Topic Word Embedding-Based Methods for Automatically Extracting Main Aspects from Product Reviews
Detecting the main aspects of a particular product from a collection of review documents is so challenging in real applications. To address this problem, we focus on utilizing existing topic models that can briefly summarize large text documents. Unlike existing approaches that are limited because o...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-05-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/10/11/3831 |
| _version_ | 1797566498390147072 |
|---|---|
| author | Sang-Min Park Sung Joon Lee Byung-Won On |
| author_facet | Sang-Min Park Sung Joon Lee Byung-Won On |
| author_sort | Sang-Min Park |
| collection | DOAJ |
| description | Detecting the main aspects of a particular product from a collection of review documents is so challenging in real applications. To address this problem, we focus on utilizing existing topic models that can briefly summarize large text documents. Unlike existing approaches that are limited because of modifying any topic model or using seed opinion words as prior knowledge, we propose a novel approach of (1) identifying starting points for learning, (2) cleaning dirty topic results through word embedding and unsupervised clustering, and (3) automatically generating right aspects using topic and head word embedding. Experimental results show that the proposed methods create more clean topics, improving about 25% of Rouge–1, compared to the baseline method. In addition, through the proposed three methods, the main aspects suitable for given data are detected automatically. |
| first_indexed | 2024-03-10T19:28:41Z |
| format | Article |
| id | doaj.art-64cef297b5f94e94b40e9d1c610a4373 |
| institution | Directory Open Access Journal |
| issn | 2076-3417 |
| language | English |
| last_indexed | 2024-03-10T19:28:41Z |
| publishDate | 2020-05-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj.art-64cef297b5f94e94b40e9d1c610a43732023-11-20T02:21:53ZengMDPI AGApplied Sciences2076-34172020-05-011011383110.3390/app10113831Topic Word Embedding-Based Methods for Automatically Extracting Main Aspects from Product ReviewsSang-Min Park0Sung Joon Lee1Byung-Won On2AI Labs, Saltlux Inc., Gangnam-gu, Seoul 06147, KoreaDepartment of Software Convergence Engineering, Kunsan National University, Gunsan, Jeollabuk-do 54150, KoreaDepartment of Software Convergence Engineering, Kunsan National University, Gunsan, Jeollabuk-do 54150, KoreaDetecting the main aspects of a particular product from a collection of review documents is so challenging in real applications. To address this problem, we focus on utilizing existing topic models that can briefly summarize large text documents. Unlike existing approaches that are limited because of modifying any topic model or using seed opinion words as prior knowledge, we propose a novel approach of (1) identifying starting points for learning, (2) cleaning dirty topic results through word embedding and unsupervised clustering, and (3) automatically generating right aspects using topic and head word embedding. Experimental results show that the proposed methods create more clean topics, improving about 25% of Rouge–1, compared to the baseline method. In addition, through the proposed three methods, the main aspects suitable for given data are detected automatically.https://www.mdpi.com/2076-3417/10/11/3831word embeddingaspect detectionopinion summarization |
| spellingShingle | Sang-Min Park Sung Joon Lee Byung-Won On Topic Word Embedding-Based Methods for Automatically Extracting Main Aspects from Product Reviews Applied Sciences word embedding aspect detection opinion summarization |
| title | Topic Word Embedding-Based Methods for Automatically Extracting Main Aspects from Product Reviews |
| title_full | Topic Word Embedding-Based Methods for Automatically Extracting Main Aspects from Product Reviews |
| title_fullStr | Topic Word Embedding-Based Methods for Automatically Extracting Main Aspects from Product Reviews |
| title_full_unstemmed | Topic Word Embedding-Based Methods for Automatically Extracting Main Aspects from Product Reviews |
| title_short | Topic Word Embedding-Based Methods for Automatically Extracting Main Aspects from Product Reviews |
| title_sort | topic word embedding based methods for automatically extracting main aspects from product reviews |
| topic | word embedding aspect detection opinion summarization |
| url | https://www.mdpi.com/2076-3417/10/11/3831 |
| work_keys_str_mv | AT sangminpark topicwordembeddingbasedmethodsforautomaticallyextractingmainaspectsfromproductreviews AT sungjoonlee topicwordembeddingbasedmethodsforautomaticallyextractingmainaspectsfromproductreviews AT byungwonon topicwordembeddingbasedmethodsforautomaticallyextractingmainaspectsfromproductreviews |