Hybrid Approach for Real-Time Emotion Recognition on Twitter Data
N. Bala Krishna, Vidhyavathi Kotha, Obula Reddy Chandana, Yasodhara Varma, Yedavali Venkat Sai, Torati Nobel Siva Sai Ganesh · 2025
Emotion recognition from textual data is a critical aspect of natural language processing, particularly in understanding user sentiment and behavior on social media platforms like Twitter. In order to successfully classify emotions in Twitter data, this study suggests a hybrid deep learning model that integrates Convolutional Neural Networks (CNNs) with Gated Recurrent Units (GRUs). Whereas the GRU component records contextual information and sequential dependencies, the CNN component extracts local features and n-gram patterns from the text. The model's capacity to recognize complex emotional cues is improved by using pre-trained embeddings for textual representation. In comparison to conventional machine learning techniques and stand-alone deep learning, the proposed method outperforms in metrics such as accuracy, precision, recall, and F1-score. when tested on a labeled Twitter dataset with several emotion classes.