An empirical study of emoji usage on Twitter in linguistic and national contexts

.

Abstract:

Emojis or ‘picture characters’ have become ubiquitous in modern-day digital communication, including social media sharing and smartphone texting. 

Despite this ubiquity, many questions remain about their usage, especially with respect to global variations in language and country. 

These questions are important, in part because they reveal how people communicate digitally on social platforms, but also because they provide a lens through which different regions and cultures can be studied. 

In this paper, we conduct a principled, quantitative study to understand emoji usage in terms of linguistic and country correlates. 

Our study involves 30 languages and countries each, and is conducted over tens of millions of tweets collected from the Twitter decahose over an entire month. 

Drawing on both statistical measures and information theory, our results reveal that, not only does emoji usage have strong dependencies at both the language and country level, but that some languages and countries are much more constrained in the diversity of their emoji usage. 

However, we also discover that the ‘popularity’ of emojis, both globally and within the context of a given language, follows a robust and invariant trend that emerges fairly quickly (over just a day’s worth of data) and cannot be explained either by a power-law or Heap’s law-like distribution.

.

https://www.sciencedirect.com/science/article/pii/S2468696421000318

.

https://www.semanticscholar.org/paper/An-empirical-study-of-emoji-usage-on-Twitter-in-and-Kejriwal-Wang/506411ea9600fb78f54aa9dcb2be537af56980b4

.

https://doi.org/10.1016/j.osnem.2021.100149

.

Read More

Emojis influence emotional communication, social attributions, and information processing

.

Abstract:

Many emojis symbolize nonverbal cues that are used during face-to-face communication. 

Despite their popularity, few studies have examined how emojis influence digital interactions. 

The present study addresses this gap by measuring the impact of emojis on emotion interpretation, social attributions, and information processing. 

Participants read messages that are typical of social exchanges in instant text messaging (IM) accompanied by emojis that mimic negative, positive and neutral facial expressions. 

Sentence valence and emoji valence were paired in a fully crossed design such that verbal and nonverbal messages were either congruent or incongruent. 

Perceived emotional state of the sender, perceived warmth, and patterns of eye movements that reflect information processing were measured. 

A negativity effect was observed whereby the sender's mood was perceived as negative when a negative emoji and/or a negative sentence were presented. 

Moreover, the presence of a negative emoji intensified the perceived negativity of negative sentences. 

Adding a positive emoji to a message increased the perceived warmth of the sender. 

Finally, processing speed and understanding of verbal messages was enhanced by the presence of congruent emojis. 

Our results therefore support the use of emojis, and in particular positive emojis, to improve communication, express feelings, and make a positive impression during socially-driven digital interactions.

.

https://www.sciencedirect.com/science/article/abs/pii/S0747563221000443

.

https://www.semanticscholar.org/paper/Emojis-influence-emotional-communication%2C-social-Boutet-LeBlanc/c7d4859af5e1f27a6605697b9c17e704d7926704

.

https://doi.org/10.1016/j.chb.2021.106722

Read More

How emotional are emoji?: Exploring the effect of emotional valence on the processing of emoji stimuli

.

Abstract:

Emoji are vastly becoming an integral part of everyday communication, yet little is understood about the extent to which these are processed emotionally. 

Previous research shows that there is a processing advantage for emotionally-valenced words over neutral ones, therefore if emoji are indeed emotional, one could expect an equivalent processing advantage. 

In the Pilot Study, participants (N = 44) completed a lexical decision task to explore accuracy and response latency of word, face and emoji stimuli. 

This stimuli varied in emotional valence (positive vs. neutral). 

Main effects were found for stimuli type and valence on both accuracy and latency, although the interaction for accuracy was not significant. 

That is, there were processing advantages of positively-valenced stimuli over neutral ones, across all stimuli types. 

Also, faces and emoji were processed significantly more quickly than words, and latencies between face and emoji stimuli, irrespective of valence were largely equivalent. 

The Main Study recruited 33 participants to undertake a modified and extended version of the lexical decision task, which included three valence conditions (positive, negative and neutral) per stimuli type. 

Although no main effects were found for accuracy, there was a significant main effect found for stimuli but not for valence on latency. 

Namely, that word stimuli irrespective of valence were processed significantly more slowly than face or emoji stimuli. 

There was not a significant interaction between stimuli and valence, however. 

Therefore, overall although there was partial support for a processing advantage of emoji stimuli, this was not replicated across the studies reported here, suggesting additional work may be needed to corroborate further evidence.

. 

https://www.sciencedirect.com/science/article/abs/pii/S0747563220303952

.

https://www.semanticscholar.org/paper/How-emotional-are-emoji%3A-Exploring-the-effect-of-on-Kaye-Rodriguez-Cuadrado/adb03320ca10aecb0f8392dd4a695baade573897

.

https://doi.org/10.1016/j.chb.2020.106648

.

Dataset:

https://osf.io/asmh7/?view_only=14c499529edf4acfb28b3c327af234ab

.

PlumX Metrics:

https://plu.mx/plum/a/?doi=10.1016/j.chb.2020.106648

.

Read More

Seq2Emoji: A hybrid sequence generation model for short text emoji prediction

 .

Abstract: 

As a new form of visual language, emojis are widely used in social media for their vivid image and rich meaning. 

Predicting the most likely emojis that fit a particular short text has become an important and challenging task in both academia and industry. 

In this paper, we propose a hybrid sequence generation model, Seq2Emoji, to predict multiple emojis based on a short text. 

Seq2Emoji is an encoder–decoder model, in which we consider the correlations between emojis and take the emoji prediction task as a sequence generation problem. 

It extracts features through a hierarchical structure and self-attention mechanism and decodes them with a composite recurrent neural network before predicting emojis. 

During the prediction, Diverse Beam Search algorithm is also introduced to increase the diversity of predicted emojis. 

Experiments are carried out on our collected Weibo dataset (Chinese) and the results show that our proposed Seq2Emoji model is superior to the competitive models in both accuracy and diversity of emoji prediction.

.

https://www.sciencedirect.com/science/article/abs/pii/S095070512030856X

.

https://www.semanticscholar.org/paper/Seq2Emoji%3A-A-hybrid-sequence-generation-model-for-Peng-Zhao/ab802b7223e33a38ceb8492c34ca0aaf45c47182

.

https://doi.org/10.1016/j.knosys.2020.106727

.

Read More