The data is basically a collection of tweets annotated with the emotions behind them. We have three columns tweet_id, sentiment, and content. In "content" we have the raw tweet. In "sentiment" we have the emotion behind the tweet. The data that we have is having 13 different emotion 40000 records. We need to use some advanced methods to build an efficient multiclass classification model.
This public domain dataset is collected from data.world platform. Thanks, data.world for releasing it under Public License.
๐ป Project Codes