https://github.com/EmreTaha/Unsupervised-Domain-Adaptation-with-BERT

Unsupervised domain adaptation with BERT for Amazon food product reviews sentiment analysis.
https://github.com/EmreTaha/Unsupervised-Domain-Adaptation-with-BERT

adversarial-learning amazon-food-reviews bert bert-model colab domain-adaptation nlp sentiment-analysis tensorflow unsupervised-learning

Last synced: 27 days ago
JSON representation

Unsupervised domain adaptation with BERT for Amazon food product reviews sentiment analysis.

• Host: GitHub
• URL: https://github.com/EmreTaha/Unsupervised-Domain-Adaptation-with-BERT
• Owner: EmreTaha
• License: gpl-3.0
• Created: 2019-09-30T14:08:15.000Z (about 5 years ago)
• Default Branch: master
• Last Pushed: 2020-10-06T00:39:09.000Z (about 4 years ago)
• Last Synced: 2024-08-03T05:02:11.679Z (4 months ago)
• Topics: adversarial-learning, amazon-food-reviews, bert, bert-model, colab, domain-adaptation, nlp, sentiment-analysis, tensorflow, unsupervised-learning
• Language: Jupyter Notebook
• Size: 156 KB
• Stars: 14
• Watchers: 2
• Forks: 2
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

• awesome-domain-adaptation-NLP - [__code__

README

        
# Unsupervised-Domain-Adaptation-with-BERT

A novel approach for BERT usage in an adversarial unsupervised domain adaptation manner for a NLP tasks. The topic is **Unsupervised domain adaptation between two Amazon product reviews categories with BERT and a domain discriminator network for the sentiment analysis**.

Please cite original [BERT](https://arxiv.org/abs/1810.04805) paper when using the code.

The code based on BERT in the TF-Hub. `BERTOptimizer.py` file is modified for freezing the network partially. 

## Getting Started

Upload ipynb file, `BERTOptimizer.py`, `utils.py` files to Google Colab. The data (from Stanford) is downloaded and processed within the code

## Requirements

* Python 3.6

* pandas 

* Tensorflow 1.x

* Numpy

* Matplotlib

* Scipy

* Google Colab