Multi-stage pattern reduction for natural language processing

• US 7,599,831 B2
• Filed: 11/10/2005
• Issued: 10/06/2009
• Est. Priority Date: 03/14/2003
• Status: Expired due to Fees

• Current Assignee: Sonim Technologies, Inc.
• Original Assignee: Sonim Technologies, Inc.
• Inventors: Ford, W. Randolph
• Primary Examiner(s): Hudspeth; David R
• Assistant Examiner(s): ALBERTALLI, BRIAN LOUIS
• Application Number:
  US11/271,202
• Publication Number:
  US 20060167678A1
• Time in Patent Office: 1,426 Days
• Field of Search: None
• 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
• CPC Class Codes: G10L 15/1815 Semantic context, e.g. disambiguation of the recognition hypotheses based on word meaning
• CPC Class Codes: Y10S 707/99933 Query processing, i.e. searching
• CPC Class Codes: Y10S 707/99942 Manipulating data structure, e.g. compression, compaction, compilation

Abstract

A computer program product for controlling the computer'"'"'s processor to perform responsive actions a natural language input has: (1) vocabulary, phrase and concept databases of words, phrase and concepts, respectively, that can be recognized in the inputted communication, wherein each of these database elements is representable by a designated semantic symbol, (2) means for searching the inputted communication to identify the words in the communication that are contained within the vocabulary database, (3) means for expressing the communication in terms of the word semantic symbols that correspond to each of the words identified in the inputted communication, (4) means for searching the communication when expressed in terms of its corresponding word semantic symbols so as to identify the phrases in the communication that are contained within the phrase database, (5) means for expressing the communication in terms of the phrase semantic symbols that correspond to each of the phrases identified in the communication, (6) means for searching the communication when expressed in terms of its corresponding phrase semantic symbols so as to identify the concepts in the communication that are contained within the concept database, and (7) means for expressing the communication in terms of the concept semantic symbols that correspond to each of the concepts identified in the inputted communication, wherein these concept semantic symbols are recognizable by the processor and can cause the processor to take action responsive to the inputted communication.

Claims

1. A method for converting a natural language communication, said method comprising:

a computer processor searching a natural language communication to identify words in said communication contained in a vocabulary database of words, wherein each of said database words is representable by a designated word semantic symbol and multiple different words in the vocabulary database determined to have a similar meaning are represented by the same semantic symbol in the vocabulary database;

expressing said communication in terms of word semantic symbols from the vocabulary database that correspond to each of said words identified in said communication;

searching said communication when expressed in terms of word semantic symbols so as to identify phrases in said communication contained in a phrase database of phrases, wherein each of said database phrases is representable by a designated phrase semantic symbol and multiple different phrases in the phrase database determined to have a similar meaning are represented by the same semantic symbol in the phrase database;

expressing said communication in terms of phrase semantic symbols from the phrase database that correspond to each of said phrases identified in said communication;

searching said communication when expressed in terms of the phrase semantic symbols so as to identify concepts in said communication; and

expressing said communication in terms of concept semantic symbols that correspond to each of said concepts identified in said communication.

2. The method of claim 1, further comprising:

upon identifying said concepts in said communication, causing the processor to take action responsive to said identified concepts into data that can be processed by the processor.

3. The method of claim 1, wherein searching said communication when expressed in terms of the phrase semantic symbols so as to identify concepts in said communication further comprises:

searching said communication when expressed in terms of the phrase semantic symbols so as to identify concepts in said communication contained in a concepts database of concepts, wherein multiple different phrases corresponding to the same concept are represented by the same semantic symbol in the concepts database.

4. The method of claim 1, wherein the natural language communication comprises an audio communication.

5. The method of claim 1, wherein the natural language communication comprises a user instruction to a computer system, and the method further comprises:

determining the instruction from the identified concepts.

6. An apparatus comprising:

means for receiving a communication;

means for storing semantic representations of at least words, phrases, and concepts, wherein multiple different words determined to have a similar meaning are represented by the same semantic representation stored in the means for storing and multiple different phrases determined to have a similar meaning are represented by the same semantic representation stored in the means for storing; and

means for determining a semantic representation of the communication using the semantic representations for words, phrases, and concepts stored in the means for storing, wherein the means for determining a semantic representation is further operable to perform at least a three-stage reduction of the communication to determine the semantic representation of the communication using words, phrases, and concepts and corresponding semantic representations stored in the means for storing.

7. The apparatus of claim 6, wherein the semantic representation of the communication includes semantic representations of stored concepts.

8. The apparatus of claim 7, further comprising:

processor means for taking action in response to the stored concepts identified from the semantic representations.

9. A method for performing a multistage reduction of a communication, the method comprising:

receiving the communication at a computer, wherein the communication includes at least one concept expressed in words;

performing a multistage reduction of the communication to determine a semantic representation of the communication, wherein the multistage reduction includes at least using stored words, phrases, and concepts and corresponding semantic representations to semantically represent the communication

wherein each of said stored words is represented by one of the stored semantic representations and multiple different words determined to have similar meanings are represented by the same stored semantic representation, and each of said stored phrases is represented by one of the stored semantic representations and multiple different phrases determined to have similar meanings are represented by the same stored semantic representation; and

using the semantic representation of the communication to identify a proper action to be taken by the computer.

10. The method of claim 9, wherein the multistage reduction includes using each of the stored words, phrases, and concepts at different stages of the multistage reduction.

11. The method of claim 9, wherein performing a multistage reduction at least comprises:

determining whether any of the stored words are in the communication;

for any of the stored words in the communication, identifying corresponding word semantic symbols representing the stored words;

using the word semantic symbols to identify stored phrases corresponding to the stored words in the communication, wherein the stored phrases are represented as phrase semantic symbols;

using the phrase semantic symbols corresponding to the identified phrases to identify corresponding stored concepts for the communication.

12. The method of claim 11, wherein the stored concepts are represented as concept semantic symbols, the method further comprising:

expressing the identified stored concepts as the corresponding concept semantic symbols.

13. The method of claim 11, wherein identifying corresponding word semantic symbols representing the stored words comprises:

searching a vocabulary database using to identify the corresponding word semantic symbols representing the stored words, wherein multiple words in the vocabulary database determined to be exchangeable without significantly changing a concept in a communication are represented by a same semantic symbol in the vocabulary database.

14. The method of claim 13, wherein identifying stored phrases corresponding to the stored words in the communication comprises:

searching a phrase database using the identified words semantic symbols to identify the corresponding phrase semantic symbols associated with the identified words, wherein multiple phrases in the phrase database determined to be exchangeable without significantly changing a concept in a communication are represented by a same semantic symbol in the phrase database.

15. The method of claim 14, wherein identify corresponding stored concepts for the communication comprises:

searching a concept database using the identified phrase semantic symbols to identify corresponding concept semantic symbols associated with the identified phrases, wherein multiple concepts in the concepts database determined to be exchangeable without significantly changing a concept in a communication are represented by a same semantic symbol in the concept database.

16. The method of claim 9, wherein the semantic representation of the communication identifies concepts in the communication, and the proper action to be taken by the computer is determined based on the identified concepts.

17. The method of claim 9, wherein the received communication comprises an audio communication.