Process for the automatic generation of a textual expression from a semantic representation using a computer system

• US 6,965,856 B1
• Filed: 06/30/1999
• Issued: 11/15/2005
• Est. Priority Date: 10/29/1998
• Status: Expired due to Term

• Current Assignee: Nuance Communications, Inc. (Microsoft Corporation)
• Original Assignee: International Business Machines Corporation
• Inventors: Stuermer, Thomas
• Primary Examiner(s): Knepper, David D.
• Assistant Examiner(s): Han, Qi
• Application Number:
  US09/343,929
• Time in Patent Office: 2,330 Days
• Field of Search: 704/1, 704/9, 704/10, 704/258, 704/260, 704/255
• US Class Current: 704/1
• CPC Class Codes: G06F 40/10 Text processing natural language analysis G06F40/20; semantic analysis G06F40/30; processing or translation of natural language G06F40/40

Abstract

A procedure for the automatic generation of a textual expression from a semantic representation by a computer-system is described. With the procedure, a statistical model is determined by the computer-system on a plurality of pre-determined pairs of semantic representations and associated expressions and stored. A semantic representation, from which an associated expression is determined by the computer system by means of the statistical model, is presented to the computer system. These steps are repeated by the computer system for further semantic representations if necessary.

Claims

1. A method for the automatic generation of a textual expression from a semantic representation, comprising the computer-executed steps of:

building a statistical model from a plurality of pre-determined pairs of semantic representations and associated expressions, wherein the building further comprises determining which information in an external format is present in a negative list and which information in the external format is absent in the negative list, wherein the negative list describes information in the external format that is irrelevant to the statistical model, wherein the statistical model comprises a decision tree; and

producing a first associated expression from a first semantic representation using the statistical model.

2. The method of claim 1, wherein the building step farther comprises:

converting the pairs of semantic representations and associated expressions from the external format into an internal format.

3. The method of claim 2, wherein the converting step further uses the negative list and a translation table.

4. The method of claim 3, wherein the converting step further comprises:

converting information that is absent in the negative list from the external format to the internal format using the translation table; and

refraining from converting information that is present in the negative list.

5. The method of claim 2, wherein the building step further comprises:

determining a plurality of questions;

classifying the information in the internal format using the plurality of questions; and

calculating the statistical model from the internal format using the plurality of questions.

6. The method of claim 5, wherein the determining step further comprises:

determining the plurality of questions from pre-determined boundary conditions.

7. An apparatus for the automatic generation of a textual expression from a semantic representation, comprising:

means for building a statistical model from a plurality of pre-determined pairs of semantic representations and associated expressions, wherein the building further comprises determining which information in an external format is present in a negative list and which information in the external format is absent in the negative list, wherein the negative list describes information in the external format that is damaging to the statistical model, wherein the statistical model comprises a decision tree; and

means for producing a first associated expression from a first semantic representation using the statistical model.

8. The apparatus of claim 7, wherein the means for building further comprises:

means for converting the pairs of semantic representations and associated expressions from the external format into an internal format.

9. The apparatus of claim 8, wherein the means for converting further uses the negative list and a translation table.

10. The apparatus of claim 9, wherein the means for converting further comprises:

means for converting information that is absent in the negative list from the external format to the internal format using the translation table; and

means for refraining from converting information that is present in the negative list.

11. The apparatus of claim 8, wherein the means for building further comprises:

means for determining a plurality of questions;

means for classifying the information in the internal format using the plurality of questions; and

means for calculating the statistical model from the internal format using the plurality of questions.

12. The apparatus of claim 11, wherein the means for determining further comprises:

means for determining the plurality of questions from predetermined boundary conditions.