This is the implementation of Aspect-based Sentiment Analysis withType-aware Graph Convolutional Networks and Layer Ensemble at NAACL 2021.

You can e-mail Yuanhe Tian at [email protected], if you have any questions.

Visit our homepage to find more our recent research and softwares for NLP (e.g., pre-trained LM, POS tagging, NER, sentiment analysis, relation extraction, datasets, etc.).

We are improving our ASA-TGCN. For updates, please visit HERE.

If you use or extend our work, please cite our paper at NAACL 2021.

@inproceedings{tian-etal-2021-aspect,
    title = "Aspect-based Sentiment Analysis with Type-aware Graph Convolutional Networks and Layer Ensemble",
    author = "Tian, Yuanhe and Chen, Guimin and Song, Yan",
    booktitle = "Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies",
    month = jun,
    year = "2021",
    address = "Online",
    publisher = "Association for Computational Linguistics",
    pages = "2910--2922"
}

Our code works with the following environment.

• python=3.7
• pytorch=1.3

To obtain the data, you can go to data directory for details.

In our paper, we use BERT (paper) as the encoder.

For BERT, please download pre-trained BERT-Base and BERT-Large English from Google or from HuggingFace. If you download it from Google, you need to convert the model from TensorFlow version to PyTorch version.

Run run_sample.sh to train a model on the small sample data under the sample_data directory.

Here are some important parameters:

• --do_train: train the model.
• --do_eval: test the model.

• Release the models.
• Regular maintenance.

You can leave comments in the Issues section, if you want us to implement any functions.