A Multi-Modality Deep Network for Cold-Start Recommendation
Collaborative filtering (CF) approaches, which provide recommendations based on ratings or purchase history, perform well for users and items with sufficient interactions. However, CF approaches suffer from the cold-start problem for users and items with few ratings. Hybrid recommender systems that...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2018-03-01
|
| Series: | Big Data and Cognitive Computing |
| Subjects: | |
| Online Access: | http://www.mdpi.com/2504-2289/2/1/7 |
| _version_ | 1811198308445585408 |
|---|---|
| author | Mingxuan Sun Fei Li Jian Zhang |
| author_facet | Mingxuan Sun Fei Li Jian Zhang |
| author_sort | Mingxuan Sun |
| collection | DOAJ |
| description | Collaborative filtering (CF) approaches, which provide recommendations based on ratings or purchase history, perform well for users and items with sufficient interactions. However, CF approaches suffer from the cold-start problem for users and items with few ratings. Hybrid recommender systems that combine collaborative filtering and content-based approaches have been proved as an effective way to alleviate the cold-start issue. Integrating contents from multiple heterogeneous data sources such as reviews and product images is challenging for two reasons. Firstly, mapping contents in different modalities from the original feature space to a joint lower-dimensional space is difficult since they have intrinsically different characteristics and statistical properties, such as sparse texts and dense images. Secondly, most algorithms only use content features as the prior knowledge to improve the estimation of user and item profiles but the ratings do not directly provide feedback to guide feature extraction. To tackle these challenges, we propose a tightly-coupled deep network model for fusing heterogeneous modalities, to avoid tedious feature extraction in specific domains, and to enable two-way information propagation from both content and rating information. Experiments on large-scale Amazon product data in book and movie domains demonstrate the effectiveness of the proposed model for cold-start recommendation. |
| first_indexed | 2024-04-12T01:29:14Z |
| format | Article |
| id | doaj.art-1a0022003a7546ca862e05858ece81d0 |
| institution | Directory Open Access Journal |
| issn | 2504-2289 |
| language | English |
| last_indexed | 2024-04-12T01:29:14Z |
| publishDate | 2018-03-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Big Data and Cognitive Computing |
| spelling | doaj.art-1a0022003a7546ca862e05858ece81d02022-12-22T03:53:32ZengMDPI AGBig Data and Cognitive Computing2504-22892018-03-0121710.3390/bdcc2010007bdcc2010007A Multi-Modality Deep Network for Cold-Start RecommendationMingxuan Sun0Fei Li1Jian Zhang2Division of Computer Science and Engineering, Louisiana State University, Baton Rouge, LA 70803, USADivision of Computer Science and Engineering, Louisiana State University, Baton Rouge, LA 70803, USADivision of Computer Science and Engineering, Louisiana State University, Baton Rouge, LA 70803, USACollaborative filtering (CF) approaches, which provide recommendations based on ratings or purchase history, perform well for users and items with sufficient interactions. However, CF approaches suffer from the cold-start problem for users and items with few ratings. Hybrid recommender systems that combine collaborative filtering and content-based approaches have been proved as an effective way to alleviate the cold-start issue. Integrating contents from multiple heterogeneous data sources such as reviews and product images is challenging for two reasons. Firstly, mapping contents in different modalities from the original feature space to a joint lower-dimensional space is difficult since they have intrinsically different characteristics and statistical properties, such as sparse texts and dense images. Secondly, most algorithms only use content features as the prior knowledge to improve the estimation of user and item profiles but the ratings do not directly provide feedback to guide feature extraction. To tackle these challenges, we propose a tightly-coupled deep network model for fusing heterogeneous modalities, to avoid tedious feature extraction in specific domains, and to enable two-way information propagation from both content and rating information. Experiments on large-scale Amazon product data in book and movie domains demonstrate the effectiveness of the proposed model for cold-start recommendation.http://www.mdpi.com/2504-2289/2/1/7recommender systemdeep learningmultimodal learning |
| spellingShingle | Mingxuan Sun Fei Li Jian Zhang A Multi-Modality Deep Network for Cold-Start Recommendation Big Data and Cognitive Computing recommender system deep learning multimodal learning |
| title | A Multi-Modality Deep Network for Cold-Start Recommendation |
| title_full | A Multi-Modality Deep Network for Cold-Start Recommendation |
| title_fullStr | A Multi-Modality Deep Network for Cold-Start Recommendation |
| title_full_unstemmed | A Multi-Modality Deep Network for Cold-Start Recommendation |
| title_short | A Multi-Modality Deep Network for Cold-Start Recommendation |
| title_sort | multi modality deep network for cold start recommendation |
| topic | recommender system deep learning multimodal learning |
| url | http://www.mdpi.com/2504-2289/2/1/7 |
| work_keys_str_mv | AT mingxuansun amultimodalitydeepnetworkforcoldstartrecommendation AT feili amultimodalitydeepnetworkforcoldstartrecommendation AT jianzhang amultimodalitydeepnetworkforcoldstartrecommendation AT mingxuansun multimodalitydeepnetworkforcoldstartrecommendation AT feili multimodalitydeepnetworkforcoldstartrecommendation AT jianzhang multimodalitydeepnetworkforcoldstartrecommendation |