ABSA-Bench: Towards the Unified Evaluation of Aspect-based Sentiment Analysis Research

Abhishek Das, Wei Emma Zhang


Abstract
Aspect-Based Sentiment Analysis (ABSA)has gained much attention in recent years. It is the task of identifying fine-grained opinionpolarity towards a specific aspect associated with a given target. However, there is a lack of benchmarking platform to provide a unified environment under consistent evaluation criteria for ABSA, resulting in the difficulties for fair comparisons. In this work, we address this issue and define a benchmark, ABSA-Bench, by unifying the evaluation protocols and the pre-processed publicly available datasets in a Web-based platform. ABSA-Bench provides two means of evaluations for participants to submit their predictions or models for online evaluation. Performances are ranked in the leader board and a discussion forum is supported to serve as a collaborative platform for academics and researchers to discuss queries.
Anthology ID:
2020.alta-1.7
Volume:
Proceedings of the The 18th Annual Workshop of the Australasian Language Technology Association
Month:
December
Year:
2020
Address:
Virtual Workshop
Venue:
ALTA
SIG:
Publisher:
Australasian Language Technology Association
Note:
Pages:
65–71
Language:
URL:
https://aclanthology.org/2020.alta-1.7
DOI:
Bibkey:
Cite (ACL):
Abhishek Das and Wei Emma Zhang. 2020. ABSA-Bench: Towards the Unified Evaluation of Aspect-based Sentiment Analysis Research. In Proceedings of the The 18th Annual Workshop of the Australasian Language Technology Association, pages 65–71, Virtual Workshop. Australasian Language Technology Association.
Cite (Informal):
ABSA-Bench: Towards the Unified Evaluation of Aspect-based Sentiment Analysis Research (Das & Zhang, ALTA 2020)
Copy Citation:
PDF:
https://preview.aclanthology.org/author-url/2020.alta-1.7.pdf
Data
LinCEQuACdecaNLP