IJLLL 2015 Vol.1(3): 164-169 ISSN: 2382-6282
DOI: 10.18178/IJLLL.2015.1.3.32

Tracking Emotions in Hot Topics: Exploiting Event-Specific Emotional Words

Yun Niu and Shihong Wang
Abstract—A word often conveys different emotions depending on its context. For example, in social media, emotional words in comments made on hot spot social events often denote emotions specific to the topic, which may be different from that when examined out of context. Nonetheless, contextual emotions of words has not been given much attention in current work of emotion analysis. This paper proposes an approach to identify event-specific emotional words and predict their emotion labels. These words are then used to track emotions in microblog posts using both rule-based and supervised approaches. Experimental results show that exploiting the detected words substantially improved the performance of emotion analysis.

Index Terms—Contextual emotion analysis, co-occurrence pattern, emotion lexicon, social media.

The authors are with Nanjing University of Aeronautics and Astronautics, Nanjing 210016, China (e-mail: yniu@nuaa.edu.cn, wsh_014@nuaa.edu.cn).

[PDF]

Cite:Yun Niu and Shihong Wang, "Tracking Emotions in Hot Topics: Exploiting Event-Specific Emotional Words," International Journal of Languages, Literature and Linguistics vol. 1, no. 3, pp. 164-169, 2015.

Copyright©2008-2017. International Journal of Languages, Literature and Linguistics. All rights reserved.
E-mail: ijlll@ejournal.net