Dive deeper: Deep Semantics for Sentiment Analysis
Nikhilkumar Jadhav, Pushpak Bhattacharyya · 2014
This paper illustrates the use of deep se-mantic processing for sentiment analysis. Existing methods for sentiment analysis use supervised approaches which take into account all the subjective words and or phrases. Due to this, the fact that not all of these words and phrases actually con-tribute to the overall sentiment of the text is ignored. We propose an unsupervised rule-based approach using deep semantic processing to identify only relevant sub-jective terms. We generate a UNL (Uni-versal Networking Language) graph for the input text. Rules are applied on the graph to extract relevant terms. The sen-timent expressed in these terms is used to figure out the overall sentiment of the text. Results on binary sentiment classification have shown promising results. 1