An Event-comment Social Media Corpus for Implicit Emotion Analysis

Proceedings of the 12th Conference on Language Resources and Evaluation (LREC 2020), pages 1633–1642 Marseille, 11–16 May 2020 c⃝ European Language Resources Association (ELRA), licensed under CC-BY-NC 1633 An Event-comment Social Media Corpus for Implicit Emotion Analysis Sophia Yat Mei LEE, Helena Yan Ping LAU The Hong Kong Polytechnic University Hong Kong ym.lee@polyu.edu.hk, helena.lau@connect.polyu.hk Abstract The classification of implicit emotions in text has always been a great challenge to emotion processing. Even though the majority of emotion expressed implicitly, most previous attempts at emotions have focused on the examination of explicit emotions. The poor performance of existing emotion identification and classification models can partly be attributed to the disregard of implicit emotions. In view of this, this paper presents the development of a Chinese event-comment social media emotion corpus. The corpus deals with both explicit and implicit emotions with more emphasis being placed on the implicit ones. This paper specifically describes the data collection and annotation of the corpus. An annotation scheme has been proposed for the annotation of emotion-related information including the emotion type, the emotion cause, the emotion reaction, the use of rhetorical question, the opinion target (i.e. the semantic role in an event that triggers an emotion), etc. Corpus data shows that the annotated items are of great value to the identification of implicit emotions. We believe that the corpus will be a useful resource for both explicit and implicit emotion classification and detection as well as event classification. Keywords: Corpus, implicit emotion, explicit emotion 1. Introduction Emotion has long been extensively studied across various disciplines, including philosophy, psychology, sociology, computer science, and linguistics. In the fields of linguistics and natural language processing (NLP), the identification of emotion in text has been a great challenge to researchers as reflected by the poor performance of existing emotion classification models. This can be attributed to the fact that emotions are frequently expressed in an implicit way without using any emotion keywords. Despite the importance of implicit emotions, most previous studies have been dedicated to the examination of explicit emotions and not much work has been done on investigating implicit ones. The term explicit emotion refers to emotion-related information denoted by the presence of emotion keywords. For instance, the emotion keyword “delighted” in “I was delighted by the news” explicitly indicates the emotion of the experiencer. Implicit emotion refers to emotion-related information which is inferred by the readers, instead of being conveyed through emotion keywords. An example of such a sentence is “You shut your mouth!” which connotes an anger emotion with none of the individual words in the sentence expressing such an emotion. Given that there is a clear gap in existing emotion research, this paper aims to construct a Chinese event- comment emotion corpus which allows users to explore the characteristics of implicit emotion with annotated emotion- related information such as emotion type, emotion keyword, emotion expression, the use of rhetorical question, opinion target, emotion cause (i.e. pre-event) and emotion reaction (i.e. post-event). Regarding emotion causes, most emotion …

Content truncated to 500 words