Emotions play a vital role in human experiences and interactions, influencing decision-making, behavior, and preferences. This research presents a real-time emotion detection system that integrates deep learning and computer vision to recognize facial expressions and recommend personalized movies accordingly. The proposed system utilizes OpenCV for face detection and a Convolutional Neural Network (CNN)-based deep learning model to classify emotions into seven categories: Angry, Disgust, Fear, Happy, Neutral, Sad, and Surprise. The model processes live video frames, detecting facial expressions and maintaining a frequency count of detected emotions. Upon identifying the most frequently occurring emotion, the system retrieves personalized movie recommendations by scraping IMDb for films corresponding to the detected mood. By mapping different emotions to relevant movie genres (e.g., Drama for sadness, Comedy for happiness, Thriller for fear), the system enhances user engagement and provides a tailored entertainment experience. The implementation demonstrates the potential of artificial intelligence (AI) in affective computing, bridging the gap between emotional recognition and real-world applications. The study also highlights the significance of real-time processing, web scraping for dynamic recommendations, and deep learning for accurate emotion classification. Future improvements may include integrating natural language processing (NLP) for multimodal emotion recognition, expanding the recommendation dataset, and refining model accuracy using larger training datasets. This research contributes to the fields of human-computer interaction (HCI), affective computing, and AI-driven recommendation systems, offering an innovative approach to emotion-based content personalization.
Volume 17 | Issue 3
Pages: 1-13