Pattern Recognition in Waste Management
A. Bhanu Prasad, Venkataramana Asha, M. Govindaraj, Angel Christina, M S Arpitha, C T Aswathi · 2025
The proper waste sorting is important for efficient waste management and the protection of the environment. This paper presents an improved ml typical for waste item organization by means of a heterogeneous dataset. Here this research compares various classification models, such as XGBoost, Random Forest, SVM, CNN, KNN and also it is used to classify the pattern. To improve model performance, the dataset is subjected to extensive preprocessing, including feature extraction and augmentation. Important evaluation metrics like accuracy, and other types of evaluation metrics and confusion matrix analysis stand employed toward measure effectiveness of separately perfect. These outcomes display that the ensemble-based algorithms such as XGBoost and deep-learning algorithms like as CNN perform better than conventional classifiers by identifying complex patterns. This study adds to smart waste management by using a data-driven method for accurate waste sorting. Through enhanced sorting accuracy, it enhances effective recycling and minimizes landfill waste, thereby advancing environmental sustainability.