🤖 AI Summary
This study addresses the lack of systematic longitudinal analysis of emoji semantic evolution over time. Method: Leveraging a large-scale Twitter corpus from 2012–2018, we apply dynamic word embedding models and computational linguistic techniques to track semantic shifts. Contribution/Results: We identify five distinct patterns of semantic change, empirically demonstrate that lower-iconicity emojis exhibit greater semantic drift, and quantify significant impacts of seasonal cycles and major socio-cultural events on emoji semantics. We introduce EmojiEvolve—the first publicly available, timestamp-annotated dataset for emoji semantic evolution—along with an interpretable, quantitative metric for measuring semantic change. Additionally, we develop an interactive web platform enabling multidimensional exploration. These contributions establish a new paradigm for modeling semantic evolution in social media, advancing cross-temporal emoji understanding and temporal representation learning in NLP.
📝 Abstract
The semantics of emoji has, to date, been considered from a static perspective. We offer the first longitudinal study of how emoji semantics changes over time, applying techniques from computational linguistics to six years of Twitter data. We identify five patterns in emoji semantic development and find evidence that the less abstract an emoji is, the more likely it is to undergo semantic change. In addition, we analyse select emoji in more detail, examining the effect of seasonality and world events on emoji semantics. To aid future work on emoji and semantics, we make our data publicly available along with a web-based interface that anyone can use to explore semantic change in emoji.
