Jointly Modeling Aspect Information and Ratings for Review Rating Prediction
Although matrix model-based approaches to collaborative filtering (CF), such as latent factor models, achieve good accuracy in review rating prediction, they still face data sparsity problems. Many recent studies have exploited review text information to improve the performance of predictions. The r...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-10-01
|
| Series: | Electronics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2079-9292/11/21/3532 |
| _version_ | 1797468516320804864 |
|---|---|
| author | Qingxi Peng Lan You Hao Feng Wei Du Kesong Zheng Fuxi Zhu Xiaoya Xu |
| author_facet | Qingxi Peng Lan You Hao Feng Wei Du Kesong Zheng Fuxi Zhu Xiaoya Xu |
| author_sort | Qingxi Peng |
| collection | DOAJ |
| description | Although matrix model-based approaches to collaborative filtering (CF), such as latent factor models, achieve good accuracy in review rating prediction, they still face data sparsity problems. Many recent studies have exploited review text information to improve the performance of predictions. The review content that they use, however, is usually on the coarse-grained text level or sentence level. In this paper, we propose a joint model that incorporates review text information with matrix factorization for review rating prediction. First, we adopt an aspect extraction method and propose a simple and practical algorithm to represent the review by aspects and sentiments. Then, we propose two similarity measures: aspect-based user similarity and aspect-based product similarity. Finally, aspect-based user and product similarity measures are incorporated into a matrix factorization to build a joint model for rating prediction. To this end, our model can alleviate the data sparsity problem and obtain interpretability for the recommendation. We conducted experiments on two datasets. The experimental results demonstrate the effectiveness of the proposed model. |
| first_indexed | 2024-03-09T19:08:31Z |
| format | Article |
| id | doaj.art-957a8d15b5ca43e6b338ce6b2b3ade89 |
| institution | Directory Open Access Journal |
| issn | 2079-9292 |
| language | English |
| last_indexed | 2024-03-09T19:08:31Z |
| publishDate | 2022-10-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Electronics |
| spelling | doaj.art-957a8d15b5ca43e6b338ce6b2b3ade892023-11-24T04:25:25ZengMDPI AGElectronics2079-92922022-10-011121353210.3390/electronics11213532Jointly Modeling Aspect Information and Ratings for Review Rating PredictionQingxi Peng0Lan You1Hao Feng2Wei Du3Kesong Zheng4Fuxi Zhu5Xiaoya Xu6School of Information Engineering, Wuhan College, Wuhan 430212, ChinaFaculty of Computer Science and Information Engineering, Hubei University, Wuhan 430062, ChinaSchool of Information Engineering, Wuhan College, Wuhan 430212, ChinaSchool of Information Engineering, Wuhan College, Wuhan 430212, ChinaSchool of Information Engineering, Wuhan College, Wuhan 430212, ChinaSchool of Information Engineering, Wuhan College, Wuhan 430212, ChinaFaculty of Computer Science and Information Engineering, Hubei University, Wuhan 430062, ChinaAlthough matrix model-based approaches to collaborative filtering (CF), such as latent factor models, achieve good accuracy in review rating prediction, they still face data sparsity problems. Many recent studies have exploited review text information to improve the performance of predictions. The review content that they use, however, is usually on the coarse-grained text level or sentence level. In this paper, we propose a joint model that incorporates review text information with matrix factorization for review rating prediction. First, we adopt an aspect extraction method and propose a simple and practical algorithm to represent the review by aspects and sentiments. Then, we propose two similarity measures: aspect-based user similarity and aspect-based product similarity. Finally, aspect-based user and product similarity measures are incorporated into a matrix factorization to build a joint model for rating prediction. To this end, our model can alleviate the data sparsity problem and obtain interpretability for the recommendation. We conducted experiments on two datasets. The experimental results demonstrate the effectiveness of the proposed model.https://www.mdpi.com/2079-9292/11/21/3532rating predictionmatrix factorizationproduct reviewaspect analysis |
| spellingShingle | Qingxi Peng Lan You Hao Feng Wei Du Kesong Zheng Fuxi Zhu Xiaoya Xu Jointly Modeling Aspect Information and Ratings for Review Rating Prediction Electronics rating prediction matrix factorization product review aspect analysis |
| title | Jointly Modeling Aspect Information and Ratings for Review Rating Prediction |
| title_full | Jointly Modeling Aspect Information and Ratings for Review Rating Prediction |
| title_fullStr | Jointly Modeling Aspect Information and Ratings for Review Rating Prediction |
| title_full_unstemmed | Jointly Modeling Aspect Information and Ratings for Review Rating Prediction |
| title_short | Jointly Modeling Aspect Information and Ratings for Review Rating Prediction |
| title_sort | jointly modeling aspect information and ratings for review rating prediction |
| topic | rating prediction matrix factorization product review aspect analysis |
| url | https://www.mdpi.com/2079-9292/11/21/3532 |
| work_keys_str_mv | AT qingxipeng jointlymodelingaspectinformationandratingsforreviewratingprediction AT lanyou jointlymodelingaspectinformationandratingsforreviewratingprediction AT haofeng jointlymodelingaspectinformationandratingsforreviewratingprediction AT weidu jointlymodelingaspectinformationandratingsforreviewratingprediction AT kesongzheng jointlymodelingaspectinformationandratingsforreviewratingprediction AT fuxizhu jointlymodelingaspectinformationandratingsforreviewratingprediction AT xiaoyaxu jointlymodelingaspectinformationandratingsforreviewratingprediction |