MiMuSA—mimicking human language understanding for fine-grained multi-class sentiment analysis
Sentiment analysis is an important natural language processing (NLP) task due to a wide range of applications. Most existing sentiment analysis techniques are limited to the analysis carried out at the aggregate level, merely providing negative, neutral and positive sentiments. The latest deep learn...
Saved in:
Main Authors: | , , , , |
---|---|
Other Authors: | |
Format: | Article |
Language: | English |
Published: |
2024
|
Subjects: | |
Online Access: | https://hdl.handle.net/10356/172942 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Institution: | Nanyang Technological University |
Language: | English |
id |
sg-ntu-dr.10356-172942 |
---|---|
record_format |
dspace |
spelling |
sg-ntu-dr.10356-1729422024-01-03T05:00:46Z MiMuSA—mimicking human language understanding for fine-grained multi-class sentiment analysis Wang, Zhaoxia Hu, Zhenda Ho, Seng-Beng Cambria, Erik Tan, Ah-Hwee School of Computer Science and Engineering Engineering::Computer science and engineering Human-Like Understanding Fine-Grained Sentiment Understanding Sentiment analysis is an important natural language processing (NLP) task due to a wide range of applications. Most existing sentiment analysis techniques are limited to the analysis carried out at the aggregate level, merely providing negative, neutral and positive sentiments. The latest deep learning-based methods have been leveraged to provide more than three sentiment classes. However, such learning-based methods are still black-box-based methods rather than explainable language processing methods. To address this gap, this paper proposes a new explainable fine-grained multi-class sentiment analysis method, namely MiMuSA, which mimics the human language understanding processes. The proposed method involves a multi-level modular structure designed to mimic human’s language understanding processes, e.g., ambivalence handling process, sentiment strength handling process, etc. Specifically, multiple knowledge bases including Basic Knowledge Base, Negation and Special Knowledge Base, Sarcasm Rule and Adversative Knowledge Base, and Sentiment Strength Knowledge Base are built to support the sentiment understanding process. Compared with other multi-class sentiment analysis methods, this method not only identifies positive or negative sentiments, but can also understand fine-grained multi-class sentiments, such as the degree of positivity (e.g., strongly positive or slightly positive) and the degree of negativity (e.g., slightly negative or strongly negative) of the sentiments involved. The experimental results demonstrate that the proposed MiMuSA outperforms other existing multi-class sentiment analysis methods in terms of accuracy and F1-Score. Agency for Science, Technology and Research (A*STAR) s This research was supported by A*STAR under its Advanced Manufacturing and Engineering (AME) Programmatic Grant (Award No.: A19E2b0098). 2024-01-03T05:00:46Z 2024-01-03T05:00:46Z 2023 Journal Article Wang, Z., Hu, Z., Ho, S., Cambria, E. & Tan, A. (2023). MiMuSA—mimicking human language understanding for fine-grained multi-class sentiment analysis. Neural Computing and Applications, 35(21), 15907-15921. https://dx.doi.org/10.1007/s00521-023-08576-z 0941-0643 https://hdl.handle.net/10356/172942 10.1007/s00521-023-08576-z 2-s2.0-85153059818 21 35 15907 15921 en A19E2b0098 Neural Computing and Applications © 2023 The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature. All rights reserved. |
institution |
Nanyang Technological University |
building |
NTU Library |
continent |
Asia |
country |
Singapore Singapore |
content_provider |
NTU Library |
collection |
DR-NTU |
language |
English |
topic |
Engineering::Computer science and engineering Human-Like Understanding Fine-Grained Sentiment Understanding |
spellingShingle |
Engineering::Computer science and engineering Human-Like Understanding Fine-Grained Sentiment Understanding Wang, Zhaoxia Hu, Zhenda Ho, Seng-Beng Cambria, Erik Tan, Ah-Hwee MiMuSA—mimicking human language understanding for fine-grained multi-class sentiment analysis |
description |
Sentiment analysis is an important natural language processing (NLP) task due to a wide range of applications. Most existing sentiment analysis techniques are limited to the analysis carried out at the aggregate level, merely providing negative, neutral and positive sentiments. The latest deep learning-based methods have been leveraged to provide more than three sentiment classes. However, such learning-based methods are still black-box-based methods rather than explainable language processing methods. To address this gap, this paper proposes a new explainable fine-grained multi-class sentiment analysis method, namely MiMuSA, which mimics the human language understanding processes. The proposed method involves a multi-level modular structure designed to mimic human’s language understanding processes, e.g., ambivalence handling process, sentiment strength handling process, etc. Specifically, multiple knowledge bases including Basic Knowledge Base, Negation and Special Knowledge Base, Sarcasm Rule and Adversative Knowledge Base, and Sentiment Strength Knowledge Base are built to support the sentiment understanding process. Compared with other multi-class sentiment analysis methods, this method not only identifies positive or negative sentiments, but can also understand fine-grained multi-class sentiments, such as the degree of positivity (e.g., strongly positive or slightly positive) and the degree of negativity (e.g., slightly negative or strongly negative) of the sentiments involved. The experimental results demonstrate that the proposed MiMuSA outperforms other existing multi-class sentiment analysis methods in terms of accuracy and F1-Score. |
author2 |
School of Computer Science and Engineering |
author_facet |
School of Computer Science and Engineering Wang, Zhaoxia Hu, Zhenda Ho, Seng-Beng Cambria, Erik Tan, Ah-Hwee |
format |
Article |
author |
Wang, Zhaoxia Hu, Zhenda Ho, Seng-Beng Cambria, Erik Tan, Ah-Hwee |
author_sort |
Wang, Zhaoxia |
title |
MiMuSA—mimicking human language understanding for fine-grained multi-class sentiment analysis |
title_short |
MiMuSA—mimicking human language understanding for fine-grained multi-class sentiment analysis |
title_full |
MiMuSA—mimicking human language understanding for fine-grained multi-class sentiment analysis |
title_fullStr |
MiMuSA—mimicking human language understanding for fine-grained multi-class sentiment analysis |
title_full_unstemmed |
MiMuSA—mimicking human language understanding for fine-grained multi-class sentiment analysis |
title_sort |
mimusa—mimicking human language understanding for fine-grained multi-class sentiment analysis |
publishDate |
2024 |
url |
https://hdl.handle.net/10356/172942 |
_version_ |
1787590730130128896 |