An Uncertainty-Aware Encoder for Aspect Detection

Thi-Nhung Nguyen, Kiem-Hieu Nguyen, Young-In Song, Tuan-Dung Cao


Abstract
Aspect detection is a fundamental task in opinion mining. Previous works use seed words either as priors of topic models, as anchors to guide the learning of aspects, or as features of aspect classifiers. This paper presents a novel weakly-supervised method to exploit seed words for aspect detection based on an encoder architecture. The encoder maps segments and aspects into a low-dimensional embedding space. The goal is approximating similarity between segments and aspects in the embedding space and their ground-truth similarity generated from seed words. An objective function is proposed to capture the uncertainty of ground-truth similarity. Our method outperforms previous works on several benchmarks in various domains.
Anthology ID:
2021.findings-emnlp.69
Volume:
Findings of the Association for Computational Linguistics: EMNLP 2021
Month:
November
Year:
2021
Address:
Punta Cana, Dominican Republic
Venues:
EMNLP | Findings
SIG:
SIGDAT
Publisher:
Association for Computational Linguistics
Note:
Pages:
797–806
Language:
URL:
https://aclanthology.org/2021.findings-emnlp.69
DOI:
10.18653/v1/2021.findings-emnlp.69
Bibkey:
Cite (ACL):
Thi-Nhung Nguyen, Kiem-Hieu Nguyen, Young-In Song, and Tuan-Dung Cao. 2021. An Uncertainty-Aware Encoder for Aspect Detection. In Findings of the Association for Computational Linguistics: EMNLP 2021, pages 797–806, Punta Cana, Dominican Republic. Association for Computational Linguistics.
Cite (Informal):
An Uncertainty-Aware Encoder for Aspect Detection (Nguyen et al., Findings 2021)
Copy Citation:
PDF:
https://preview.aclanthology.org/update-css-js/2021.findings-emnlp.69.pdf
Data
OpoSum