Abstract: The Student Counselling System through Artificial Intelligence (AI) is a cutting-edge platform designed to enhance student well-being, and career readiness. Leveraging advanced AI algorithms, the system provides personalized counselling, emotional support, and career guidance to students. A key feature of this system is its ability to generate comprehensive reports of counselling conversations, allowing both students and educators to track progress and identify areas of concern. The platform supports both voice and text input during counselling sessions, ensuring flexibility and accessibility for a diverse range of users. The role of Artificial Intelligence in human monitoring and recognition is taking advanced steps on every progress. This technology makes a greater impact on student’s life in helping parents and teachers understand and realize their panic situations. This project introduces a student counseling system integrating Convolutional Neural Networks (CNNs) for emotion recognition from facial expressions. Utilizing open-source Face Emotion Recognition (FER) dataset, the system classifies seven basic emotions: angry, disgusted, fearful, happy, neutral, sad, and surprised. The classifications guide personalized counseling sessions conducted through a chatbot interface integrated with the RASA framework. Interactions are securely stored in a database accessible only to teachers, offering insights into students' emotional states. The CNN model achieved an overall accuracy of 80%, with varying precision and F1-score metrics across emotion categories. The model achieved high accuracy rates of approximately 75% for happiness and 67% for surprise, while demonstrating moderate accuracy of around 53% for neutral and 48% for angry emotions. However, it showed lower accuracy rates of approximately 65% for disgust, 38% for fear, and 49% for sadness. Despite variations in accuracy across emotions, the system aims to enhance student counseling efficacy, promoting holistic well-being and academic success in educational settings.
Keywords: Chat2bot Interface; Convolutional Neural Networks; Emotion Recognition; Face Emotion Recognition (FER) ;RASA framework.
|
DOI:
10.17148/IJARCCE.2024.134226
