In this project, we attempt to solve the problem of Emotion detection in text based data. Recent developments in neural networks and sequence based models have shown great success in being able to detect emotions like fear, joy, happiness, sadness, etc from text. We compare 4 types of Deep Learning models on this problem- Recurrent Neural Network (RNN), Long Short Term Memory (LSTM), Gated recurrent Unit (GRU), and BERT, and contrast them with the baseline model of logistic regression. We demonstrate the abilities of these models on two datasets of different sample spaces.

1. Download the 300 dimensional Glove embeddings
2. Place the file in the Embeddings folder
3. You can run the othermodels.eld_othermodels.py file
4. You can run the othermodels.isears_othermodels
5. Wait for results. Check the summary of results.