🤩 Sentiment2Emoji 🤩
Find an emoji for a passage of text. This project uses machine learning to predict the sentiment of a given string and return an appropriate emoji. There is a web service provided using flask that is can be found at the demo link below.
Note — Emojis are best seen on updated iOS or a mac, sometimes the conversion for anger/surprised doesn't show up correctly
Emotions:
- Love
- Anger
- Joy
- Surprise
- Sadness
- Fear
input: "The last jedi was the greatest movie!"
output: 😍
input: "Collard greens are absolutely disgusting "
output: 😡
python app.py
https://sentiment2emoji.herokuapp.com/predict
Thank you
- Python
- Pandas
- Scikit-learn
- Numpy
- Matplotlib
- Data:
F. Calefato, F. Lanubile, N. Novielli. “[EmoTxt: A Toolkit for Emotion Recognition from Text](https://arxiv.org/abs/1708.03892)” To appear in *Proceedings of ACII 2017, the 7th International Conference on Affective Computing and Intelligent Interaction*, San Antonio, USA, Oct. 23-26, 2017.