Title:
MULTI-MODAL NEWS RECOMMENDATION METHOD AND APPARATUS BASED ON MULTI-HEAD SELF-ATTENTION NEURAL MECHANISM
Document Type and Number:
WIPO Patent Application WO/2023/065618
Kind Code:
A1
Abstract:
The present invention provides a multi-modal news recommendation method and apparatus based on a multi-head self-attention neural mechanism. The method comprises steps: collecting data information, comprising news data, feature data, and trace data; fusing the data information into unified news features on the basis of a multi-component feature cross model of a view-level attention mechanism, a hot news real-time prediction technology of streaming data, and a multi-modal information fusion technology of intelligent frame extraction; and taking the unified news features as model input, and completing a personalized accurate recommendation function by means of a user interest representation model and in combination with a highest future influence strategy.
Inventors:
OU ZHONGHONG (CN)
LIU PEIHANG (CN)
HAN ZONGZHI (CN)
SONG MEINA (CN)
ZHONG MAOHUA (CN)
LIANG HAOGUANG (CN)
LIU PEIHANG (CN)
HAN ZONGZHI (CN)
SONG MEINA (CN)
ZHONG MAOHUA (CN)
LIANG HAOGUANG (CN)
Application Number:
PCT/CN2022/087220
Publication Date:
April 27, 2023
Filing Date:
April 15, 2022
Export Citation:
Assignee:
BEIJING UNIV OF POSTS&TELECOMMUNICATIONS (CN)
International Classes:
G06F16/9535
Foreign References:
CN114154054A | 2022-03-08 | |||
CN111444428A | 2020-07-24 | |||
CN111488931A | 2020-08-04 | |||
CN114693397A | 2022-07-01 | |||
US20120330936A1 | 2012-12-27 |
Attorney, Agent or Firm:
TSINGYIHUA INTELLECTUAL PROPERTY LLC (CN)
Download PDF: