Learning multi-grained aspect target sequence for Chinese sentiment analysis
Aspect-based sentiment analysis aims at identifying sentiment polarity towards aspect targets in a sentence. Previously, the task was modeled as a sentence-level sentiment classification problem that treated aspect targets as a hint. Such approaches oversimplify the problem by averaging word embeddi...
| Main Authors: | , , , |
|---|---|
| Other Authors: | |
| Format: | Journal Article |
| Language: | English |
| Published: |
2020
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/139596 |
| _version_ | 1811685765707464704 |
|---|---|
| author | Peng, Haiyun Ma, Yukun Li, Yang Cambria, Erik |
| author2 | School of Computer Science and Engineering |
| author_facet | School of Computer Science and Engineering Peng, Haiyun Ma, Yukun Li, Yang Cambria, Erik |
| author_sort | Peng, Haiyun |
| collection | NTU |
| description | Aspect-based sentiment analysis aims at identifying sentiment polarity towards aspect targets in a sentence. Previously, the task was modeled as a sentence-level sentiment classification problem that treated aspect targets as a hint. Such approaches oversimplify the problem by averaging word embeddings when the aspect target is a multi-word sequence. In this paper, we formalize the problem from a different perspective, i.e., that sentiment at aspect target level should be the main focus. Due to the fact that written Chinese is very rich and complex, Chinese aspect targets can be studied at three different levels of granularity: radical, character and word. Thus, we propose to explicitly model the aspect target and conduct sentiment classification directly at the aspect target level via three granularities. Moreover, we study two fusion methods for such granularities in the task of Chinese aspect-level sentiment analysis. Experimental results on a multi-word aspect target subset from SemEval2014 and four Chinese review datasets validate our claims and show the improved performance of our model over the state of the art. |
| first_indexed | 2024-10-01T04:49:44Z |
| format | Journal Article |
| id | ntu-10356/139596 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2024-10-01T04:49:44Z |
| publishDate | 2020 |
| record_format | dspace |
| spelling | ntu-10356/1395962020-05-20T07:36:05Z Learning multi-grained aspect target sequence for Chinese sentiment analysis Peng, Haiyun Ma, Yukun Li, Yang Cambria, Erik School of Computer Science and Engineering Engineering::Computer science and engineering Aspect-based Sentiment Analysis Chinese NLP Aspect-based sentiment analysis aims at identifying sentiment polarity towards aspect targets in a sentence. Previously, the task was modeled as a sentence-level sentiment classification problem that treated aspect targets as a hint. Such approaches oversimplify the problem by averaging word embeddings when the aspect target is a multi-word sequence. In this paper, we formalize the problem from a different perspective, i.e., that sentiment at aspect target level should be the main focus. Due to the fact that written Chinese is very rich and complex, Chinese aspect targets can be studied at three different levels of granularity: radical, character and word. Thus, we propose to explicitly model the aspect target and conduct sentiment classification directly at the aspect target level via three granularities. Moreover, we study two fusion methods for such granularities in the task of Chinese aspect-level sentiment analysis. Experimental results on a multi-word aspect target subset from SemEval2014 and four Chinese review datasets validate our claims and show the improved performance of our model over the state of the art. 2020-05-20T07:36:05Z 2020-05-20T07:36:05Z 2018 Journal Article Peng, H., Ma, Y., Li, Y., & Cambria, E. (2018). Learning multi-grained aspect target sequence for Chinese sentiment analysis. Knowledge-Based Systems, 148, 167-176. doi:10.1016/j.knosys.2018.02.034 0950-7051 https://hdl.handle.net/10356/139596 10.1016/j.knosys.2018.02.034 2-s2.0-85042932426 148 167 176 en Knowledge-Based Systems © 2018 Elsevier B.V. All rights reserved. |
| spellingShingle | Engineering::Computer science and engineering Aspect-based Sentiment Analysis Chinese NLP Peng, Haiyun Ma, Yukun Li, Yang Cambria, Erik Learning multi-grained aspect target sequence for Chinese sentiment analysis |
| title | Learning multi-grained aspect target sequence for Chinese sentiment analysis |
| title_full | Learning multi-grained aspect target sequence for Chinese sentiment analysis |
| title_fullStr | Learning multi-grained aspect target sequence for Chinese sentiment analysis |
| title_full_unstemmed | Learning multi-grained aspect target sequence for Chinese sentiment analysis |
| title_short | Learning multi-grained aspect target sequence for Chinese sentiment analysis |
| title_sort | learning multi grained aspect target sequence for chinese sentiment analysis |
| topic | Engineering::Computer science and engineering Aspect-based Sentiment Analysis Chinese NLP |
| url | https://hdl.handle.net/10356/139596 |
| work_keys_str_mv | AT penghaiyun learningmultigrainedaspecttargetsequenceforchinesesentimentanalysis AT mayukun learningmultigrainedaspecttargetsequenceforchinesesentimentanalysis AT liyang learningmultigrainedaspecttargetsequenceforchinesesentimentanalysis AT cambriaerik learningmultigrainedaspecttargetsequenceforchinesesentimentanalysis |