Back to results
US 11,232,264 B2Granted
Natural language processing with non-ontological hierarchy models
AssigneeVerint Americas Inc.
InventorTimothy James Hewitt
Life of this patent
Estimating the term…
Abstract
A non-ontological hierarchy for language models is based on established psycholinguistic and neuro-linguistic evidences. By using non-ontological hierarchies, a more natural understanding of user's inputs and intents improve toward a better potential for producing intelligent responses in a conversational situation.
Claims15
A computerized system for conducting human-machine conversations, comprising:
- a computer comprising a processor connected to memory storing computer-implemented language processing software comprising a network of associations comprising target words arranged according to functions-like relationship definitions in respective domains of words;
- at least one data structure storing the network of associations in the memory;
- a user interface connected to the computer and configured to receive natural language inputs and to provide responses from the computer using the language processing software, wherein the language processing software implements a method comprising the steps of:
- identifying at least one base term and phrasal structures in a respective natural language input;
- identifying respective functional items in the phrasal structures used with the at least one base term in the natural language input and comparing the respective functional items with the functions-like relationship definitions in the network of associations to identify similarities between the at least one base term and the target words in the network of associations and to identify the respective domain of words that includes the target words;
- computing a response with words from the respective domain of words.
3 more elements
A computer implemented method of formulating a computerized response to a natural language input to a computer, the method comprising:
- storing a network of associations between target words arranged according to functions-like relationship definitions in a respective domain of words;
- identifying at least one base term and phrasal structures in a respective natural language input;
- identifying respective functional items in the phrasal structures used with the at least one base term in the natural language input;
- identifying similarities between the at least one base term and the target words in the network of associations by comparing the respective functional items from the natural language input with the functions-like relationship definitions in the network of associations;
- identifying the respective domain of words that includes the target words;
- and
- computing a response with words from the respective domain of words.
4 more elements
Dependencies are inferred from the claim text.