Citation

Ng, Hu and Chong, Wing Kin and Chia, Yu Zhang and Yap, Timothy Tzen Vun and Goh, Vik Tor and Wong, Lai Kuan and Tan, Ian Kim Teck and Cher, Dong Theng (2024) Aspect-Based Subjectivity Analysis Using a BERT-based Approach. In: 2024 IEEE 14th Symposium on Computer Applications & Industrial Electronics (ISCAIE), 24-25 May 2024, Penang, Malaysia.

Text Aspect-Based Subjectivity Analysis Using a BERT-based Approach.pdf - Published Version Restricted to Repository staff only Download (1MB)

Abstract

Aspect-based subjectivity analysis stands as an important task in natural language processing, seeking to identify the subjectivity of various aspects or features within a text. A new method for aspect-based subjectivity analysis using BERT is introduced in this paper. BERT has demonstrated impressive performance across various NLP tasks, and its capabilities are utilized to accurately ascertain the subjectivity of specific aspects within a given text. The approach involves fine-tuning BERT on a sizable dataset annotated with aspect-level subjectivity labels, enabling the model to grasp the subtleties of aspect-based subjectivity analysis. Extensive experiments on benchmark datasets are conducted to showcase the effectiveness of this approach and compare it with existing methods. The results reveal that this proposed approach surpasses state-of-the-art techniques in aspect-based subjectivity analysis, underscoring the potential of leveraging BERT for such purposes.

Item Type:	Conference or Workshop Item (Paper)
Uncontrolled Keywords:	Subjectivity Analysis , NLP , Aspect-based
Subjects:	Q Science > QA Mathematics > QA71-90 Instruments and machines
Divisions:	Faculty of Computing and Informatics (FCI)
Depositing User:	Ms Nurul Iqtiani Ahmad
Date Deposited:	31 Jul 2024 03:55
Last Modified:	31 Jul 2024 03:55
URII:	http://shdl.mmu.edu.my/id/eprint/12676

Downloads

Edit (login required)