This application creates the dataset through fetching tweets by their IDs.
There are some policies to avoid tweets content directly. Thus, there are many datasets with just tweets IDs and their labels. Hence, I've prepared an application to fetch these tweets context by their IDs using Twitter API. Consequently, you need a Twitter API Token http://apps.twitter.com.
In ./dataset
directory exists a dataset contains 1.3 million tweets in seven categories label (\ anger
\ , thankfulness
\ , joy
\ , sadness
\ , fear
\ , love
\ , surprise
\ ) by their IDs.
Installation
.. code-block::
pip install -r requirements.txt
Configuration
Customize the configuration file (\ ./configuration/configs.py
\ )
Run
.. code-block:: python
python -m tweet_getter.dataset_maker
[\ NOTE\ ]:
This package is compatible on Python 3