/sentiAnnotator140

Text annotation tool.

Primary LanguagePython

sentiAnnotator140

image

This can be use as a text labeling tool which has:

  • An input text box for setting your raw dataset as '.csv' file.
  • 4 kinds of indicators:

  • A numerical indicator which display predicted sentiment value of specified text(this attain from text blob libray)

  • A green/red rectangular indicator which determine if there is a positive emotional emoji in the specified text or negative emotional emoji.

  • Two textual indicators which display what you selected for labels (the left one display numerical label (1-5) and the right one display textual label ("Buy/Sell").

  • 5 buttons to assigan a numerical label (1 to 5) to the text.
  • 2 buttons to assign "Buy/Sell" label to specified text. (It could be helpful if you are labeling financial tweets like)
  • Next and Previous button to switch between tweets

Usage:

1. In windows open cmd and run pip install -r requirement.txt to install dependencies

2. Make a ".csv" file (must be in ".csv" format and having three columns as bellow):

  • Signal (leave it's value to be None)
  • handySentiment (leave it's value to be None)
  • tweet (fill this by your dataset tweets)

2. Run "SentiAnnotator.py"

3. Enter ".csv" file which created in the first step and click "Enter" button

4. Do Some labeling

5. Finalize your changes on dataset by click on Finish button and exit.