Multi-dimensional method and apparatus for automated language interpretation

• US 7,403,890 B2
• Filed: 05/13/2002
• Issued: 07/22/2008
• Est. Priority Date: 05/13/2002
• Status: Expired due to Fees

• Current Assignee: Knowledgenetica Corporation
• Original Assignee: Knowledgenetica Corporation
• Inventors: Roushar, Joseph C.
• Primary Examiner(s): Lerner; Martin
• Application Number:
  US10/144,739
• Publication Number:
  US 20030216919A1
• Time in Patent Office: 2,262 Days
• Field of Search: 704/1, 704/9, 704/10, 704/270, 707/1, 707/2, 707/3, 707/100, 707/101, 707/104.1, 706/55, 706/56, 706/45
• US Class Current: 704/9
• CPC Class Codes: G06F 40/284 Lexical analysis, e.g. tokenisation or collocates
• CPC Class Codes: G06F 40/289 Phrasal analysis, e.g. finite state techniques or chunking
• CPC Class Codes: G06F 40/30 Semantic analysis

Abstract

A method and apparatus for natural language interpretation are described. The invention includes a schema and apparatus for storing, in digital, analog, or other machine-readable format, a network of propositions formed of a plurality of text and/or non-text objects, and the steps of retrieving a string of input text, and locating all associated propositions in the network for each word in the input string. Embodiments of the invention also include optimization steps for locating said propositions, and specialized structures for storing them in a ready access storage area simulating human short-term memory. The schema and steps may also include structures and processes for obtaining and adjusting the weights of said propositions to determine posterior probabilities representing the intended meaning. Embodiments of the invention also include an apparatus designed to apply an automated interpretation algorithm to automated voice response systems and portable knowledge appliance devices.

Claims

1. A method of deducing the meaning of a sentence or phrase comprising the steps of:

A) forming, a priori, a set of propositions each comprising at least an x, y, r and c object wherein x is a first object, y is a second object, r is a relationship between the first and second objects and c is a context of the relationship between the first and second objects;

B) assigning, a priori, a weight to each proposition formed in step A to represent a confidence value of the validity of such proposition as compared to other propositions with the same first object;

C) retrieving, upon receiving input of the sentence or phrase, applicable propositions by matching each word in the sentence or phrase to all propositions with the same first object;

D) modifying the a priori weight of the propositions retrieved in step C by applying applicable heuristics through one or more of the steps of;

1) applying heuristics to analyze the morphology of each word in a sentence or phrase to identify if a prefix or suffix has been added to a root word that may impact the meaning of the root word;

2) applying semantic heuristics to the sentence or phrase to create a hierarchy of statistically likely semantic roles;

3) applying syntax heuristics to the sentence or phrase to create a hierarchy of statistically likely semantic roles;

4) applying context heuristics to the “

c”

objects of all applicable propositions of step C to create a hierarchy of statistically probable contexts of the sentence or phrase;

E) applying the modified weights to the applicable propositions; and

F) selecting the propositions with the highest modified weights;

whereby the meaning of a phrase or sentence and its constituent words is deduced.