Sarcasm Detection Using Emoji and Text Analysis.
The dataset.xlsx file has the data stats and the data before and after preprocessing. I have taken the 12.9K input data (emoji + text) separately and put it in a csv file in the 'Data to work with' folder(zip). These are supposed to be the input files for word_to_vec and emoji_to_vec codes.
- Numpy
- Keras
- NLTK
- emoji
- gensim
- scikit-learn
- pandas
- codecs
- enchant
- Place data in /Data directory and run:
python Baseline/Run_ML.py
Change file names in settings.json and run:
python Model/BiLSTM+Attention.py