Super Characters: A Conversion from Sentiment Classification to Image Classification
This addresses sentiment classification for social media users by offering a novel approach that eliminates explicit word embeddings, though it is incremental in applying image classification techniques to text.
The authors tackled sentiment classification by converting text into images and using CNNs for classification, achieving consistent outperformance over other methods on ten large social media datasets across four languages.
We propose a method named Super Characters for sentiment classification. This method converts the sentiment classification problem into image classification problem by projecting texts into images and then applying CNN models for classification. Text features are extracted automatically from the generated Super Characters images, hence there is no need of any explicit step of embedding the words or characters into numerical vector representations. Experimental results on large social media corpus show that the Super Characters method consistently outperforms other methods for sentiment classification and topic classification tasks on ten large social media datasets of millions of contents in four different languages, including Chinese, Japanese, Korean and English.