Classification of Organic and Recyclable Waste based on Feature Extraction and Machine Learning Algorithms
Yazarlar (2)
Dr. Öğr. Üyesi Elham Tahsın Yasın YASIN Sinop Üniversitesi, Türkiye
Doç. Dr. Murat Köklü Selçuk Üniversitesi, Türkiye
| Bildiri Türü | Tebliğ/Bildiri | Bildiri Dili | İngilizce |
| Bildiri Alt Türü | Tam Metin Olarak Yayınlanan Tebliğ (Uluslararası Kongre/Sempozyum) | ||
| Bildiri Niteliği | Alanında Hakemli Uluslararası Kongre/Sempozyum | ||
| Kongre Adı | International Conference on Intelligent Systems and New Applications (ICISNA’23) | ||
| Kongre Tarihi | 28-04-2023 / 30-04-2023 | ||
| Basıldığı Ülke | İngiltere | Basıldığı Şehir | Liverpool |
| UAK Araştırma Alanları |
|
||
| Özet |
| Managing solid waste effectively requires the proper classification of waste. To determine whether a waste is organic or recyclable, machine learning methods can be used. This study extracted features from waste samples using the InceptionV3 feature extraction method, and three machine learning classifiers were used to compare accuracy. The study utilized the InceptionV3 deep convolutional neural network, which was pretrained on large-scale image datasets and fine-tuned on waste images to extract features. The extracted features were used to train three machine-learning classifiers. The performance of the classifiers was evaluated using a labeled waste image dataset. As a result of our experiments, we found that SVMs achieved an accuracy of 96.3% without any feature selection, Decision Trees achieved a result of 85.8%, and KNNs achieved a result of 94.9%. Based on our study, we demonstrate that it is feasible to classify solid waste using machine learning algorithms. A waste classification and management system that achieves optimum efficiency can be implemented with the help of the findings of this study. |
| Anahtar Kelimeler |
| Solid waste classification | Feature extraction | Waste management | Machine learning algorithms | Deep learning-based approach |
