Skip to content

Unsupervised Style Transfer, transferring sentiment of reviews using classification attention weights.

License

Notifications You must be signed in to change notification settings

cs-mac/Unsupervised_Style_Transfer

Repository files navigation

Unsupervised Style Transfer: Automatic Sentiment Transfer Using Classification Attention Weights

Code for my Master's Thesis Information Science at the University of Groningen.

header image

How to run the models

First preprocess the data using: preprocess.py

Before we can train the HAN model we first need to get the POS-tags using: HAN/POS.ipynb

Next we can train the HAN model using: HAN/HAN+POS_attention_mechanism.ipynb

We can then use the style_generator.ipynb to generate sentences from one sentiment to the opposing sentiment

We can use train_evaluation.ipynb to train the classifiers for automatic classification

Lastly, we can use evaluation.ipynb and all_evaluation.ipynb to evaluate the human and automatic evaluations

Note:
- This is research code and might therefore not be fully complete. 
- For questions and full results contact the author.

Authors

License

This project is licensed under the Apache 2.0 License - see the LICENSE file for details

Acknowledgments

  • M. Nissim for mentoring my project giving me guidance and tips

About

Unsupervised Style Transfer, transferring sentiment of reviews using classification attention weights.

Topics

Resources

License

Stars

Watchers

Forks

Languages