# A `Sourceful' Twist: Emoji Prediction Based on Sentiment, Hashtags and Application Source

arXiv: https://arxiv.org/abs/2103.07833 · HTML: https://mlsanity.com/paper/2103.07833/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 5 of its 16 references that are in the map.
- First posted on arXiv 14 Mar 2021.

## Builds on (most cited first, 5 of 5)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Using millions of emoji occurrences to learn any-domain representations for detecting sentiment, emotion and sarcasm](https://mlsanity.com/paper/1708.00524/) (2017)
- [Sentiment of Emojis](https://mlsanity.com/paper/1509.07761/) (2015)
- [emoji2vec: Learning Emoji Representations from their Description](https://mlsanity.com/paper/1609.08359/) (2016)
- [Twitter Sentiment Analysis via Bi-sense Emoji Embedding and Attention-based LSTM](https://mlsanity.com/paper/1807.07961/) (2018)

## Built on by (most cited first, 3 of 3)

- [The Sentiment Problem: A Critical Survey towards Deconstructing Sentiment Analysis](https://mlsanity.com/paper/2310.12318/) (2023)
- [Federated Learning Based Multilingual Emoji Prediction In Clean and Attack Scenarios](https://mlsanity.com/paper/2304.01005/) (2023)
- [Machine learning and emoji prediction: How much accuracy can MARBERT achieve?](https://mlsanity.com/paper/2604.21108/) (2026)

## Abstract (arXiv)

We widely use emojis in social networking to heighten, mitigate or negate the sentiment of the text. Emoji suggestions already exist in many cross-platform applications but an emoji is predicted solely based a few prominent words instead of understanding the subject and substance of the text. Through this paper, we showcase the importance of using Twitter features to help the model understand the sentiment involved and hence to predict the most suitable emoji for the text. Hashtags and Application Sources like Android, etc. are two features which we found to be important yet underused in emoji prediction and Twitter sentiment analysis on the whole. To approach this shortcoming and to further understand emoji behavioral patterns, we propose a more balanced dataset by crawling additional Twitter data, including timestamp, hashtags, and application source acting as additional attributes to the tweet. Our data analysis and neural network model performance evaluations depict that using hashtags and application sources as features allows to encode different information and is effective in emoji prediction.