Gated Mechanism for Attention Based Multimodal Sentiment Analysis
This work addresses sentiment analysis for social media and video content, but it is incremental as it builds on existing methods with modest performance gains.
The paper tackled multimodal sentiment analysis by addressing cross-modal interaction learning, long-term dependencies, and fusion of cues, finding that cross-modal interactions are beneficial, and achieved accuracies of 83.9% and 81.1% on benchmark datasets, with absolute improvements of 1.6% and 1.34% over state-of-the-art.
Multimodal sentiment analysis has recently gained popularity because of its relevance to social media posts, customer service calls and video blogs. In this paper, we address three aspects of multimodal sentiment analysis; 1. Cross modal interaction learning, i.e. how multiple modalities contribute to the sentiment, 2. Learning long-term dependencies in multimodal interactions and 3. Fusion of unimodal and cross modal cues. Out of these three, we find that learning cross modal interactions is beneficial for this problem. We perform experiments on two benchmark datasets, CMU Multimodal Opinion level Sentiment Intensity (CMU-MOSI) and CMU Multimodal Opinion Sentiment and Emotion Intensity (CMU-MOSEI) corpus. Our approach on both these tasks yields accuracies of 83.9% and 81.1% respectively, which is 1.6% and 1.34% absolute improvement over current state-of-the-art.