Content-based video relevance prediction with multi-view multi-level deep interest network

Zeyuan Chen; Kai Xu; Wei Zhang

doi:10.1145/3343031.3356068

Content-based video relevance prediction with multi-view multi-level deep interest network

Zeyuan Chen^*
, Kai Xu
, Wei Zhang

^*Corresponding author for this work

School of Computer Science and Technology

East China Normal University

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

2 Scopus citations

Abstract

This paper presents our solution for the Hulu Content-Based Video Relevance Prediction (CBVRP) challenge, which focuses on cold-start videos as candidates. The keys to success of this prediction scenario are to learn effective user and video representations. To this end, we develop a multi-view multi-level deep interest network (MMDIN), which involves a multi-level deep interest network to learn user and video representations in a single-view, and a late fusion technique to integrate their multi-view representations corresponding to different types of video features. Through the above manner, the cold-start video prediction could be handled well with representations through their past interaction behaviors with videos and video representations based on their multiple types of content profiles.

Original language	English
Title of host publication	MM 2019 - Proceedings of the 27th ACM International Conference on Multimedia
Publisher	Association for Computing Machinery, Inc
Pages	2607-2611
Number of pages	5
ISBN (Electronic)	9781450368896
DOIs	https://doi.org/10.1145/3343031.3356068
State	Published - 15 Oct 2019
Event	27th ACM International Conference on Multimedia, MM 2019 - Nice, France Duration: 21 Oct 2019 → 25 Oct 2019

Publication series

Name	MM 2019 - Proceedings of the 27th ACM International Conference on Multimedia

Conference

Conference	27th ACM International Conference on Multimedia, MM 2019
Country/Territory	France
City	Nice
Period	21/10/19 → 25/10/19

Keywords

Content-based video recommendation
Deep learning
Sequence modeling

Access to Document

10.1145/3343031.3356068

Cite this

Chen, Z., Xu, K., & Zhang, W. (2019). Content-based video relevance prediction with multi-view multi-level deep interest network. In MM 2019 - Proceedings of the 27th ACM International Conference on Multimedia (pp. 2607-2611). (MM 2019 - Proceedings of the 27th ACM International Conference on Multimedia). Association for Computing Machinery, Inc. https://doi.org/10.1145/3343031.3356068

@inproceedings{c1cdde4492524dbba640fe2d1cb3581f,

title = "Content-based video relevance prediction with multi-view multi-level deep interest network",

abstract = "This paper presents our solution for the Hulu Content-Based Video Relevance Prediction (CBVRP) challenge, which focuses on cold-start videos as candidates. The keys to success of this prediction scenario are to learn effective user and video representations. To this end, we develop a multi-view multi-level deep interest network (MMDIN), which involves a multi-level deep interest network to learn user and video representations in a single-view, and a late fusion technique to integrate their multi-view representations corresponding to different types of video features. Through the above manner, the cold-start video prediction could be handled well with representations through their past interaction behaviors with videos and video representations based on their multiple types of content profiles.",

keywords = "Content-based video recommendation, Deep learning, Sequence modeling",

author = "Zeyuan Chen and Kai Xu and Wei Zhang",

note = "Publisher Copyright: {\textcopyright} 2019 Association for Computing Machinery. ACM ISBN 978-1-4503-6889-6/19/10...\$15.00; 27th ACM International Conference on Multimedia, MM 2019 ; Conference date: 21-10-2019 Through 25-10-2019",

year = "2019",

month = oct,

day = "15",

doi = "10.1145/3343031.3356068",

language = "英语",

series = "MM 2019 - Proceedings of the 27th ACM International Conference on Multimedia",

publisher = "Association for Computing Machinery, Inc",

pages = "2607--2611",

booktitle = "MM 2019 - Proceedings of the 27th ACM International Conference on Multimedia",

}

Chen, Z, Xu, K & Zhang, W 2019, Content-based video relevance prediction with multi-view multi-level deep interest network. in MM 2019 - Proceedings of the 27th ACM International Conference on Multimedia. MM 2019 - Proceedings of the 27th ACM International Conference on Multimedia, Association for Computing Machinery, Inc, pp. 2607-2611, 27th ACM International Conference on Multimedia, MM 2019, Nice, France, 21/10/19. https://doi.org/10.1145/3343031.3356068

Content-based video relevance prediction with multi-view multi-level deep interest network. / Chen, Zeyuan; Xu, Kai; Zhang, Wei.
MM 2019 - Proceedings of the 27th ACM International Conference on Multimedia. Association for Computing Machinery, Inc, 2019. p. 2607-2611 (MM 2019 - Proceedings of the 27th ACM International Conference on Multimedia).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Content-based video relevance prediction with multi-view multi-level deep interest network

AU - Chen, Zeyuan

AU - Xu, Kai

AU - Zhang, Wei

PY - 2019/10/15

Y1 - 2019/10/15

N2 - This paper presents our solution for the Hulu Content-Based Video Relevance Prediction (CBVRP) challenge, which focuses on cold-start videos as candidates. The keys to success of this prediction scenario are to learn effective user and video representations. To this end, we develop a multi-view multi-level deep interest network (MMDIN), which involves a multi-level deep interest network to learn user and video representations in a single-view, and a late fusion technique to integrate their multi-view representations corresponding to different types of video features. Through the above manner, the cold-start video prediction could be handled well with representations through their past interaction behaviors with videos and video representations based on their multiple types of content profiles.

AB - This paper presents our solution for the Hulu Content-Based Video Relevance Prediction (CBVRP) challenge, which focuses on cold-start videos as candidates. The keys to success of this prediction scenario are to learn effective user and video representations. To this end, we develop a multi-view multi-level deep interest network (MMDIN), which involves a multi-level deep interest network to learn user and video representations in a single-view, and a late fusion technique to integrate their multi-view representations corresponding to different types of video features. Through the above manner, the cold-start video prediction could be handled well with representations through their past interaction behaviors with videos and video representations based on their multiple types of content profiles.

KW - Content-based video recommendation

KW - Deep learning

KW - Sequence modeling

UR - https://www.scopus.com/pages/publications/85074824889

U2 - 10.1145/3343031.3356068

DO - 10.1145/3343031.3356068

M3 - 会议稿件

AN - SCOPUS:85074824889

T3 - MM 2019 - Proceedings of the 27th ACM International Conference on Multimedia

SP - 2607

EP - 2611

BT - MM 2019 - Proceedings of the 27th ACM International Conference on Multimedia

PB - Association for Computing Machinery, Inc

T2 - 27th ACM International Conference on Multimedia, MM 2019

Y2 - 21 October 2019 through 25 October 2019

ER -

Content-based video relevance prediction with multi-view multi-level deep interest network

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this