Title:
WORD EMBEDDING-BASED SEARCH METHOD, APPARATUS AND DEVICE, AND STORAGE MEDIUM
Document Type and Number:
WIPO Patent Application WO/2022/141876
Kind Code:
A1
Abstract:
A word embedding-based search method, apparatus and device, and a storage medium, relating to the technical field of natural language processing. The method comprises: in response to index content input by a user, determining keywords of the index content (S101); searching a pre-stored inverted index table for word embeddings of the keywords (S102); calculating similarity between each of the word embeddings and all target long texts (S103), the target long texts being all pre-stored long texts associated with the index content; and displaying, on the basis of the similarity, a search result matching the index content (S104), all the pre-stored long texts associated with the index content being all long texts comprising the keywords of the index content and obtained by analyzing the index content on the basis of an XLNet model. The method does not increase the calculation overhead while ensuring the search precision.
Inventors:
CHEN ZHENBO (CN)
ZHENG LIYING (CN)
XU LIANG (CN)
ZHENG LIYING (CN)
XU LIANG (CN)
Application Number:
PCT/CN2021/084253
Publication Date:
July 07, 2022
Filing Date:
March 31, 2021
Export Citation:
Assignee:
PING AN TECH SHENZHEN CO LTD (CN)
International Classes:
G06F16/33
Foreign References:
| CN110019668A | 2019-07-16 | |||
| CN111967258A | 2020-11-20 | |||
| CN110362678A | 2019-10-22 | |||
| CN112149005A | 2020-12-29 | |||
| US20190370273A1 | 2019-12-05 |
Attorney, Agent or Firm:
SHENZHEN ZHONGYI UNION INTELLECTUAL PROPERTY AGENCY CO., LTD. (CN)
Download PDF: