Title:
MODEL GENERATION METHOD, RECOMMENDATION METHOD AND CORRESPONDING APPARATUSES, DEVICE AND STORAGE MEDIUM
Document Type and Number:
WIPO Patent Application WO/2017/084362
Kind Code:
A1
Abstract:
Provided are a model generation method, a recommendation method and corresponding apparatuses, device and storage medium. In one aspect, the model generation method of the embodiments of the present invention comprises: obtaining at least one of: a document content feature vector of various entities, a logic association relationship feature vector of various entities, a user behaviour relationship feature vector of various entities and a feature vector of various entities in a knowledge domain; and performing machine learning to generate a depth fusion model according to at least one of the document content feature vector, the logic association relationship feature vector, the user behaviour relationship feature vector and the feature vector. Therefore, the technical solution provided by embodiments of the present invention can generate a depth fusion model by integrating various relationships between entities. The depth fusion model can be used for obtaining a surprise degree between the entities, so that an entity can be recommended to a user based on the surprise degree, satisfying a search recommendation demand of the user and improving the click rate of a recommendation entity.
Inventors:
HUANG JIZHOU (CN)
SUN MINGMING (CN)
DING SHIQIANG (CN)
WANG HAIFENG (CN)
SUN MINGMING (CN)
DING SHIQIANG (CN)
WANG HAIFENG (CN)
Application Number:
PCT/CN2016/089648
Publication Date:
May 26, 2017
Filing Date:
July 11, 2016
Export Citation:
Assignee:
BAIDU ONLINE NETWORK TECHNOLOGY BEIJING CO LTD (CN)
International Classes:
G06F17/30
Foreign References:
| CN101114294A | 2008-01-30 | |||
| CN104035917A | 2014-09-10 | |||
| CN104462293A | 2015-03-25 | |||
| CN105335519A | 2016-02-17 |
Attorney, Agent or Firm:
BEIJING WISPRO INTELLECTUAL PROPERTY LLP. (CN)
Download PDF: