<![CDATA[This study implements a direct facial expression detection system via the web using teachable machine and tensorflow.js. This system utilizes machine learning technology that operates directly in the browser without the need for a special server. With the transfer learning method, the model is trained to recognize various facial expressions such as happy, sad, angry, and neutral. This implementation uses a convolutional neural network (cnn) architecture that has been optimized for web activities. The results of the test show a detection accuracy level of 85-90% with a response time of under 200ms. This solution provides a lightweight option for emotion recognition applications that can be easily accessed via a web browser. The main advantages of this system are ease of implementation, cross-platform support, and maintaining data privacy because the process is carried out locally.]]>
Copyrights © 2025
