The Emoji Sentiment Lexicon: Analysing Consumer Emotions in Social Media Communication
(A2020-63029)

AUTHORS

Lisa Holthoff, University of Duisburg-Essen

KEYWORDS

Categorical Sentiment Analysis; Social Media Communication; Emoji Lexicon

ABSTRACT

Due to the ongoing digitalization of communication, many companies are increasingly interested in tools that allow them to analyze their consumers’ brand-related online messages. Current sentiment analysis tools are able to analyze the emotional tone of written text automatically. However, since consumers often use emojis to express emotional subtext in social media messages, I propose the additional integration of emojis into these tools. Study 1, an online experiment with 1,519 participants, confirms that the inclusion of emojis increases the correct interpretation of brand-related tweets (with regard to polarity as well as emotion category). In study 2, a qualitative study with 1,157 participants, I develop a categorical emoji lexicon for unsupervised sentiment analyses. Finally, I validate the resulting lexicon in a quantitative study with 1,926 participants. The lexicon improves and simplifies the identification of emotional consumer reactions in social media communication.