A Quantitative Analysis of Comparison of Emoji Sentiment: Taiwan Mandarin Users and English Users

Abstract Emojis have become essential components in our digital communication. Emojis, especially smiley face emojis and heart emojis, are considered the ones conveying more emotions. In this paper, two functions of emoji usages are discussed across two languages, Taiwanese Mandarin and English. The first function discussed here is sentiment enhancement and the other is sentiment modification. Multilingual language model is adopted for seeing the probability distribution of the text sentiment, and relative entropy is used to quantify the degree of changes. The results support the previous research that emojis are more frequently-used in positive contexts, smileys tend to be used for expressing emotions and prove the language-independent nature of emojis. Keywords: emoji, sentiment enhancement, sentiment modification 1 Introduction With the considerable growth of social media, emojis have become increasingly popular and widely used across the world. During the last few years, emojis change the way we communicate online. They allow us to interact with each other more clearly when we struggle to express our emotions through pure texts. It is an explicit acknowledgment that emojis are now part of how we express our emotions, intents and feelings. Aside from emotion expression, emojis are believed to enhance and modify the sentiment of a text. A sentence may convey an emotion, but its emotion would be strengthened after the addition of emojis, which is called sentiment enhancement. On the other hand, if the emotion of the sentence is weakened or altered to the opposite emotion, that is called sentiment modification. Moreover, research shows that emojis used to convey emotions are mostly smileys and hearts, which belong to the Smiley & Emotion group in the Unicode emoji categories. However, though emojis’ non-verbal nature suggests that they are universal across cultures, their usages may change from language to language and culture to culture. Over the past few years, great concern has arisen in the research of cross-cultural or cross-language emoji usage. Some of them discuss which kinds of emoji patterns are the same or different across cultures. Some studies build the emoji sentiment lexicon with the data from different languages. But little was done on the degree of how each emoji can enhance or modify the text across languages. This paper aims to compare the degree of sentiment enhancement and sentiment modification of emojis which are used by Taiwanese Mandarin users and English users with the help of a multilingual language model. Recently, large pre-trained neural models such as BERT have achieved great success in NLP, motivating more and more research to investigate what aspects of language they are able to learn from unlabeled data. 2 Background and Related Work Many people believe emojis are like the older emoticons, which provide a visual representation using punctuation marks. Emojis and emoticons function as the non-verbal cues (paralanguages) in face-to-face communication, which are believed to convey emotions more effectively and efficiently than the words themselves are saying. In fact, the reason punctuation marks came into existence was to complement emotional engagement in written texts. (Evans, 2017) …

Content truncated to 500 words