Deep Learning Algorithms for Skin Condition Classification

Abstract

Skincare is an essential aspect of personal care, but selecting suitable products remains challenging due to individual variations in skin type and condition. Existing skincare recommendation systems rely on questionnaires, which may lead to inaccurate recommendations. This study explores the application of machine learning algorithms, particularly Convolutional Neural Networks (CNNs), for automated skin analysis and personalised skincare recommendations. By analysing images of users’ skin, the system can classify skin types, detect conditions such as acne or dryness, and suggest suitable products. The study evaluates different deep learning models, including VGG-16, ResNet-50, and MobileNetV2, comparing their accuracy and efficiency. Experimental results indicate that the proposed model achieves high accuracy in classifying skin conditions, demonstrating the potential of machine learning in revolutionising personalised skincare solutions.