Abstract: A method for unsupervised language understanding may be carried out in a software environment and is operative to identify pairs of semantically-related words within a corpus of documents, extract candidate intents from the pairs of semantically-related words, and define the most general intents from the candidate intents, without human guidance or supervision.
Type:
Grant
Filed:
February 25, 2019
Date of Patent:
April 13, 2021
Assignee:
LANG ARTIFICIAL INTELLIGENCE INC.
Inventors:
Henry Anaya, Jorge Miguel PeƱalva, III, Enrique Fueyo, III