Sentiment Classification using Images and Label Embeddings
This work addresses sentiment analysis for applications combining text and images, but it appears incremental as it focuses on comparing existing data types without introducing new methods.
The study investigated the semantic information in images and their added value for sentiment analysis of associated text, comparing models using only images, only text, or both, and assessing generalization to unknown sentiments.
In this project we analysed how much semantic information images carry, and how much value image data can add to sentiment analysis of the text associated with the images. To better understand the contribution from images, we compared models which only made use of image data, models which only made use of text data, and models which combined both data types. We also analysed if this approach could help sentiment classifiers generalize to unknown sentiments.