Adversarial Training for Aspect-Based Sentiment Analysis with BERT

Code for "Adversarial Training for Aspect-Based Sentiment Analysis with BERT".

We have used the codebase from the following paper and improved upon their results by applying adversarial training. "BERT Post-Training for Review Reading Comprehension and Aspect-based Sentiment Analysis".

Running

Place laptop and restaurant post-trained BERTs into pt_model/laptop_pt and pt_model/rest_pt, respectively. The post-trained Laptop weights can be download here and restaurant here.

Execute the following command to run the model for Aspect Extraction task:

script\run_ae.bat ae laptop_pt laptop pt_ae 9

Here, laptop_pt is the post-trained weights for laptop, laptop is the domain, pt_ae is the fine-tuned folder in run/, 9 means run 9 times.

Similarly,

script\run_ae.bat ae rest_pt rest pt_ae 9

Evaluation

Execute the following command to evaluate the model for Aspect Extraction task:

eval\run_ae_eval.bat laptop 9

Here laptop is the domain, 9 means run 9 predictions corresponding to 9 runs

The evaluation additionally needs Java JRE/JDK to be installed.

Open result.ipynb and check the results.

Citation

@misc{karimi2020adversarial,
    title={Adversarial Training for Aspect-Based Sentiment Analysis with BERT},
    author={Akbar Karimi and Leonardo Rossi and Andrea Prati and Katharina Full},
    year={2020},
    eprint={2001.11316},
    archivePrefix={arXiv},
    primaryClass={cs.LG}
}