Comparative Opinion Quintuple Extraction (COQE)

This repo contains the annotated data and code for our paper [eGen: An Enhanced Generative Framework for Comparative Opinion Quintuple Extraction]

Short Summary

• We aim to tackle COQE task: given a sentence, we predict all comparative quads subject (\textit{sub}), object (\textit{obj}), comparative aspect (\textit{ca}), comparative opinion (\textit{co}), and comparative preference (\textit{cp})

Data

• We use two three datasets, Camera-COQE, Car-COQE and Ele-COQE:

• Camera-COQE: On basis of the Camera domain corpus released by Kessler and Kuhn (2014), we completed the annotation of Comparative Opinion and Comparative Preference for 1705 comparative sentences, and introducing 1599 non-comparative sentences.

• Car-COQE: Based on the COAE 2012/2013 Car domain corpus, we supplemented with the annotation of Comparative Opinion and Comparative Preference.

• Ele-COQE: Similar to Car-COQE, we construct the Ele-COQE dataset based on the COAE 2012/2013 Electronics (Ele) domain corpus.

Requirements

We highly recommend you to install the specified version of the following packages to avoid unnecessary troubles:

• transformers==4.0.0
• sentencepiece==0.1.91
• pytorch_lightning==0.8.1

Quick Start

• Set up the environment as described in the above section
• Download the pre-trained T5-base model (you can also use larger versions for better performance depending on the availability of the computation resource), put it under the folder T5-base.
  • You can also skip this step and the pre-trained model would be automatically downloaded to the cache in the next step
• Run command sh run.sh.
• More details can be found in the paper and the help info in the main.py.

Citation

If the code is used in your research, please star our repo and cite our paper as follows: