Semantic Journeys: Quantifying Change in Emoji Meaning from 2012-2018

TL;DR

This work quantifies how emoji meaning evolves over six years by translating word-level semantic-change techniques to emoji via a local neighbourhood SC score and time-series clustering. It reveals that most emoji remain semantically stable, while a minority show substantive, pattern-driven changes including gradual establishment, sudden peaks, and seasonality, with concrete emoji more likely to shift than abstract ones. The study also demonstrates that some semantic changes correlate with world events and memes (e.g., Pepe the Frog) and that these trajectories can be linked to external signals like Google Trends. By releasing a public dataset and an interactive website, the authors provide a resource for researchers and practitioners to analyze emoji semantics over time and to incorporate diachronic semantics into NLP systems. Overall, the paper establishes a foundation for understanding diachronic emoji semantics and suggests directions for extending linguistic theories of semantic change to emoji."

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.

Figures (8)

• Figure 1: Mean semantic change change score of emoji over time. The standard deviation of the data at each point is shown by the shaded areas. Upper-left plot: all emoji. Others: specific emoji releases. Dashed vertical lines denote addition of new emoji to the Unicode Standard by the Unicode Consortium. The solid vertical line denotes the date of release for that Unicode/Emoji version. First occurrences of emoji lag behind their Unicode addition as they are not immediately implemented by vendors such as Apple and Google.
• Figure 2: Mean semantic score change over time for Unicode 6.x emoji, grouped by percentile based on standard deviation of their month-to-month semantic change scores. The standard deviation of the data at each point is shown by the shaded areas.
• Figure 3: Semantic change over time for 5 random emoji in the 50th (top), 75th, 90th, 95th and 99th (bottom) percentile, based on standard deviation of month-to-month semantic change for Unicode 6.x emoji with 2012 anchor points.
• Figure 4: Progression of semantic change in 348 Unicode 6.x emoji, with 2012 anchor points, from 2012 to 2018. Each plot shows a cluster's characteristic shape (dashed lines) and the mean of the actual observed semantic change scores (solid lines). Top left shows the averages over all clusters. The standard deviation of the data at each point is shown by the shaded areas.
• Figure 5: Distribution of the number of senses per emoji, based on EmojiNet data linked to BabelNet senses.