Classification of Skin Allergies Using Convolutional Neural Network and Support Vector Machine
Hazel F. Anuncio, Arjo R. Ladia, Nicko James T. Pama · 2024
This study presents the development of a mobile application tailored for the classification of skin allergies, leveraging Convolutional Neural Network (CNN) and Support Vector Machine (SVM) techniques. The application utilizes datasets sourced from Kaggle, with a rigorous partitioning of 70% for training, 20% for testing, and 10% for validation. Through the integration of CNN and SVM algorithms, where CNN is responsible for feature extraction and SVM for classification, the application demonstrates robust performance in accurately identifying various skin allergy conditions. By harnessing the power of deep learning and traditional machine learning approaches, coupled with an optimized dataset split, the developed application showcases promising results in the realm of dermatological diagnosis and healthcare accessibility.