Dual granularity multi-layer attention networks for group event recommendation

Xiaobin Deng

doi:10.1117/12.3017548

15 January 2024 Dual granularity multi-layer attention networks for group event recommendation

Xiaobin Deng

Proceedings Volume 12983, Second International Conference on Electrical, Electronics, and Information Engineering (EEIE 2023); 1298316 (2024) https://doi.org/10.1117/12.3017548
Event: Second International Conference on Electrical, Electronics, and Information Engineering (EEIE 2023), 2023, Wuhan, China

Abstract

Recently, there is a popular type of social network called Event-based Social Networks (EBSN), such as Meetup, Plancast and Douban. In EBSN, users often participate in events as group members. Therefore, studying recommending events to groups is a very meaningful thing. In order to reflect the different weights of different factors and consider the influence of the internal features both of the groups and events, we propose Dual Granularity Multi-layer Attention Networks (DGMAN) for group event recommendation in EBSN. Firstly, we construct an attention network layer to learn the vector representation both of the groups and events through coarse-grained learning. Secondly, we build a two-layer attention network to further learn the vectors both of groups and events through fine-grained learning, which considers the influence of the internal features both of the groups and events. Finally, high-quality feature representations of groups and events are generated. We evaluate performance of DGMAN on three real-world datasets, and the extensive experimental results show that DGMAN outperforms the state-of-the-art approaches.

(2024) Published by SPIE. Downloading of the abstract is permitted for personal use only.

Citation Download Citation

Xiaobin Deng "Dual granularity multi-layer attention networks for group event recommendation", Proc. SPIE 12983, Second International Conference on Electrical, Electronics, and Information Engineering (EEIE 2023), 1298316 (15 January 2024); https://doi.org/10.1117/12.3017548

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available