Aspect-Category-Opinion-Sentiment (ACOS) Quadruple Extraction

This repo contains the data sets and source code of our paper:

Aspect-Category-Opinion-Sentiment Quadruple Extraction with Implicit Aspects and Opinions [ACL 2021].

• We introduce a new ABSA task, named Aspect-Category-Opinion-Sentiment Quadruple (ACOS) Extraction, to extract fine-grained ABSA Quadruples from product reviews;
• We construct two new datasets for the task, with ACOS quadruple annotations, and benchmark the task with four baseline systems;
• Our task and datasets provide a good support for discovering implicit opinion targets and implicit opinion expressions in product reviews.

Task

The Aspect-Category-Opinion-Sentiment (ACOS) Quadruple Extraction aims to extract all aspect-category-opinion-sentiment quadruples in a review sentence and provide full support for aspect-based sentiment analysis with implicit aspects and opinions.

Datasets

Two new datasets, Restaurant-ACOS and Laptop-ACOS, are constructed for the ACOS Quadruple Extraction task:

• Restaurant-ACOS is an extension of the existing SemEval Restaurant dataset, based on which we add the annotation of implicit aspects, implicit opinions, and the quadruples;
• Laptop-ACOS is a brand new one collected from the Amazon Laptop domain. It has twice size of the SemEval Loptop dataset, and is annotated with quadruples containing all explicit/implicit aspects and opinions.

The following table shows the comparison between our two ACOS Quadruple datasets and existing representative ABSA datasets.

Methods

We benchmark the ACOS Quadruple Extraction task with four baseline systems:

• Double-Propagation-ACOS
• JET-ACOS
• TAS-BERT-ACOS
• Extract-Classify-ACOS

We provided the source code of Extract-Classify-ACOS. The source code of the other three methods will be provided soon.

Overview of our Extract-Classify-ACOS method. The first step performs aspect-opinion co-extraction, and the second step predicts category-sentiment given the aspect-opinion pairs.

Results

The ACOS quadruple extraction performance of four different systems on the two datasets:

We further investigate the ability of different systems in addressing the implicit aspects/opinion problem:

Citation

If you use the data and code in your research, please cite our paper as follows:

@inproceedings{cai2021aspect,
  title={Aspect-Category-Opinion-Sentiment Quadruple Extraction with Implicit Aspects and Opinions},
  author={Cai, Hongjie and Xia, Rui and Yu, Jianfei},
  booktitle={Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers)},
  pages={340--350},
  year={2021}
}