CD Skripsi
Implementasi Deep Learning Untuk Klasifikasi Jenis Sampah Organik Dan Anorganik Menggunakan Transfer Learning Vgg16
ABSTRACT
Waste refers to objects that are no longer used, have no utility, are undesirable, or are deliberately discarded. Waste is generated from human activities and does not occur naturally. At the national level, Indonesia faced a waste emergency in 2022, marked by a total waste volume reaching 36.19 million tons. Of this amount, approximately 35.99%, or around 13,025,812.81 tons, was unmanaged waste. One effort that can aid in waste management is developing a machine learning model capable of recognizing waste types. In this study, a pre-trained VGG16 model was utilized as the foundation for creating a model that can detect waste types. The resulting model achieved an accuracy of 94.67%, while the validation score was 91.3%. The trained model demonstrated optimal performance in detecting waste types.
Kata kunci : Waste, Pre-trained model, VGG16
Tidak tersedia versi lain