Title:
METHOD AND DEVICE FOR WORD EMBEDDING ON BASIS OF CONTEXT INFORMATION AND MORPHOLOGICAL INFORMATION OF WORD
Document Type and Number:
WIPO Patent Application WO/2020/204364
Kind Code:
A2
Abstract:
The present invention relates to a method and device for word embedding on the basis of context information and morphological information of a word. A method for word embedding according to one embodiment of the present invention comprises the steps of: processing a sentence by replacing an out of vocabulary (OOV) word in the sentence to be learned with an unknown token; inputting characters of a target word excluding the out of vocabulary word in the processed sentence as an input of a context character model to be learned; combining surrounding context vectors for surrounding words of the target word in the sentence so as to set the context character model as an initial state; and learning the context character model such that an error can be minimized between predicted embedding of the target word and real embedding of the target word, the predicted embedding being generated by connecting a forward hidden state and a backward hidden state calculated from the context character model.
More Like This:
Inventors:
WON MIN SUB (KR)
LEE JEE HYONG (KR)
LEE SANG HEON (KR)
SHIN YUN SEOB (KR)
JEONG DONG EON (KR)
LEE JEE HYONG (KR)
LEE SANG HEON (KR)
SHIN YUN SEOB (KR)
JEONG DONG EON (KR)
Application Number:
PCT/KR2020/003000
Publication Date:
October 08, 2020
Filing Date:
March 03, 2020
Export Citation:
Assignee:
RESEARCH & BUSINESS FOUND SUNGKYUNKWAN UNIV (KR)
International Classes:
G06F40/205; G06F40/289; G06N20/00
Attorney, Agent or Firm:
ENVISION PATENT & LAW FIRM (KR)
Download PDF: