Attention-Based Personalized Compatibility Learning for Fashion Matching
The fashion industry has a critical need for fashion compatibility. Modeling compatibility is a challenging task that involves extracting (in)compatible features of pairs, obtaining compatible relationships between matching items, and applying them to personalized recommendation tasks. Measuring com...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-08-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/13/17/9638 |
| _version_ | 1827728327719256064 |
|---|---|
| author | Xiaozhe Nie Zhijie Xu Jianqin Zhang Yu Tian |
| author_facet | Xiaozhe Nie Zhijie Xu Jianqin Zhang Yu Tian |
| author_sort | Xiaozhe Nie |
| collection | DOAJ |
| description | The fashion industry has a critical need for fashion compatibility. Modeling compatibility is a challenging task that involves extracting (in)compatible features of pairs, obtaining compatible relationships between matching items, and applying them to personalized recommendation tasks. Measuring compatibility is a complex and subjective concept in general. The complexity is reflected in the fact that relationships between fashion items are determined by multiple matching rules, such as color, shape, and material. Each personal aesthetic style and fashion preference differs, adding subjectivity to the compatibility concept. As a result, personalized factors must be considered. Previous works mainly utilize a convolutional neural network to measure compatibility by extracting general features, but they ignore fine-grained compatibility features and only model overall compatibility. We propose a novel neural network framework called the Attention-based Personalized Compatibility Embedding Network (PCE-Net). It comprises two components: attention-based compatibility embedding modeling and attention-based personal preference modeling. In the second part, we utilize matrix factorization and content-based features to obtain user preferences. Both pieces are jointly trained using the BPR framework in an end-to-end method. Extensive experiments on the IQON3000 dataset demonstrate that PCE-Net significantly outperforms most baseline methods. |
| first_indexed | 2024-03-10T23:27:41Z |
| format | Article |
| id | doaj.art-67390d590af144f69aa7de91fc5164a7 |
| institution | Directory Open Access Journal |
| issn | 2076-3417 |
| language | English |
| last_indexed | 2024-03-10T23:27:41Z |
| publishDate | 2023-08-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj.art-67390d590af144f69aa7de91fc5164a72023-11-19T07:49:47ZengMDPI AGApplied Sciences2076-34172023-08-011317963810.3390/app13179638Attention-Based Personalized Compatibility Learning for Fashion MatchingXiaozhe Nie0Zhijie Xu1Jianqin Zhang2Yu Tian3School of Science, Beijing University of Civil Engineering and Architecture, Beijing 102616, ChinaSchool of Science, Beijing University of Civil Engineering and Architecture, Beijing 102616, ChinaSchool of Geomatics and Urban Spatial Informatics, Beijing University of Civil Engineering and Architecture, Beijing 102616, ChinaSchool of Science, Beijing University of Civil Engineering and Architecture, Beijing 102616, ChinaThe fashion industry has a critical need for fashion compatibility. Modeling compatibility is a challenging task that involves extracting (in)compatible features of pairs, obtaining compatible relationships between matching items, and applying them to personalized recommendation tasks. Measuring compatibility is a complex and subjective concept in general. The complexity is reflected in the fact that relationships between fashion items are determined by multiple matching rules, such as color, shape, and material. Each personal aesthetic style and fashion preference differs, adding subjectivity to the compatibility concept. As a result, personalized factors must be considered. Previous works mainly utilize a convolutional neural network to measure compatibility by extracting general features, but they ignore fine-grained compatibility features and only model overall compatibility. We propose a novel neural network framework called the Attention-based Personalized Compatibility Embedding Network (PCE-Net). It comprises two components: attention-based compatibility embedding modeling and attention-based personal preference modeling. In the second part, we utilize matrix factorization and content-based features to obtain user preferences. Both pieces are jointly trained using the BPR framework in an end-to-end method. Extensive experiments on the IQON3000 dataset demonstrate that PCE-Net significantly outperforms most baseline methods.https://www.mdpi.com/2076-3417/13/17/9638fashion analysispersonalized compatibility embedding modelingattention mechanismmulti-modal |
| spellingShingle | Xiaozhe Nie Zhijie Xu Jianqin Zhang Yu Tian Attention-Based Personalized Compatibility Learning for Fashion Matching Applied Sciences fashion analysis personalized compatibility embedding modeling attention mechanism multi-modal |
| title | Attention-Based Personalized Compatibility Learning for Fashion Matching |
| title_full | Attention-Based Personalized Compatibility Learning for Fashion Matching |
| title_fullStr | Attention-Based Personalized Compatibility Learning for Fashion Matching |
| title_full_unstemmed | Attention-Based Personalized Compatibility Learning for Fashion Matching |
| title_short | Attention-Based Personalized Compatibility Learning for Fashion Matching |
| title_sort | attention based personalized compatibility learning for fashion matching |
| topic | fashion analysis personalized compatibility embedding modeling attention mechanism multi-modal |
| url | https://www.mdpi.com/2076-3417/13/17/9638 |
| work_keys_str_mv | AT xiaozhenie attentionbasedpersonalizedcompatibilitylearningforfashionmatching AT zhijiexu attentionbasedpersonalizedcompatibilitylearningforfashionmatching AT jianqinzhang attentionbasedpersonalizedcompatibilitylearningforfashionmatching AT yutian attentionbasedpersonalizedcompatibilitylearningforfashionmatching |