waste Management: A Focus on Data Preprocessing and Labeling

Efficient waste management is essential for sustainability and urban planning, especially with increasing waste complexity. While traditional methods often rely on manual processes, machine learning offers a promising approach to automate waste classification, improving accuracy and reducing costs. Building on our previous research, which utilized real-time data collection through Internet of Things (IoT) devices, this study focuses on enhancing machine learning models by improving the quality and diversity of the underlying dataset. We developed a large pre-processed dataset, initially analyzed using unsupervised clustering with the K-Means algorithm, and labeled various waste types. The dataset integrates primary data from IoT devices, secondary data from public repositories, and tertiary data from related studies, creating a comprehensive resource for model training. Key preprocessing techniques, including data cleaning, annotation, and normalization, were applied to improve data quality. Experiments with different machine learning models, such as Random Forest and Support Vector Machines, demonstrate that diverse and well-preprocessed data significantly enhances model performance, leading to better classification accuracy. This study contributes to advancing waste management systems by providing a robust dataset and insights into data preprocessing, offering a foundation for further research and practical applications in smart cities.