#Emotional Tweets

First Joint Conference on Lexical and Computational Semantics (*SEM), pages 246–255, Montr´eal, Canada, June 7-8, 2012. c⃝2012 Association for Computational Linguistics #Emotional Tweets Saif M. Mohammad Emerging Technologies National Research Council Canada Ottawa, Ontario, Canada K1A 0R6 saif.mohammad@nrc-cnrc.gc.ca Abstract Detecting emotions in microblogs and so- cial media posts has applications for industry, health, and security. However, there exists no microblog corpus with instances labeled for emotions for developing supervised systems. In this paper, we describe how we created such a corpus from Twitter posts using emotion- word hashtags. We conduct experiments to show that the self-labeled hashtag annotations are consistent and match with the annotations of trained judges. We also show how the Twit- ter emotion corpus can be used to improve emotion classification accuracy in a different domain. Finally, we extract a word–emotion association lexicon from this Twitter corpus, and show that it leads to significantly better results than the manually crafted WordNet Af- fect lexicon in an emotion classification task.1 1 Introduction We use language not just to convey facts, but also our emotions. Automatically identifying emotions expressed in text has a number of applications, in- cluding customer relation management (Bougie et al., 2003), determining popularity of products and governments (Mohammad and Yang, 2011), and improving human-computer interaction (Vel´asquez, 1997; Ravaja et al., 2006). Twitter is an online social networking and mi- croblogging service where users post and read mes- sages that are up to 140 characters long. The mes- sages are called tweets. 1Email the author to obtain a copy of the hash-tagged tweets or the emotion lexicon: saif.mohammad@nrc-cnrc.gc.ca. Often a tweet may include one or more words im- mediately preceded with a hash symbol (#). These words are called hashtags. Hashtags serve many purposes, but most notably they are used to indicate the topic. Often these words add to the information in the tweet: for example, hashtags indicating the tone of the message or their internal emotions. From the perspective of one consuming tweets, hashtags play a role in search: Twitter allows peo- ple to search tweets not only through words in the tweets, but also through hashtagged words. Con- sider the tweet below: We are fighting for the 99% that have been left behind. #OWS #anger A number of people tweeting about the Occupy Wall Street movement added the hashtag #OWS to their tweets. This allowed people searching for tweets about the movement to access them simply by searching for the #OWS hashtag. In this partic- ular instance, the tweeter (one who tweets) has also added an emotion-word hashtag #anger, possibly to convey that he or she is angry. Currently there are more than 200 million Twitter accounts, 180 thousand tweets posted every day, and 18 thousand Twitter search queries every second. Socio-linguistic researchers point out that Twitter is primarily a means for people to converse with other individuals, groups, and the world in general (Boyd et al., 2010). As tweets are freely accessible to all, the conversations can take on non-traditional forms such as …

Content truncated to 500 words