<![CDATA[Guidance and Counseling is a service provided to students to help them develop their potential optimally. Detecting students' facial expressions in the counseling room plays a crucial role in assisting counselors in understanding the emotional state of students who may need help, such as depression, anxiety, or stress, as they often find it difficult to express their feelings verbally. Therefore, this research will focus on 7 types of facial expressions: Anger, Disgust, Fear, Happiness, Neutral, Sadness, and Surprise. To classify these facial expressions, a Convolutional Neural Network (CNN) technique will be used, which identifies objects based on color and contours in an image. The aim of this research is to create a CNN model that can detect students' facial expressions during counseling sessions. In this study, the machine learning life cycle method is also employed as a stage in building the CNN model, starting with data collection with a total of 618 images, data cleaning, labeling the data, splitting the data into training and testing data with an 80% training data and 20% testing data ratio, creating the CNN architecture, training and evaluating the created model, and finally implementing it using a webcam. The results of this research show that the model achieved an accuracy of 33%. However, the facial expression detection features using the CNN model successfully detected students' facial expressions despite having a low prediction accuracy rate. Keywords— Convolutional Neural Network, Facial Expression Detection, Guidance Counseling, Machine Learning, Webcam]]>
