Content processing system for discourse

Claim 2

An automated method for determining content of discourse input to a computer system, the method comprising the steps of: storing, in computer readable medium, a plurality of hierarchically arranged categories that form a static hierarchy for classifying the content of the discourse input to the computer system; processing the input discourse to generate a plurality of thematic tags for terms including words and phrases, in the input discourse, wherein the thematic tags indicate applicability of a plurality of thematic constructions that define content of the input discourse; processing the thematic tags to generate theme terms from words and phrases in the input discourse based on existence or non existence of the thematic constructions as indicated by the thematic tags, wherein the theme terms identify overall content of the input discourse; selecting a category in the static hierarchy stored in the computer readable medium to classify at least one theme of the input discourse; generating at least one dynamic category from the words and phrases in the input discourse, wherein the dynamic category defines a greater level of detail than the categories in the static hierarchy; and generating a logical link between the dynamic category and the static hierarchy to form a more detailed classification hierarchy, wherein classification of the themes in categories and generation of the dynamic category defines the content of the input discourse.

Claim 3

The method as set forth in claim 1, further comprising the step of generating a kernel sentence by removing low content grammatically neutral words through use of the thematic tags.

Claim 4

The method as set forth in claim 1, further comprising the step of extracting at least one topic for a sentence of the input discourse including extracting context words that support the topic.

Claim 5

The method as set forth in claim 1, wherein the step of generating a plurality of themes comprises the step of generating a theme strength associated with each theme identified, based on the thematic tags, wherein the theme strength indicates relative thematic importance among themes generated from the input discourse.

Claim 8

The method as set forth in claim 5, wherein the step of generating a plurality of themes comprises the step of generating a theme strength associated with each theme identified, based on the thematic tags, wherein the theme strength indicates relative thematic importance among themes that defines content for the input discourse.

Claim 9

The method as set forth in claim 5, wherein the step of storing a knowledge catalog comprises the step of storing a plurality of independent static ontologies that arrange the tasks hierarchically.

Claim 10

The method as set forth in claim 7, further comprising the step of generating a theme concept for each theme identified by extracting a topic from a higher level node of a static ontology.

Claim 11

The method as set forth in claim 5, further comprising the step of generating at least one dynamic hierarchy based on themes not contained as topics in the knowledge catalog.

Claim 12

The method as set forth in claim 9, wherein the step of generating at least one dynamic hierarchy comprises the steps of: generating a plurality of head words from the input discourse, wherein the head words represent secondary themes of the input discourse that do not correlate to topics in the knowledge catalog; generating, for each head word, contextual relationships based on use of each head word in the input discourse; and generating a dynamic hierarchy for a head word based on the contextual relationships for the head word.

Claim 13

The method as set forth in claim 9, further comprising the step of expanding the predefined topics of the knowledge catalog to include the at least one dynamic hierarchy.

Claim 14

The method as set forth in claim 11, further comprising the step of classifying the themes of the input discourse by cross referencing dynamic hierarchies to generate a more detailed classification criteria.

Claim 15

The method as set forth in claim 5, wherein the step of generating a plurality of themes from the thematic tags comprises the steps of: identifying content carrying words and phrases in the input discourse; and generating the themes based on the thematic tags and the content carrying words.

Claim 16

The method as set forth in claim 13, wherein the step of identifying content carrying words and phrases in the input discourse comprises the step of identifying non-ambiguous content carrying words and phrases in the input discourse.

Claim 17

The method as set forth in claim 13, further comprising the steps of: generating for each content carrying word a nominalized or noun form for the content carrying word; and generating, if applicable, a mass sense and a count sense for each content carrying word.

Claim 18

A computer readable medium containing program instructions for execution on a computer system, which when executed by a computer, cause the computer to perform the steps of: storing, in computer readable medium, a plurality of hierarchically arranged categories that form a static hierarchy for classifying the content of the discourse input to the computer system; processing the input discourse to generate a plurality of thematic tags for terms including words and phrases, in the input discourse, wherein the thematic tags indicate applicability of a plurality of thematic constructions; processing the thematic tags to generate theme terms from words and phrases in the input discourse based on existence or non existence of the thematic constructions as indicated by the thematic tags, wherein the theme terms identify overall content of the input discourse; selecting a category in the static hierarchy stored in the computer readable medium to classify at least one theme of the input discourse; generating at least one dynamic category from the words and phrases in the input discourse, wherein the dynamic category defines a greater level of detail than the categories in the static hierarchy; and generating a logical link between the dynamic category and the static hierarchy to form a more detailed classification hierarchy, wherein classification of the themes in categories and generation of the dynamic category defines the content of the input discourse.

Claim 19

A computer readable medium containing program instructions for execution on a computer system, which when executed by a computer, cause the computer to perform the steps of: storing, in computer readable medium, a plurality of hierarchically arranged categories that form a static hierarchy for classifying the content of the discourse input to the computer system; processing the input discourse to generate a plurality of thematic tags for terms including words and phrases, in the input discourse, wherein the thematic tags indicate applicability of a plurality of thematic constructions; processing the thematic tags to generate theme terms from words and phrases in the input discourse based on existence or non existence of the thematic constructions as indicated by the thematic tags, wherein the theme terms identify overall content of the input discourse; selecting a category in the static hierarchy stored in the computer readable medium to classify at least one theme of the input discourse; generating at least one dynamic category from the words and phrases in the input discourse, wherein the dynamic category defines a greater level of detail than the categories in the static hierarchy; and generating a logical link between the dynamic category and the static hierarchy to form a more detailed classification hierarchy, wherein classification of the themes in categories and generation of the dynamic category defines the content of the input discourse.

Claim 20

The computer readable medium as set forth in claim 16, further comprising instructions for generating a kernel sentence by removing low content grammatically neutral words through use of the thematic tags.

Claim 21

The computer readable medium as set forth in claim 16, further comprising instructions for extracting at least one topic for a sentence of the input discourse including extracting context words that support the topic.

Claim 22

The computer readable medium as set forth in claim 16, wherein instructions for generating a plurality of themes comprises instructions for generating a theme strength associated with each theme identified, based on the thematic tags, wherein the theme strength indicates relative thematic importance among themes generated from the input discourse.

Claim 23

A computer readable medium containing program instructions for execution on a computer system, which when executed by a computer, cause the computer to perform the steps of: storing, in computer readable medium, a knowledge catalog that includes a plurality of pre-defined topics for classifying the content of the input discourse; processing said input discourse by generating a plurality of thematic tags for terms in the input discourse, wherein the thematic tags indicate applicability of a plurality of thematic constructions that define content of the input discourse; processing the thematic tags to generate a plurality of themes of the input discourse; and generating at least one classification from a theme of the input discourse by correlating the theme to a topic stored in the computer readable medium, wherein classification of the themes in the topics defines the content of the input discourse.

Claim 24

A computer readable medium containing program instructions for execution on a computer system, which when executed by a computer, cause the computer to perform the steps of: storing, in computer readable medium, a knowledge catalog that includes a plurality of pre-defined topics for classifying the content of the input discourse; processing said input discourse by generating a plurality of thematic tags for terms in the input discourse, wherein the thematic tags indicate applicability of a plurality of thematic constructions that define content of the input discourse; processing the thematic tags to generate a plurality of themes of the input discourse; and generating at least one classification from a theme of the input discourse by correlating the theme to a topic stored in the computer readable medium, wherein classification of the themes in the topics defines the content of the input discourse.

Claim 25

The computer readable medium as set forth in claim 20, wherein instructions for generating a plurality of themes comprise instructions for generating a theme strength associated with each theme identified, based on the thematic tags, wherein the theme strength indicates relative thematic importance among themes that defines content for the input discourse.

Claim 26

The computer readable medium as set forth in claim 20, wherein the instructions for storing a knowledge catalog comprise instructions for storing a plurality of independent static ontologies that arrange the tasks hierarchically.

Claim 27

The computer readable medium as set forth in claim 22, further comprising instructions for generating a theme concept for each theme identified by extracting a topic from a higher level node of a static ontology.

Claim 28

The computer readable medium as set forth in claim 20, further comprising instructions for generating at least one dynamic hierarchy based on themes that are not pre-defined topics in the knowledge catalog.

Claim 29

The computer readable medium as set forth in claim 24, wherein the instructions for generating at least one dynamic hierarchy comprises instructions for: generating a plurality of head words from the input discourse, wherein the head words represent secondary themes of the input discourse that do not correlate to the pre-defined topics in the knowledge catalog; generating, for each head word, contextual relationships based on use of each head word in the input discourse; and generating a dynamic hierarchy for a head word based on the contextual relationships for the head word.

Claim 30

The computer readable medium as set forth in claim 24, further comprising instructions for expanding the pre-defined topics of the knowledge catalog to include the at least one dynamic hierarchy.

Claim 31

The computer readable medium as set forth in claim 26, further comprising instructions for classifying the themes of the input discourse by cross referencing dynamic hierarchies to generate a more detailed classification criteria.

Claim 32

The computer readable medium as set forth in claim 20, wherein the instructions for generating a plurality of themes from the thematic tags comprises instructions for: identifying content carrying words and phrases in the input discourse; and generating the themes based on the thematic tags and the content carrying words.

Claim 33

The computer readable medium as set forth in claim 28, wherein the instructions for identifying content carrying words and phrases in the input discourse comprises instructions for identifying non-ambiguous content carrying words and phrases in the input discourse.

Claim 34

The computer readable medium as set forth in claim 28, further comprising instructions for: generating for each content carrying word a nominalized or noun form for the content carrying word; and generating, if applicable, a mass sense and a count sense for each content carrying word.

Claim 35

A content processing system for determining thematic content of input discourse, said system comprising: computer readable medium for storing a knowledge catalog that includes a plurality of pre-defined topics for classifying the content of the input discourse; a theme parser processor for processing said input discourse by generating a plurality of thematic tags for terms in the input discourse, wherein the thematic tags indicate applicability of a plurality of thematic constructions that define content of the input discourse; and a theme vector processor for processing the thematic tags to generate a plurality of themes of the input discourse, and for generating at least one classification for a theme of the input discourse by correlating the theme to a topic stored in the computer readable medium, wherein classification of the themes in the knowledge catalog defines the content of the input discourse.

Claim 36

A content processing system for determining thematic content of input discourse, said system comprising: computer readable medium for storing a knowledge catalog that includes a plurality of pre-defined topics for classifying the content of the input discourse; a theme parser processor for processing said input discourse by generating a plurality of thematic tags for terms in the input discourse, wherein the thematic tags indicate applicability of a plurality of thematic constructions that define content of the input discourse; and a theme vector processor for processing the thematic tags to generate a plurality of themes of the input discourse, and for generating at least one classification for a theme of the input discourse by correlating the theme to a topic stored in the computer readable medium, wherein classification of the themes in the knowledge catalog defines the content of the input discourse.

Claim 37

The content processing system as set forth in claim 31, further comprising a context indexing processor for generating at least one dynamic category from words and phrases in the input discourse, wherein the dynamic category defines a greater level of detail than the categories in the knowledge, and for generating a logical link between the dynamic category and the knowledge catalog to form a more detailed classification hierarchy, wherein classification of the themes in categories and generation of the dynamic category defines the content of the input discourse.