🤩 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.

The project uses Flask for the backend server along with Swagger docs for the service. It also uses a Vue frontend hosted on github.

Website

API Docs

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

Sample

input: "The last jedi was the greatest movie!"

output: 😍

input: "Collard greens are absolutely disgusting "

output: 😡

Quick Start for Flask app

python app.py

Demo

https://sentiment2emoji.herokuapp.com/predict

Frontend

Using Vue and Vuetify

vue create frontend
vue add vuetify
npm run build

Resources and References

Thank you

• Python
• Flask restplus
• Vue
• 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.