<![CDATA[This study aims to develop a skin disease classification model based on Convolutional Neural Networks (CNN) specifically designed to classify three types of skin diseases: acne, ringworm, and tinea versicolor. Unlike some previous studies that utilized datasets from various domains such as textile images, plants, and blood cell images, this research specifically employs a dataset relevant to skin diseases. The dataset used in this study consists of 810 skin disease images, divided into 600 images for training (200 images each for acne, tinea versicolor, and ringworm) and 210 images for testing. To enhance data variation and support model generalization, the dataset was processed using augmentation techniques. The model's performance evaluation showed promising results, with an average accuracy of 87.14%. Additionally, the model achieved precision, recall, and F1-score values of 87% each, demonstrating its ability to detect and classify skin diseases consistently. This study is expected to serve as a foundation for developing more accurate and efficient technology-based diagnostic tools, particularly for skin diseases, in the future]]>
Copyrights © 2025
