Comparative Study of Various Machine Learning Algorithms for American Sign Language Recognition

This research paper presents a comparative study on various machine learning algorithms for sign language detection. The objective of this study is to find the sign language identification method that is most accurate and effective for usage in real-time applications. The Residual Network (ResNet), Artificial Neural Network (ANN), Convolutional Neural Networks (CNNs), VGG16, and MobileNet are five well-known machine learning methods whose performance is compared. The dataset utilized in this study comprises camera-captured sign language motions made by diverse people. We assess each algorithm’s performance using a variety of parameters, including accuracy. According to our findings, VGG16 performs better than the other four algorithms, with an accuracy rate of 99%. Therefore, we conclude that VGG16 is the most suitable algorithm for sign language detection, which can be used in real-time applications for the deaf and hearing-impaired community.