Attention-based sequential recommendation system using multimodal data
Fuente:
arXiv
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909212119924736 |
|---|---|
| author | Oh, Hyungtaik Jo, Wonkeun Kim, Dongil |
| author_facet | Oh, Hyungtaik Jo, Wonkeun Kim, Dongil |
| contents | Sequential recommendation systems that model dynamic preferences based on a use's past behavior are crucial to e-commerce. Recent studies on these systems have considered various types of information such as images and texts. However, multimodal data have not yet been utilized directly to recommend products to users. In this study, we propose an attention-based sequential recommendation method that employs multimodal data of items such as images, texts, and categories. First, we extract image and text features from pre-trained VGG and BERT and convert categories into multi-labeled forms. Subsequently, attention operations are performed independent of the item sequence and multimodal representations. Finally, the individual attention information is integrated through an attention fusion function. In addition, we apply multitask learning loss for each modality to improve the generalization performance. The experimental results obtained from the Amazon datasets show that the proposed method outperforms those of conventional sequential recommendation systems. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2405_17959 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Attention-based sequential recommendation system using multimodal data Oh, Hyungtaik Jo, Wonkeun Kim, Dongil Information Retrieval Artificial Intelligence I.2.1; I.2.4; I.2.7 Sequential recommendation systems that model dynamic preferences based on a use's past behavior are crucial to e-commerce. Recent studies on these systems have considered various types of information such as images and texts. However, multimodal data have not yet been utilized directly to recommend products to users. In this study, we propose an attention-based sequential recommendation method that employs multimodal data of items such as images, texts, and categories. First, we extract image and text features from pre-trained VGG and BERT and convert categories into multi-labeled forms. Subsequently, attention operations are performed independent of the item sequence and multimodal representations. Finally, the individual attention information is integrated through an attention fusion function. In addition, we apply multitask learning loss for each modality to improve the generalization performance. The experimental results obtained from the Amazon datasets show that the proposed method outperforms those of conventional sequential recommendation systems. |
| title | Attention-based sequential recommendation system using multimodal data |
| topic | Information Retrieval Artificial Intelligence I.2.1; I.2.4; I.2.7 |
| url | https://arxiv.org/abs/2405.17959 |