Skripsi
KLASIFIKASI KANKER KULIT MENGGUNAKAN VISION TRANSFORMER (VIT)
This research presents the implementation and evaluation of a LightWeight Vision Transformer (ViT) architecture for classifying seven different types of skin cancer using medical dermoscopic images from the HAM10000 dataset. The methodology includes data preprocessing, class balancing using Random Oversampling, and splitting the dataset into training and testing sets. Several hyperparameter configurations were applied during model training, and the performance was assessed using accuracy, precision, recall, and F1-score to determine the optimal model. The best performance was achieved by Model 19 with a batch size of 64, learning rate of 1×10⁻³, and 100 epochs, resulting in an accuracy of 0.9909, precision of 0.9910, recall of 0.9911, and F1-score of 0.9909. These findings indicate that the LightWeight ViT architecture is capable of delivering highly accurate and stable performance in skin cancer image classification across multiple training configurations, making it suitable for implementation as an automated decision-support system in medical diagnosis.
| Title | Edition | Language |
|---|---|---|
| KLASIFIKASI CITRA DERMOSKOPI KANKER KULIT DENGAN MENGGUNAKAN CONVOLUTIONAL NEURAL NETWORK (CNN) | id |