TABAS: Text augmentation based on attention score for text classification model
To improve the performance of text classification, we propose text augmentation based on attention score (TABAS). We recognized that a criterion for selecting a replacement word rather than a random selection was necessary. Therefore, TABAS utilizes attention scores for text modification, processing...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Elsevier
2022-12-01
|
Series: | ICT Express |
Subjects: | |
Online Access: | http://www.sciencedirect.com/science/article/pii/S2405959521001454 |
_version_ | 1797986963864682496 |
---|---|
author | Yeong Jae Yu Seung Joo Yoon So Young Jun Jong Woo Kim |
author_facet | Yeong Jae Yu Seung Joo Yoon So Young Jun Jong Woo Kim |
author_sort | Yeong Jae Yu |
collection | DOAJ |
description | To improve the performance of text classification, we propose text augmentation based on attention score (TABAS). We recognized that a criterion for selecting a replacement word rather than a random selection was necessary. Therefore, TABAS utilizes attention scores for text modification, processing only words with the same entity and part-of-speech tags to consider informational aspects. To verify this approach, we used two benchmark tasks. As a result, TABAS can significantly improve performance, both recurrent and convolutional neural networks. Furthermore, we confirm that it provides a practical way to develop deep-learning models by saving costs on making additional datasets. |
first_indexed | 2024-04-11T07:41:19Z |
format | Article |
id | doaj.art-fd5c620fa53646e6b48711b4eec7d38c |
institution | Directory Open Access Journal |
issn | 2405-9595 |
language | English |
last_indexed | 2024-04-11T07:41:19Z |
publishDate | 2022-12-01 |
publisher | Elsevier |
record_format | Article |
series | ICT Express |
spelling | doaj.art-fd5c620fa53646e6b48711b4eec7d38c2022-12-22T04:36:35ZengElsevierICT Express2405-95952022-12-0184549554TABAS: Text augmentation based on attention score for text classification modelYeong Jae Yu0Seung Joo Yoon1So Young Jun2Jong Woo Kim3School of Business, Hanyang University, Seoul, Republic of KoreaDepartment of Business Informatics, Hanyang University, Seoul, Republic of KoreaDepartment of Business Informatics, Hanyang University, Seoul, Republic of KoreaSchool of Business, Hanyang University, Seoul, Republic of Korea; Correspondence to: School of Business, Hanyang University, 222 Wangsimni-ro, Seongdong-gu, Seoul 04763, Republic of Korea.To improve the performance of text classification, we propose text augmentation based on attention score (TABAS). We recognized that a criterion for selecting a replacement word rather than a random selection was necessary. Therefore, TABAS utilizes attention scores for text modification, processing only words with the same entity and part-of-speech tags to consider informational aspects. To verify this approach, we used two benchmark tasks. As a result, TABAS can significantly improve performance, both recurrent and convolutional neural networks. Furthermore, we confirm that it provides a practical way to develop deep-learning models by saving costs on making additional datasets.http://www.sciencedirect.com/science/article/pii/S2405959521001454Attention mechanismData augmentationNatural language processingText classification |
spellingShingle | Yeong Jae Yu Seung Joo Yoon So Young Jun Jong Woo Kim TABAS: Text augmentation based on attention score for text classification model ICT Express Attention mechanism Data augmentation Natural language processing Text classification |
title | TABAS: Text augmentation based on attention score for text classification model |
title_full | TABAS: Text augmentation based on attention score for text classification model |
title_fullStr | TABAS: Text augmentation based on attention score for text classification model |
title_full_unstemmed | TABAS: Text augmentation based on attention score for text classification model |
title_short | TABAS: Text augmentation based on attention score for text classification model |
title_sort | tabas text augmentation based on attention score for text classification model |
topic | Attention mechanism Data augmentation Natural language processing Text classification |
url | http://www.sciencedirect.com/science/article/pii/S2405959521001454 |
work_keys_str_mv | AT yeongjaeyu tabastextaugmentationbasedonattentionscorefortextclassificationmodel AT seungjooyoon tabastextaugmentationbasedonattentionscorefortextclassificationmodel AT soyoungjun tabastextaugmentationbasedonattentionscorefortextclassificationmodel AT jongwookim tabastextaugmentationbasedonattentionscorefortextclassificationmodel |