Weakly Supervised Learning Approach for Implicit Aspect Extraction
Aspect-based sentiment analysis (ABSA) is a process to extract an aspect of a product from a customer review and identify its polarity. Most previous studies of ABSA focused on explicit aspects, but implicit aspects have not yet been the subject of much attention. This paper proposes a novel weakly...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-11-01
|
| Series: | Information |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2078-2489/14/11/612 |
| _version_ | 1797458930120523776 |
|---|---|
| author | Aye Aye Mar Kiyoaki Shirai Natthawut Kertkeidkachorn |
| author_facet | Aye Aye Mar Kiyoaki Shirai Natthawut Kertkeidkachorn |
| author_sort | Aye Aye Mar |
| collection | DOAJ |
| description | Aspect-based sentiment analysis (ABSA) is a process to extract an aspect of a product from a customer review and identify its polarity. Most previous studies of ABSA focused on explicit aspects, but implicit aspects have not yet been the subject of much attention. This paper proposes a novel weakly supervised method for implicit aspect extraction, which is a task to classify a sentence into a pre-defined implicit aspect category. A dataset labeled with implicit aspects is automatically constructed from unlabeled sentences as follows. First, explicit sentences are obtained by extracting explicit aspects from unlabeled sentences, while sentences that do not contain explicit aspects are preserved as candidates of implicit sentences. Second, clustering is performed to merge the explicit and implicit sentences that share the same aspect. Third, the aspect of the explicit sentence is assigned to the implicit sentences in the same cluster as the implicit aspect label. Then, the BERT model is fine-tuned for implicit aspect extraction using the constructed dataset. The results of the experiments show that our method achieves 82% and 84% accuracy for mobile phone and PC reviews, respectively, which are 20 and 21 percentage points higher than the baseline. |
| first_indexed | 2024-03-09T16:44:11Z |
| format | Article |
| id | doaj.art-9cd2cdff7c4d4c43be75879bdea69c73 |
| institution | Directory Open Access Journal |
| issn | 2078-2489 |
| language | English |
| last_indexed | 2024-03-09T16:44:11Z |
| publishDate | 2023-11-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Information |
| spelling | doaj.art-9cd2cdff7c4d4c43be75879bdea69c732023-11-24T14:48:16ZengMDPI AGInformation2078-24892023-11-01141161210.3390/info14110612Weakly Supervised Learning Approach for Implicit Aspect ExtractionAye Aye Mar0Kiyoaki Shirai1Natthawut Kertkeidkachorn2Graduate School of Advanced Science and Technology, Japan Advanced Institute of Science and Technology, 1-1 Asahidai, Nomi 923-1292, Ishikawa, JapanGraduate School of Advanced Science and Technology, Japan Advanced Institute of Science and Technology, 1-1 Asahidai, Nomi 923-1292, Ishikawa, JapanGraduate School of Advanced Science and Technology, Japan Advanced Institute of Science and Technology, 1-1 Asahidai, Nomi 923-1292, Ishikawa, JapanAspect-based sentiment analysis (ABSA) is a process to extract an aspect of a product from a customer review and identify its polarity. Most previous studies of ABSA focused on explicit aspects, but implicit aspects have not yet been the subject of much attention. This paper proposes a novel weakly supervised method for implicit aspect extraction, which is a task to classify a sentence into a pre-defined implicit aspect category. A dataset labeled with implicit aspects is automatically constructed from unlabeled sentences as follows. First, explicit sentences are obtained by extracting explicit aspects from unlabeled sentences, while sentences that do not contain explicit aspects are preserved as candidates of implicit sentences. Second, clustering is performed to merge the explicit and implicit sentences that share the same aspect. Third, the aspect of the explicit sentence is assigned to the implicit sentences in the same cluster as the implicit aspect label. Then, the BERT model is fine-tuned for implicit aspect extraction using the constructed dataset. The results of the experiments show that our method achieves 82% and 84% accuracy for mobile phone and PC reviews, respectively, which are 20 and 21 percentage points higher than the baseline.https://www.mdpi.com/2078-2489/14/11/612aspect-based sentiment analysisaspect extractionimplicit aspectweakly supervised learning |
| spellingShingle | Aye Aye Mar Kiyoaki Shirai Natthawut Kertkeidkachorn Weakly Supervised Learning Approach for Implicit Aspect Extraction Information aspect-based sentiment analysis aspect extraction implicit aspect weakly supervised learning |
| title | Weakly Supervised Learning Approach for Implicit Aspect Extraction |
| title_full | Weakly Supervised Learning Approach for Implicit Aspect Extraction |
| title_fullStr | Weakly Supervised Learning Approach for Implicit Aspect Extraction |
| title_full_unstemmed | Weakly Supervised Learning Approach for Implicit Aspect Extraction |
| title_short | Weakly Supervised Learning Approach for Implicit Aspect Extraction |
| title_sort | weakly supervised learning approach for implicit aspect extraction |
| topic | aspect-based sentiment analysis aspect extraction implicit aspect weakly supervised learning |
| url | https://www.mdpi.com/2078-2489/14/11/612 |
| work_keys_str_mv | AT ayeayemar weaklysupervisedlearningapproachforimplicitaspectextraction AT kiyoakishirai weaklysupervisedlearningapproachforimplicitaspectextraction AT natthawutkertkeidkachorn weaklysupervisedlearningapproachforimplicitaspectextraction |