Multi-Level Fine-Grained Interactions for Collaborative Filtering
In recent years, review-based collaborative filtering (CF) has been extensively studied, which is an combination between natural language processing (NLP) and recommender systems. The core pattern behind CF is to first model user and item, and then adopts a relatively primitive interaction between t...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2019-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/8844254/ |
| _version_ | 1818611556000202752 |
|---|---|
| author | Xingjie Feng Yunze Zeng |
| author_facet | Xingjie Feng Yunze Zeng |
| author_sort | Xingjie Feng |
| collection | DOAJ |
| description | In recent years, review-based collaborative filtering (CF) has been extensively studied, which is an combination between natural language processing (NLP) and recommender systems. The core pattern behind CF is to first model user and item, and then adopts a relatively primitive interaction between them for personalized recommendation. This pattern is very similar to the issue of sequence matching in NLP, where sequence 1 and sequence 2 are matched with a fine-grained interaction leading to a better result. Therefore, there is a tremendous room for further improvement in current review-based CF to release the power of fine-grained interaction. To this end, we treat the user review set and item review set as two sequences, and design a multi-level matching attention layer for fine-grained interaction. In addition, we devise the aspect-level and review-level attention to measure the contribution of each review. Extensive experiments on 24 public datasets show that the proposed model consistently outperforms the state-of-the-art approaches. More importantly, by selecting the relevant reviews according to the aspect attention score and review attention score, we can observe which specific item aspects that user mainly concerned and which item characteristic highly matched with the user preference, in which the recommendation interpretability can be enhanced. |
| first_indexed | 2024-12-16T15:32:12Z |
| format | Article |
| id | doaj.art-b0c63cad6d254276a425bf23345af190 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-12-16T15:32:12Z |
| publishDate | 2019-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-b0c63cad6d254276a425bf23345af1902022-12-21T22:26:18ZengIEEEIEEE Access2169-35362019-01-01714316914318410.1109/ACCESS.2019.29412368844254Multi-Level Fine-Grained Interactions for Collaborative FilteringXingjie Feng0Yunze Zeng1https://orcid.org/0000-0003-2081-0505College of Computer Science and Technology, Civil Aviation University of China, Tianjin, ChinaCollege of Computer Science and Technology, Civil Aviation University of China, Tianjin, ChinaIn recent years, review-based collaborative filtering (CF) has been extensively studied, which is an combination between natural language processing (NLP) and recommender systems. The core pattern behind CF is to first model user and item, and then adopts a relatively primitive interaction between them for personalized recommendation. This pattern is very similar to the issue of sequence matching in NLP, where sequence 1 and sequence 2 are matched with a fine-grained interaction leading to a better result. Therefore, there is a tremendous room for further improvement in current review-based CF to release the power of fine-grained interaction. To this end, we treat the user review set and item review set as two sequences, and design a multi-level matching attention layer for fine-grained interaction. In addition, we devise the aspect-level and review-level attention to measure the contribution of each review. Extensive experiments on 24 public datasets show that the proposed model consistently outperforms the state-of-the-art approaches. More importantly, by selecting the relevant reviews according to the aspect attention score and review attention score, we can observe which specific item aspects that user mainly concerned and which item characteristic highly matched with the user preference, in which the recommendation interpretability can be enhanced.https://ieeexplore.ieee.org/document/8844254/Recommender systemcollaborative filteringdeep learningattention mechanism |
| spellingShingle | Xingjie Feng Yunze Zeng Multi-Level Fine-Grained Interactions for Collaborative Filtering IEEE Access Recommender system collaborative filtering deep learning attention mechanism |
| title | Multi-Level Fine-Grained Interactions for Collaborative Filtering |
| title_full | Multi-Level Fine-Grained Interactions for Collaborative Filtering |
| title_fullStr | Multi-Level Fine-Grained Interactions for Collaborative Filtering |
| title_full_unstemmed | Multi-Level Fine-Grained Interactions for Collaborative Filtering |
| title_short | Multi-Level Fine-Grained Interactions for Collaborative Filtering |
| title_sort | multi level fine grained interactions for collaborative filtering |
| topic | Recommender system collaborative filtering deep learning attention mechanism |
| url | https://ieeexplore.ieee.org/document/8844254/ |
| work_keys_str_mv | AT xingjiefeng multilevelfinegrainedinteractionsforcollaborativefiltering AT yunzezeng multilevelfinegrainedinteractionsforcollaborativefiltering |