Kaggle Toxic Comment Classification Challenge: (https://www.kaggle.com/c/jigsaw-toxic-comment-classification-challenge)
Single LSTM + GRU Model with 10 fold CV yields a ROC-AUC score of 0.9871 against Public LB highest of 0.9890 with current solution ranked 300th on Public LB
Additional Details:
- Embedding Vectors - fastText & GloVe Twitter (200d)
- Implementation Libraries - Pytorch (Model) & Keras (Text Pre-processing)
Potential Areas of Improvement:
- Modifying model architecture with focus on better regularization
- Ensembling (though ensembling with NB-SVM baseline did not help improve the score)
Note - Did not use BERT baseline since it wasn't released at the time of competition