Exploring zero-shot and joint training cross-lingual strategies for aspect-based sentiment analysis based on contextualized multilingual language models
ABSTRACTAspect-based sentiment analysis (ABSA) has attracted many researchers' attention in recent years. However, the lack of benchmark datasets for specific languages is a common challenge because of the prohibitive cost of manual annotation. The zero-shot cross-lingual strategy can be applie...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Taylor & Francis Group
2023-04-01
|
Series: | Journal of Information and Telecommunication |
Subjects: | |
Online Access: | https://www.tandfonline.com/doi/10.1080/24751839.2023.2173843 |
_version_ | 1797837434876067840 |
---|---|
author | Dang Van Thin Hung Quoc Ngo Duong Ngoc Hao Ngan Luu-Thuy Nguyen |
author_facet | Dang Van Thin Hung Quoc Ngo Duong Ngoc Hao Ngan Luu-Thuy Nguyen |
author_sort | Dang Van Thin |
collection | DOAJ |
description | ABSTRACTAspect-based sentiment analysis (ABSA) has attracted many researchers' attention in recent years. However, the lack of benchmark datasets for specific languages is a common challenge because of the prohibitive cost of manual annotation. The zero-shot cross-lingual strategy can be applied to solve this gap in research. Moreover, previous works mainly focus on improving the performance of supervised ABSA with pre-trained languages. Therefore, there are few to no systematic comparisons of the benefits of multilingual models in zero-shot and joint training cross-lingual for the ABSA task. In this paper, we focus on the zero-shot and joint training cross-lingual transfer task for the ABSA. We fine-tune the latest pre-trained multilingual language models on the source language, and then it is directly predicted in the target language. For the joint learning scenario, the models are trained on the combination of multiple source languages. Our experimental results show that (1) fine-tuning multilingual models achieve promising performances in the zero-shot cross-lingual scenario; (2) fine-tuning models on the combination training data of multiple source languages outperforms monolingual data in the joint training scenario. Furthermore, the experimental results indicated that choosing other languages instead of English as the source language can give promising results in the low-resource languages scenario. |
first_indexed | 2024-04-09T15:25:42Z |
format | Article |
id | doaj.art-47c8fab5e56648948a68813f7d56036f |
institution | Directory Open Access Journal |
issn | 2475-1839 2475-1847 |
language | English |
last_indexed | 2024-04-09T15:25:42Z |
publishDate | 2023-04-01 |
publisher | Taylor & Francis Group |
record_format | Article |
series | Journal of Information and Telecommunication |
spelling | doaj.art-47c8fab5e56648948a68813f7d56036f2023-04-28T18:16:02ZengTaylor & Francis GroupJournal of Information and Telecommunication2475-18392475-18472023-04-017212114310.1080/24751839.2023.2173843Exploring zero-shot and joint training cross-lingual strategies for aspect-based sentiment analysis based on contextualized multilingual language modelsDang Van Thin0Hung Quoc Ngo1Duong Ngoc Hao2Ngan Luu-Thuy Nguyen3Multimedia Communications Laboratory, University of Information Technology, Ho Chi Minh city, VietnamSchool of Business Technology, Retail, and Supply Chain, Technological University Dublin, Dublin, IrelandMultimedia Communications Laboratory, University of Information Technology, Ho Chi Minh city, VietnamMultimedia Communications Laboratory, University of Information Technology, Ho Chi Minh city, VietnamABSTRACTAspect-based sentiment analysis (ABSA) has attracted many researchers' attention in recent years. However, the lack of benchmark datasets for specific languages is a common challenge because of the prohibitive cost of manual annotation. The zero-shot cross-lingual strategy can be applied to solve this gap in research. Moreover, previous works mainly focus on improving the performance of supervised ABSA with pre-trained languages. Therefore, there are few to no systematic comparisons of the benefits of multilingual models in zero-shot and joint training cross-lingual for the ABSA task. In this paper, we focus on the zero-shot and joint training cross-lingual transfer task for the ABSA. We fine-tune the latest pre-trained multilingual language models on the source language, and then it is directly predicted in the target language. For the joint learning scenario, the models are trained on the combination of multiple source languages. Our experimental results show that (1) fine-tuning multilingual models achieve promising performances in the zero-shot cross-lingual scenario; (2) fine-tuning models on the combination training data of multiple source languages outperforms monolingual data in the joint training scenario. Furthermore, the experimental results indicated that choosing other languages instead of English as the source language can give promising results in the low-resource languages scenario.https://www.tandfonline.com/doi/10.1080/24751839.2023.2173843Aspect-based sentiment analysiszero-shot cross-lingualmultilingual modelsjoint learning |
spellingShingle | Dang Van Thin Hung Quoc Ngo Duong Ngoc Hao Ngan Luu-Thuy Nguyen Exploring zero-shot and joint training cross-lingual strategies for aspect-based sentiment analysis based on contextualized multilingual language models Journal of Information and Telecommunication Aspect-based sentiment analysis zero-shot cross-lingual multilingual models joint learning |
title | Exploring zero-shot and joint training cross-lingual strategies for aspect-based sentiment analysis based on contextualized multilingual language models |
title_full | Exploring zero-shot and joint training cross-lingual strategies for aspect-based sentiment analysis based on contextualized multilingual language models |
title_fullStr | Exploring zero-shot and joint training cross-lingual strategies for aspect-based sentiment analysis based on contextualized multilingual language models |
title_full_unstemmed | Exploring zero-shot and joint training cross-lingual strategies for aspect-based sentiment analysis based on contextualized multilingual language models |
title_short | Exploring zero-shot and joint training cross-lingual strategies for aspect-based sentiment analysis based on contextualized multilingual language models |
title_sort | exploring zero shot and joint training cross lingual strategies for aspect based sentiment analysis based on contextualized multilingual language models |
topic | Aspect-based sentiment analysis zero-shot cross-lingual multilingual models joint learning |
url | https://www.tandfonline.com/doi/10.1080/24751839.2023.2173843 |
work_keys_str_mv | AT dangvanthin exploringzeroshotandjointtrainingcrosslingualstrategiesforaspectbasedsentimentanalysisbasedoncontextualizedmultilinguallanguagemodels AT hungquocngo exploringzeroshotandjointtrainingcrosslingualstrategiesforaspectbasedsentimentanalysisbasedoncontextualizedmultilinguallanguagemodels AT duongngochao exploringzeroshotandjointtrainingcrosslingualstrategiesforaspectbasedsentimentanalysisbasedoncontextualizedmultilinguallanguagemodels AT nganluuthuynguyen exploringzeroshotandjointtrainingcrosslingualstrategiesforaspectbasedsentimentanalysisbasedoncontextualizedmultilinguallanguagemodels |