Skip to main content
Figure 3 | EPJ Data Science

Figure 3

From: Sentiment analysis methods for understanding large-scale texts: a case for using continuum-scored words and word shift graphs

Figure 3

New York Times (NYT) ‘Society’ section shifted against the entire NYT corpus for each of the six dictionaries listed in tiles A-F. We provide a detailed analysis in Section 3.1. Generally, we are able to glean the greatest understanding of the sentiment texture associated with this NYT section using the labMT dictionary. We note that the labMT dictionary has the most coverage quantified by word match count (Figure in S3 Appendix), that we are able to identify and correct problematic words scores in the OL dictionary, and that we see that the MPQA dictionary disagrees entirely with the others because of an overly broad stem match.

Back to article page