Title:
TEXT ENTITY EXTRACTION METHOD, DEVICE, APPARATUS, AND STORAGE MEDIA
Document Type and Number:
WIPO Patent Application WO/2018/153295
Kind Code:
A1
Abstract:
The present invention relates to the field of text extraction. Provided are a text entity extraction method, device, apparatus, and storage media. The method comprises: determining a candidate text entity contained within a target text (301); combining the candidate text entity and generating a candidate word segmentation combination corresponding to the target text, wherein each of the candidate word segmentation combinations comprises different candidate text entities (302); calculating a combination probability for each of the candidate word segmentation combinations, wherein the combination probability refers to the probability of the candidate word segmentation combination being grammatical when used in the target text (303); determining, by using the combination probability, a target word segmentation combination corresponding to the target text (304); and extracting, according to the target word segmentation combination, a text entity from the target text (305). Text entity extraction does not rely on a preconfigured template, reducing the length of time required for pre-configuration. When compared to template-based matching, a method adopting probability calculation is faster when determining a target word segmentation combination, and can address an issue of text entity extraction failure due to inability to find a matching template.
Inventors:
BAO HENGYAO (CN)
SU KE (CN)
CHEN YI (CN)
RAO MENGLIANG (CN)
SU KE (CN)
CHEN YI (CN)
RAO MENGLIANG (CN)
Application Number:
PCT/CN2018/076274
Publication Date:
August 30, 2018
Filing Date:
February 11, 2018
Export Citation:
Assignee:
TENCENT TECH SHENZHEN CO LTD (CN)
International Classes:
G06F17/21; G10L15/22
Foreign References:
| CN106910501A | 2017-06-30 | |||
| CN105988978A | 2016-10-05 | |||
| CN105279149A | 2016-01-27 | |||
| CN1819018A | 2006-08-16 | |||
| CN101208689A | 2008-06-25 | |||
| US6415250B1 | 2002-07-02 |
Attorney, Agent or Firm:
BEIJING SAN GAO YONG XIN INTELLECTUAL PROPERTY AGENCY CO., LTD. (CN)
Download PDF:
Previous Patent: FACE TRACKING METHOD, STORAGE MEDIUM, AND TERMINAL DEVICE
Next Patent: METHOD AND DEVICE FOR FINDING VEHICLE QUEUE
Next Patent: METHOD AND DEVICE FOR FINDING VEHICLE QUEUE