Module for creating a language neutral syntax representation using a language particular syntax tree

• US 20060004563A1
• Filed: 06/30/2004
• Published: 01/05/2006
• Est. Priority Date: 06/30/2004
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Ringger, Eric K., Campbell, Richard G.
• Granted Patent: US 7,596,485 B2
• Time in Patent Office: Days
• US Class Current: 704/9
• CPC Class Codes: G06F 40/211 Syntactic parsing, e.g. based on context-free grammar [CFG] or unification grammars
• CPC Class Codes: G06F 40/268 Morphological analysis
• CPC Class Codes: G06F 40/30 Semantic analysis

Abstract

A method or module for creating an Language Neutral Syntax (LNS) representation of a sentence from a language particular syntax representation such as found in the Penn Treebank for use by different applications. The method or module includes a node generator configured to create hierarchical and dependent nodes using phrasal and constituent nodes of the language particular syntax. A node dependency generator is configured to create an unordered hierarchical dependency structure for the hierarchical and dependent nodes using a semantic relation between the hierarchical and dependent nodes derived from the language particular syntax.

Claims

1. A computer executable medium having instructions which when implemented by a computer generates an unordered hierarchical dependency structure for a natural language sentence or phrase using a language particular syntactical representation of that sentence or phrase comprising:

a node generator configured to create hierarchical and dependent nodes using phrasal and constituent nodes of the language particular syntactical representation; and

a node dependency generator configured to create the unordered hierarchical nodal dependency structure for the hierarchical and dependent nodes using a semantic relation between the hierarchical and dependent nodes derived from the language particular syntactical representation.

2. The computer executable medium of claim 1 wherein the node generator uses a morphological analysis to derive a language neutral representation of a word or constituent for terminal nodes of the unordered hierarchical dependency structure using grammatical tags of the language particular syntactical representation.

3. The computer executable medium of claim 1 wherein the node generator uses a head analysis of the language particular syntactical representation to create the hierarchical and dependent nodes for the unordered hierarchical dependency structure.

4. The computer executable medium of claim 3 wherein the head analysis uses grammatical tags and language order syntax of the language particular syntactical representation.

5. The computer executable medium of claim 3 wherein the node generator creates a head node for each nonterminal node of the language particular syntactical representation based upon the head analysis.

6. The computer executable medium of claim 5 wherein the node generator derives a base form of a terminal node constituent of the language particular syntactical representation and corresponding grammatical features or attributes of the base form constituent is represented in a hierarchical or head node corresponding to the terminal node constituent.

7. The computer executable medium of claim 1 wherein the node generator provides a node derivation for auxiliary or complex verb construction including a main verb and aspectual and voice features.

8. The computer executable medium of claim 1 wherein the node generator provides a derivation to represent scope and logical relation of constituents of a compound and coordinating noun phrase in the hierarchical dependency structure.

9. The computer executable medium of claim 1 wherein the node dependency generator create semantic arcs between the hierarchical and dependent nodes using grammatical labels and syntactical analysis of the language particular syntactical representation to create the unordered hierarchical dependency structure.

10. The computer executable medium of claim 1 wherein the node dependency generator uses grammatical labels of the language particular syntactical representation to create the unordered hierarchical dependency structure.

11. The computer executable medium of claim 1 wherein the node dependency generator is configured to create the hierarchical dependency structure to represent logical scope and relation of constituents of a noun phrase or verb phrase using grammatical labels and syntax of the language particular syntactical representation and scope assignment criteria.

12. The computer executable medium of claim 1 wherein the node dependency generator is configured to create semantic arcs semantically linking the hierarchical and dependent nodes of the unordered hierarchical dependency structure.

13. The computer executable medium of claim 1 wherein the node dependency generator creates abstract nodes to represent tense, comparison, superlative, negation, and unspecified arguments, to the extent these are not represented by individual words or phrases in the [original]natural language sentence or phrase.

14. The computer executable medium of claim 1 wherein the language particular syntactical representation is a Penn Treebank annotation.

15. A method for creating an unordered hierarchical dependency structure to represent a natural language using a language particular syntactical representation comprising steps of:

deriving hierarchical and dependent nodes using phrasal and constituent nodes of the language particular syntactical representation; and

creating a hierarchical dependency structure between the hierarchical and dependent nodes including semantic links between the hierarchical and dependent nodes using grammatical tags and syntax of the language particular representation.

16. The method of claim 15 wherein the step of creating the hierarchical and dependent nodes further comprises:

lemmatizing constituents of terminal nodes of the unordered hierarchical dependency structure and assigning attributes to a nonterminal node corresponding to the terminal node for the lemmatized constituent to provide a language neutral representation.

17. The method of claim 15 and further comprising the step of:

performing a head analysis of the language particular syntactical representation; and

creating the unordered hierarchical dependency structure based upon a semantic relation between constituents of the language particular syntactical representation derived using the head analysis.

18. The method of claim 17 wherein hierarchical head nodes are created for each terminal or constituent node based upon the head analysis.

19. The method of claim 15 and further comprising the step of:

creating abstract nodes for the unordered hierarchical structure to represent negation, tense or modal operator.

20. The method of claim 15 and further comprising the step of:

converting a bracketed data string into a language particular syntax tree.

21. The method of claim 15 and further comprising the step of:

performing a scope analysis to create the hierarchical dependency structure to represent logical and grammatical relation of constituents of a noun or verb phrase.