Computer mediated natural language based communication augmented by arbitrary and flexibly assigned personality classification systems

• US 8,145,474 B1
• Filed: 12/21/2007
• Issued: 03/27/2012
• Est. Priority Date: 12/22/2006
• Status: Expired due to Fees

• Current Assignee: Avaya Incorporated
• Original Assignee: Avaya Incorporated
• Inventors: Gentle, Christopher R., Daily, Daniel J., Thomas, Michael J., MAITY, Ashis, Kawahara, Lisa Y.
• Primary Examiner(s): RIDER, JUSTIN W
• Application Number:
  US11/962,821
• Time in Patent Office: 1,558 Days
• Field of Search: 704/9, 704/10, 704/275
• US Class Current: 704/9
• CPC Class Codes: G06F 11/2257 using expert systems
• CPC Class Codes: G06N 5/01 Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound

Abstract

The present invention is directed, in one embodiment, to an system that lexically analyzes a behavior-descriptive input stream to provide a tokenized equivalent, syntactically parses the tokenized equivalent to provide a parser output, semantically analyzes the parser output to provide an indication of at least one of a behavior and type of behavior described by the input stream, and, based on the indication, selects an appropriate response.

Claims

1. A method, comprising:

(a) receiving a behavior-descriptive input stream;

determining, by a meta-parser that performs a preliminary parse on one or more portions of the input stream, a structure of the input stream for either subsequent selection of a parser or subsequent creation of a parser, the meta-parser outputting the input stream and an indication of whether the meta-parser was able to successfully parse the input stream, the input stream output from the meta-parser being passed to the parser;

(b) lexically analyzing the behavior-descriptive input stream to provide a tokenized equivalent;

(c) syntactically parsing the tokenized equivalent to provide a parser output;

(d) semantically analyzing, by a processor, the parser output to provide an indication of at least one of a behavior and type of behavior described by the input stream; and

(e) based on the indication, selecting an appropriate response.

2. The method of claim 1, wherein the input stream comprises contextual information, wherein the contextual information comprises a plurality of a descriptor of a physical action, a physical trait, a physical characteristic, a logical trait, a logical characteristic, a logical location, a spatial location, an identity, and a role.

3. The method of claim 1, wherein the input stream is defined by a plurality of grammars and further comprising for a selected portion of the input stream:

(f) determining a set of grammars potentially defining the input stream portion;

(g) selecting a set of parsers, each parser set member being operable to parse according to at least one grammar set member; and

(h) providing the input stream portion to at least one member of the set of parsers.

4. The method of claim 1, wherein the input stream is potentially defined by a plurality of grammars and further comprising:

(f) determining that at least a portion of the input stream does not correspond to a known grammar;

(g) in response, creating a grammar defining the at least a portion of the input stream;

(h) performing at least one of the following sub-steps;

(H1) creating a parser operable to apply syntactical rules associated with the grammar;

(H2) creating a set of tokens associated with the created grammar; and

(H3) creating a semantic encoder operable to apply semantical rules associated with the grammar.

5. The method of claim 1, wherein the at least one of a behavior and type of behavior is at least one of a personality classification, type, and function, wherein at least some of the tokens in the tokenized equivalent are associated with one or more natural language words, wherein the input stream is an interpersonal communication, and wherein the input stream is defined by a plurality of grammars, each grammar being associated with the at least one of a personality classification, type, and function.

6. The method of claim 5, wherein the appropriate response is at least one of:

(i) providing to a recipient of the interpersonal communication a recommended response to the interpersonal communication, the recommended response reflecting at least one of a personality classification, type, and function determined to be associated with a source of the interpersonal communication;

(ii) providing to the recipient a description of at least one of a personality classification, type, and function determined to be associated with the source of the interpersonal communication;

(iii) providing to the source a recommended selection of words, phrases, and/or sentences to be used in the interpersonal communication, the recommendation reflecting at least one of a personality classification, type, and function determined to be associated with the recipient; and

(iv) providing to the source a description of at least one of a personality classification, type, and function determined to be associated with the recipient of the interpersonal communication.

7. The method of claim 5, wherein, in step (d), at least one of a personality classification, type, and function associated with a source of the interpersonal communication is determined.

8. The method of claim 5, wherein step (d) outputs a personality dynamic associated with a source of the interpersonal communication and wherein step (e) comprises the substeps:

(E1) selecting a set of user interface transformation rules;

(E2) applying the personality dynamic to the set of user interface transformation rules to select a set of commands to provide to a user interface.

9. A non-transitory computer readable medium comprising processor executable instructions operable to perform the steps of claim 1.

10. A system, comprising:

(a) an input operable to receive a behavior-descriptive input stream;

determining, by a meta-parser that performs a preliminary parse on one or more portions of the input stream, a structure of the input stream for either subsequent selection of a parser or subsequent creation of a parser, the meta-parser outputting the input stream and an indication of whether the meta-parser was able to successfully parse the input stream, the input stream output from the meta-parser being passed to the parser;

(b) a lexical analyzer operable to analyze the behavior-descriptive input stream to provide a tokenized equivalent;

(c) the parser operable to syntactically parse the tokenized equivalent to provide a parser output;

(d) a semantic encoder and processor operable to semantically analyze the parser output to provide an indication of at least one of a behavior and type of behavior described by the input stream; and

(e) a semantic evaluator operable, based on the indication, to select an appropriate response.

11. The system of claim 10, wherein the input stream comprises contextual information, wherein the contextual information comprises a plurality of a descriptor of a physical action, a physical trait, a physical characteristic, a logical trait, a logical characteristic, a logical location, a spatial location, an identity, and a role.

12. The system of claim 10, wherein the input stream is defined by a plurality of grammars and further comprising:

(f) a meta-parser operable, for a selected portion of the input stream, to determine a set of grammars potentially defining the input stream portion; and

(g) a rules engine operable to select a set of parsers, each parser set member being operable to parse according to at least one grammar set member and provide the input stream portion to at least one member of the set of parsers.

13. The system of claim 10, wherein the input stream is potentially defined by a plurality of grammars and further comprising:

(f) a rules engine operable to determine that at least a portion of the input stream does not correspond to a known grammar and, in response, create a grammar defining the at least a portion of the input stream; and

(g) a parser generator operable to create a parser operable to apply syntactical rules associated with the grammar; and

(h) a semantic encoder generator operable to create a semantic encoder operable to apply semantical rules associated with the grammar.

14. The system of claim 10, wherein the at least one of a behavior and type of behavior is at least one of a personality classification, type, and function, wherein at least some of the tokens in the tokenized equivalent are associated with one or more natural language words, wherein the input stream is an interpersonal communication, and wherein the input stream is defined by a plurality of grammars, each grammar being associated with the at least one of a personality classification, type, and function.

15. The system of claim 14, wherein the appropriate response is at least one of:

(i) provide to a recipient of the interpersonal communication a recommended response to the interpersonal communication, the recommended response reflecting at least one of a personality classification, type, and function determined to be associated with a source of the interpersonal communication;

(ii) provide to the recipient a description of at least one of a personality classification, type, and function determined to be associated with the source of the interpersonal communication;

(iii) provide to the source a recommended selection of words, phrases, and/or sentences to be used in the interpersonal communication, the recommendation reflecting at least one of a personality classification, type, and function determined to be associated with the recipient; and

(iv) provide to the source a description of at least one of a personality classification, type, and function determined to be associated with the recipient of the interpersonal communication.

16. The system of claim 14, wherein at least one of the semantic encoder and evaluator determines at least one of a personality classification, type, and function associated with a source of the interpersonal communication.

17. The system of claim 14, wherein the semantic encoder outputs a personality dynamic associated with a source of the interpersonal communication and the semantic evaluator applies the personality dynamic to a set of user interface transformation rules to select a set of commands to provide to a user interface.

18. A method, comprising:

(a) receiving an input stream including an interpersonal communication;

determining, by a meta-parser that performs a preliminary parse on one or more portions of the input stream, a structure of the input stream for either subsequent selection of a parser or subsequent creation of a parser, the meta-parser outputting the input stream and an indication of whether the meta-parser was able to successfully parse the input stream, the input stream output from the meta-parser being passed to the parser;

(b) lexically analyzing, by a processor, the input stream to provide a tokenized equivalent;

(c) syntactically parsing, by the parser, the tokenized equivalent to provide a parser output; and

(d) semantically analyzing, by the processor, the parser output to provide an indication of at least one of a personality classification, type, and function associated with the input stream.

19. The method of claim 18, wherein the input stream comprises contextual information, wherein the contextual information comprises a plurality of a descriptor of a physical action, a physical trait, a physical characteristic, a logical trait, a logical characteristic, a logical location, a spatial location, an identity, and a role and further comprising:

(e) providing, by a user interface, the indication to a user.

20. The method of claim 18, wherein the input stream is defined by a plurality of grammars, wherein the plurality of grammars are associated with at least one of a plurality of personality classification systems, a plurality of personality types, and a plurality of personality functions and further comprising for a selected portion of the input stream:

(e) determining a set of grammars potentially defining the input stream portion;

(f) selecting a set of parsers, each parser set member being operable to parse according to at least one grammar set member; and

(g) providing the input stream portion to at least one member of the set of parsers.

21. The method of claim 18, wherein the input stream is potentially defined by a plurality of grammars, wherein the plurality of grammars are associated with at least one of a plurality of personality classification systems, a plurality of personality types, and a plurality of personality functions, and further comprising:

(e) determining that at least a portion of the input stream does not correspond to a known grammar;

(f) in response, creating a grammar defining the at least a portion of the input stream;

(g) performing at least one of the following sub-steps;

(g1) creating a parser operable to apply syntactical rules associated with the grammar;

(g2) creating a set of tokens associated with the created grammar; and

(g3) creating a semantic encoder operable to apply semantical rules associated with the grammar.

22. The method of claim 18, wherein at least some of the tokens in the tokenized equivalent are associated with one or more natural language words and wherein the input stream is defined by a plurality of grammars, each grammar being associated with the at least one of a personality classification, type, and function.

23. The method of claim 18, further comprising at least one of:

(D1) providing to a recipient of the interpersonal communication a recommended response to the interpersonal communication, the recommended response reflecting at least one of a personality classification, type, and function determined to be associated with a source of the interpersonal communication;

(D2) providing to the recipient a description of at least one of a personality classification, type, and function determined to be associated with the source of the interpersonal communication;

(D3) providing to the source a recommended selection of words, phrases, and/or sentences to be used in the interpersonal communication, the recommendation reflecting at least one of a personality classification, type, and function determined to be associated with the recipient; and

(D4) providing to the source a description of at least one of a personality classification, type, and function determined to be associated with the recipient of the interpersonal communication.

24. The method of claim 18, wherein, in step (d), at least one of a personality classification, type, and function associated with a source of the interpersonal communication is determined.

25. The method of claim 18, wherein step (d) outputs a personality dynamic associated with a source of the interpersonal communication and further comprising:

(e) selecting a set of user interface transformation rules; and

(f) applying the personality dynamic to the set of user interface transformation rules to select a set of commands to provide to a user interface.

26. A computer readable medium comprising processor executable instructions operable to perform the steps of claim 18.

27. A method, comprising:

receiving a communication from an originator of the communication;

determining, by a meta-parser that performs a preliminary parse on one or more portions of the communication, a structure of the communication for either subsequent selection of a parser or subsequent creation of a parser, the meta-parser outputting the input stream and an indication of whether the meta-parser was able to successfully parse the communication, the communication output from the meta-parser being passed to the parser;

analyzing the communication, by a processor, to determine from the communication personality characteristics of the originator;

in response to the communication and the characteristics, indicating, by the processor, a form of a response to the communication that is compatible with the characteristics; and

sending a response having the indicated form to the originator.

28. A non-transitory computer readable medium comprising processor executable instructions operable to perform the steps of claim 27.