Skip to main navigation Skip to search Skip to main content

Outperforming state-of-the-art systems for aspect-based sentiment analysis

  • Jordan University of Science and Technology
  • Princess Sumaya University for Technology

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

1 Scopus citations

Abstract

Aspect-Based Sentiment Analysis (ABSA) is a very important problem with numerous applications. The three editions of SemEval's ABSA Shared Task have been instrumental in fostering the development in this field. One of its sub-tasks is the sentence-level ABSA. This sub-task has received a lot of attention and new techniques and better results are reported on it frequently. The purpose of this work is to achieve the highest accuracy for this problem. We follow a state-of-the-art (SOTA) approach that is based on multi-grain attention networks and infuse it with better embedding mechanisms in order to improve the results. For the famous SemEval's ABSA Shared Task, the results of the SOTA approaches reach 81.25 accuracy and 71.94 F1 score, whereas our approach surpasses them with 83.75 accuracy and 75.75 F1 score.

Original languageEnglish
Title of host publication16th ACS/IEEE International Conference on Computer Systems and Applications, AICCSA 2019
PublisherIEEE Computer Society
ISBN (Electronic)9781728150529
DOIs
StatePublished - Nov 2019
Externally publishedYes
Event16th ACS/IEEE International Conference on Computer Systems and Applications, AICCSA 2019 - Abu Dhabi, United Arab Emirates
Duration: 3 Nov 20197 Nov 2019

Publication series

NameProceedings of IEEE/ACS International Conference on Computer Systems and Applications, AICCSA
Volume2019-November
ISSN (Print)2161-5322
ISSN (Electronic)2161-5330

Conference

Conference16th ACS/IEEE International Conference on Computer Systems and Applications, AICCSA 2019
Country/TerritoryUnited Arab Emirates
CityAbu Dhabi
Period3/11/197/11/19

Keywords

  • ELMo
  • FastText
  • GloVe
  • MGAN
  • SemEval ABSA

Fingerprint

Dive into the research topics of 'Outperforming state-of-the-art systems for aspect-based sentiment analysis'. Together they form a unique fingerprint.

Cite this