Skip to content

Latest commit

 

History

History
32 lines (23 loc) · 945 Bytes

readme.md

File metadata and controls

32 lines (23 loc) · 945 Bytes

Feiwen

A tool for modeliing documents containing emojis.

Setup

This project partially depends on python, R and emoji2vec, which has been added as a git submodule in this project. In order to install all dependencies, you need to:

  1. Have mini conda installed and executable your command line prompt.
  2. Create the feiwen python virtual environment: conda env create -f environment.yml
  3. git submodule update --init --recursive (This pulls emoji2vec into libs/emoji2vec)

TODO:

  1. Get some data, the format should be
{
	"id": STRING, 
	"text": "....",
	"sentiment": "POS" || "NEG",
	"emoji_class": "+" || "-"
}
  1. Split the data by emoji class
  2. Get emoji2vec vectors
  3. Generate doc2vec based on (2) and (3)
  4. Visualise the data in 2 by something like tsne

References: