Skip to content

A machine learning model built to identify and classify the primary emotion expressed by tweets, designed with a CNN-BiLSTM based architecture using PyTorch and Keras. Trained on a dataset of over 200000 tweets collected and preprocessed with Twitter's API, achieving accuracies of ~70%!

Notifications You must be signed in to change notification settings

g-x-w/tweet_emoclass

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

42 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Primary Emotion Identification & Classification in Tweets

This github repository will hold the code and data used by Nicole Beri, Ruile Lu, and Gary Wei for our joint project for the course ECE324H1 at the University of Toronto held in the Winter session of the 2021-2022 academic year.

For the parties reviewing this repo and the accompanying report, here is a summary of the repository structure.

  • actual_code directory contains python files written to collect, aggregate, and process data, as well as the jupyter notebook file main_model.ipynb which houses our model declaration and training process
  • data_from_api directory contains csv files with the tweets pulled using the twitter API
  • manually_labelled directory contains tweets that were manually collected and labelled for testing purposes
  • tweets_labels directory contains the filtered tweets with only the tweet contents and label for training purposes

About

A machine learning model built to identify and classify the primary emotion expressed by tweets, designed with a CNN-BiLSTM based architecture using PyTorch and Keras. Trained on a dataset of over 200000 tweets collected and preprocessed with Twitter's API, achieving accuracies of ~70%!

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 3

  •  
  •  
  •