Please use this identifier to cite or link to this item: https://doi.org/10.1145/3077136.3080797
Title: Attentive Collaborative Filtering: Multimedia Recommendation with Feature- and Item-level Attention
Authors: Jingyuan Chen
Hanwang Zhang 
Xiangnan He 
Liqiang Nie 
Wei Liu
Tat-Seng Chua 
Keywords: Attention
Collaborative Filtering
Implicit Feedback
Multimedia Recommendation
Issue Date: 7-Aug-2017
Publisher: Association for Computing Machinery, Inc
Citation: Jingyuan Chen, Hanwang Zhang, Xiangnan He, Liqiang Nie, Wei Liu, Tat-Seng Chua (2017-08-07). Attentive Collaborative Filtering: Multimedia Recommendation with Feature- and Item-level Attention. ACM SIGIR 2017 : 335-344. ScholarBank@NUS Repository. https://doi.org/10.1145/3077136.3080797
Abstract: Multimedia content is dominating today's Web information.The nature of multimedia user-item interactions is 1/0 binary implicit feedback (e.g., photo likes, video views, song downloads, etc.), which can be collected at a larger scale with a much lower cost than explicit feedback (e.g., product ratings). However, the majority of existing collaborative flltering (CF) systems are not well-designed for multimedia recommendation, since they ignore the implicitness in users' interactions with multimedia content. We argue that, in multimedia recommendation, there exists item-And component-level implicitness which blurs the underlying users' preferences. The item-level implicitness means that users' preferences on items (e.g., photos, videos, songs, etc.) are unknown, while the componentlevel implicitness means that inside each item users' preferences on different components (e.g., regions in an image, frames of a video, etc.) are unknown. For example, a "view" on a video does not provide any speci.c information about how the user likes the video (i.e., item-level) and which parts of the video the user is interested in (i.e., component-level). In this paper, we introduce a novel attention mechanism in CF to address the challenging item-And component-level implicit feedback in multimedia recommendation, dubbed A.entive Collaborative Filtering (ACF). Speci.cally, our attention model is a neural network that consists of two attention modules: The component-level attention module, starting from any content feature extraction network (e.g., CNN for images/videos), which learns to select informative components of multimedia items, and the item-level attention module, which learns to score the item preferences. ACF can be seamlessly incorporated into classic CF models with implicit feedback, such as BPR and SVD++, and effciently trained using SGD. Through extensive experiments on two real-world multimedia Web services: Vine and Pinterest, we show that ACF significantly outperforms state-of-The-Art CF methods. © 2017 Copyright held by the owner/author(s).
Source Title: ACM SIGIR 2017
URI: https://scholarbank.nus.edu.sg/handle/10635/167394
ISBN: 9781450350228
DOI: 10.1145/3077136.3080797
Appears in Collections:Staff Publications
Elements

Show full item record
Files in This Item:
File Description SizeFormatAccess SettingsVersion 
Attentive Collaborative Filtering- Multimedia Recommendation with Feature- and Item-level Attention.pdf3.11 MBAdobe PDF

OPEN

NoneView/Download

SCOPUSTM   
Citations

374
checked on Jul 29, 2021

Page view(s)

126
checked on Jul 30, 2021

Download(s)

1
checked on Jul 30, 2021

Google ScholarTM

Check

Altmetric


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.