"Blissfully Happy" or "Ready toFight": Varying Interpretations of Emoji
By Hannah Jean Miller, Jacob Thebault-Spieker, Shuo Chang, Isaac L. Johnson, Loreen Terveen, Brent J. Hecht. 2016
405 citations recorded by the source corpus.
Emoji are commonly used in modern text communication. However, as graphics with nuanced details, emoji may be open to interpretation. Emoji also render differently on different viewing platforms (e.g., Apple’s iPhone vs. Google’s Nexus phone), potentially leading to communication errors. We explore whether emoji renderings or differences across platforms give rise to diverse interpretations of emoji. Through an online survey, we solicit people’s interpretations of a sample of the most popular emoji characters, each rendered for multiple platforms. Both in terms of sentiment and semantics, we analyze the variance in interpretation of the emoji, quantifying which emoji are most (and least) likely to be misinterpreted. In cases in which participants rated the same emoji rendering, they disagreed on whether the sentiment was positive, neutral, or negative 25% of the time. When considering renderings across platforms, these disagreements only increase. Overall, we find significant potential for miscommunication, both for individual emoji renderings and for different emoji renderings across platforms.
Subfields: Artificial Intelligence, Natural Language Processing, Sentiment & Opinion Analysis, Human-Computer Interaction
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Sebastian Riedel
- Mohit Bansal
- Tim Rocktaschel
- José Camacho-Collados
- Luke S. Zettlemoyer
- Francesco Barbieri
- Isabelle Augenstein
- Yejin Choi
In Natural Language Processing
- Sebastian Riedel
- Tim Rocktaschel
- Viviana Patti
- Valerio Basile
- Isabelle Augenstein
- Luis Espinosa Anke
- José Camacho-Collados
- Francesco Barbieri
In Sentiment & Opinion Analysis
In Human-Computer Interaction
- José Camacho-Collados
- Francesco Barbieri
- Isabelle Augenstein
- Patrick Hall
- Qi Dan
- Zhe Mu
- Maokun Yang
- Emre Kıcıman
Citations in the corpus
- Social Data: Biases, Methodological Pitfalls, and Ethical Boundaries (2019, background): “For instance, Miller et al. (2016) show that variations in how emojis are displayed across smartphones can lead to confusion among users, as different renderings of the same concept are so different that they might be interpreted as having different meanings and emotional valence.”
- SemEval 2018 Task 2: Multilingual Emoji Prediction (2018, background): “) or app (Whatsapp, Twitter, Facebook, and so forth)3 (Tigwell and Flatla, 2016; Miller et al., 2016).”
- An exploratory study of COVID-19 misinformation on Twitter (2020, neutral): “Both are culturally and contextually bound, as well as influenced by age and gender [96] and open to changes in their interpretation over time [97].”
- A Systematic Review of Emoji: Current Research and Future Perspectives (2019, methodology): “The specific meanings that users want to express by emoji may be different from their official definitions, resulting in different interpretations of the same emoji (Miller et al., 2016) ( Table 5 ).”
- Domain-Specific Use Cases for Knowledge-Enabled Social Media Analysis (2018, background): “For example, (face with tears of joy) can be used for expressing happiness (using senses such as laugh and joy) as well as sadness (using senses such as cry and tears) [22].”
- Towards a Standard for Identifying and Managing Bias in Artificial Intelligence (2022, background): “behavioral bias Systematic distortions in user behavior across platforms or contexts, or across users represented in different datasets [147, 273].”
- emoji2vec: Learning Emoji Representations from their Description (2016, result): “(Lebduska, 2014) showed that emoji are culturally and contextually bound, and are open to reinterpretation and misinterpretation, a result confirmed by (Miller et al., 2016).”