MNI: An enhanced multi-task neighborhood interaction model for recommendation on knowledge graph
To alleviate the data sparsity and cold start problems for collaborative filtering in recommendation systems, side information is usually leveraged by researchers to improve the recommendation performance. The utility of knowledge graph regards the side information as part of the graph structure and...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Public Library of Science (PLoS)
2021-01-01
|
| Series: | PLoS ONE |
| Online Access: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8553089/?tool=EBI |
| _version_ | 1818718943199625216 |
|---|---|
| author | Xintao Ma Liyan Dong Yuequn Wang Yongli Li Hao Zhang |
| author_facet | Xintao Ma Liyan Dong Yuequn Wang Yongli Li Hao Zhang |
| author_sort | Xintao Ma |
| collection | DOAJ |
| description | To alleviate the data sparsity and cold start problems for collaborative filtering in recommendation systems, side information is usually leveraged by researchers to improve the recommendation performance. The utility of knowledge graph regards the side information as part of the graph structure and gives an explanation for recommendation results. In this paper, we propose an enhanced multi-task neighborhood interaction (MNI) model for recommendation on knowledge graphs. MNI explores not only the user-item interaction but also the neighbor-neighbor interactions, capturing a more sophisticated local structure. Besides, the entities and relations are also semantically embedded. And with the cross&compress unit, items in the recommendation system and entities in the knowledge graph can share latent features, and thus high-order interactions can be investigated. Through extensive experiments on real-world datasets, we demonstrate that MNI outperforms some of the state-of-the-art baselines both for CTR prediction and top-N recommendation. |
| first_indexed | 2024-12-17T19:59:04Z |
| format | Article |
| id | doaj.art-d6c77c0f53d2478db014905a0c8693f7 |
| institution | Directory Open Access Journal |
| issn | 1932-6203 |
| language | English |
| last_indexed | 2024-12-17T19:59:04Z |
| publishDate | 2021-01-01 |
| publisher | Public Library of Science (PLoS) |
| record_format | Article |
| series | PLoS ONE |
| spelling | doaj.art-d6c77c0f53d2478db014905a0c8693f72022-12-21T21:34:31ZengPublic Library of Science (PLoS)PLoS ONE1932-62032021-01-011610MNI: An enhanced multi-task neighborhood interaction model for recommendation on knowledge graphXintao MaLiyan DongYuequn WangYongli LiHao ZhangTo alleviate the data sparsity and cold start problems for collaborative filtering in recommendation systems, side information is usually leveraged by researchers to improve the recommendation performance. The utility of knowledge graph regards the side information as part of the graph structure and gives an explanation for recommendation results. In this paper, we propose an enhanced multi-task neighborhood interaction (MNI) model for recommendation on knowledge graphs. MNI explores not only the user-item interaction but also the neighbor-neighbor interactions, capturing a more sophisticated local structure. Besides, the entities and relations are also semantically embedded. And with the cross&compress unit, items in the recommendation system and entities in the knowledge graph can share latent features, and thus high-order interactions can be investigated. Through extensive experiments on real-world datasets, we demonstrate that MNI outperforms some of the state-of-the-art baselines both for CTR prediction and top-N recommendation.https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8553089/?tool=EBI |
| spellingShingle | Xintao Ma Liyan Dong Yuequn Wang Yongli Li Hao Zhang MNI: An enhanced multi-task neighborhood interaction model for recommendation on knowledge graph PLoS ONE |
| title | MNI: An enhanced multi-task neighborhood interaction model for recommendation on knowledge graph |
| title_full | MNI: An enhanced multi-task neighborhood interaction model for recommendation on knowledge graph |
| title_fullStr | MNI: An enhanced multi-task neighborhood interaction model for recommendation on knowledge graph |
| title_full_unstemmed | MNI: An enhanced multi-task neighborhood interaction model for recommendation on knowledge graph |
| title_short | MNI: An enhanced multi-task neighborhood interaction model for recommendation on knowledge graph |
| title_sort | mni an enhanced multi task neighborhood interaction model for recommendation on knowledge graph |
| url | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8553089/?tool=EBI |
| work_keys_str_mv | AT xintaoma mnianenhancedmultitaskneighborhoodinteractionmodelforrecommendationonknowledgegraph AT liyandong mnianenhancedmultitaskneighborhoodinteractionmodelforrecommendationonknowledgegraph AT yuequnwang mnianenhancedmultitaskneighborhoodinteractionmodelforrecommendationonknowledgegraph AT yonglili mnianenhancedmultitaskneighborhoodinteractionmodelforrecommendationonknowledgegraph AT haozhang mnianenhancedmultitaskneighborhoodinteractionmodelforrecommendationonknowledgegraph |