Enthusiasm and support

Abstract

We present a novel sentiment classifier particularly designed for modeling and analyzing social movements; capturing levels of support (supportive versus non-supportive) and degrees of enthusiasm (enthusiastic versus passive). The resulting computational solution can help organizations involved with social causes to disseminate messages in a more informed and effective fashion; potentially leading to greater impact. Our findings suggest that enthusiastic and supportive tweets are more prevalent in tweets about social causes than other types of tweets on Twitter.

Information extraction from digital social trace data with applications to social media and scholarly communication data

“…Consider the tweet "Roger Federer killed this, Nadals sucks." [Mishra et al 2014]. In the presented tweet, suppose the author was a fan of Federer, and Federer won the match with Nadal.…”

“…Considering opinion mining or sentiment analysis as an IE task, earlier research has shown that sentiment in social media is more nuanced, and harder to predict compared to newswire or review corpora [Mishra et al 2014;2015, Maynard et al 2012, Aue and Gamon 2005.…”

“…The preceding works have motivated us to devise a simple yet consistent method of quantifying novelty of articles. We build upon some preliminary work presented in [Mishra et al 2014, Mishra andTorvik 2016] and present the details of our methods and the corresponding results in the sections which follow. was started in 1965 and a lot of noise -with regards to many MeSH terms being wrongly spelled or older articles being index by too few MeSH terms -is present in our corpus around those years.…”

“…In this work, we study the extension of the two approaches using the framework proposed in [Mishra et al 2014] for tagging tweets across the two dimensions of enthusiasm and support. We start by investigating the quality of the annotated data introduced in [Mishra et al 2014], followed by assessing its quality and robustness for the task of annotating tweets for enthusiasm and support.…”

“…In this work, we study the extension of the two approaches using the framework proposed in [Mishra et al 2014] for tagging tweets across the two dimensions of enthusiasm and support. We start by investigating the quality of the annotated data introduced in [Mishra et al 2014], followed by assessing its quality and robustness for the task of annotating tweets for enthusiasm and support. Next, an algorithm based on a weighted version of personalized page rank [Page et al 1998, Xing and Ghorbani 2004 is introduced for combining the annotated tweets with user and hashtag mention networks.…”