Fusion-Based AI for Sentiment and Emotion Understanding in Social Media
Abstract
Analyzing user sentiment and emotions in digital conversations is essential for understanding online behavior. This study introducesa novel AI-driven framework that integrates multimodal deep learning techniques to enhance sentiment, emotion, and desireclassification from social media content. By fusing text and image-based features using transformer-based architectures, our approachoutperforms traditional unimodal models in accuracy and robustness. Extensive evaluations on diverse datasets demonstratethe effectiveness of our fusion strategy, paving the way for improved sentiment analytics in social media research and real-timeemotion tracking.
Fields of Study & Subjects
Sentiment Analysis and Opinion Mining
Emotion and Mood Recognition
Hate Speech and Cyberbullying Detection
Publication details
- Journal / source
- International Journal of Progressive Sciences and Technologies
- Volume
- 55
- Issue
- 2
- Pages
- 446–446
- Publication date
- 29 January 2026
- Language
- EN