Computation linguistics engine

• US 7,461,033 B1
• Filed: 12/22/2005
• Issued: 12/02/2008
• Est. Priority Date: 12/22/2005
• Status: Active Grant

• Current Assignee: T-Mobile Innovations LLC (Deutsche Telekom AG)
• Original Assignee: Sprint Communications Company LP (Deutsche Telekom AG)
• Inventors: McConnell, Von, Cope, Warren
• Primary Examiner(s): Vincent; David R
• Assistant Examiner(s): Bharadwaj; Kalpana
• Application Number:
  US11/315,958
• Time in Patent Office: 1,076 Days
• Field of Search: 706/60, 706/50, 706/12, 707/5, 707/4, 707/3, 715/716
• US Class Current: 706/12
• CPC Class Codes: G06F 16/345 Summarisation for human users
• CPC Class Codes: Y10S 707/99935 Query augmenting and refining, e.g. inexact access

Abstract

A system and method for generating textual structures describing the information contained on multiple web pages and blogs. The system comprises an ontology module, summation module, language generation module, and language bias module. The method comprises receiving a request for summarized web information, accumulating text from a plurality of web pages, parsing the accumulated text, indexing the text into a plurality of information sets, storing the plurality of information sets into a memory structure, aggregating information contained in the plurality of information sets to create a structure interpretation to satisfy the request, and creating at least one new textual structure from the structure interpretation. The ontology module parses and tags accumulated web text. The summation module creates a structure interpretation of the parsed and tagged web text. The language generation module creates textual structure describing the structure interpretation.

Claims

1. A computer system including a microprocessor for generating and presenting textual structures that summarize web content relevant to a topic, wherein the web content is posted on multiple web pages, comprising:

A plurality of computing units comprising the modules;

an ontology module to parse and classify text retrieved from at least one web page into at least one information set, wherein the at least one information set is stored;

a summation module to aggregate the at least one information set and create at least one structure interpretation using the at least one information set;

and

a language generation module to create at least one new textual structure describing the web content relevant to the topic by incorporating the at least one structure interpretation into one or more grammatical templates that include basic grammar confines; and

a language bias module including language bias rules that formats the at least one new textual structure to account for at least one of a user bias preference or a system bias preference, wherein the language bias rules module assigns values to a web page corresponding to it relative bias.

2. The system of claim 1, wherein the language bias module reformats the at least one new textual structure according to biases of information sources that are determined using predefined parameters.

3. The system of claim 1, further comprising:

a database to store commonly used ontology information.

4. The system of claim 1, further comprising:

an archive database to store the at least one new textual structure for a period of time determined by a parameter.

5. The system of claim 1, wherein the ontology module, the summation module, and the generation module share a common storage module.

6. The system of claim 1, further comprising:

a personal profile manager to allow continual parsing of web text and creation of a new textual structure according to user-specific parameters.

7. The system of claim 1, further comprising:

a storage module to store the at least one information set.

8. The system of claim 1, further comprising:

a storage module to store the at least one structure interpretation.

9. The system of claim 1, further comprising:

a storage module to store the at least one new textual structure.

10. A computerized method for generating and presenting one or more textual structures that summarize web content relevant to a topic, comprising:

receiving a request for summarized web information from a user;

accumulating text from a plurality of web pages.

11. The method of claim 10, further comprising:

referencing an archived database to determine whether the archived database contains a generated textual structure that satisfies the user'"'"'s request.

12. The method of claim 10, wherein the one or more textual structures includes at least one grammatically correct sentence.

13. A module being executed on a computer to construct at least one grammatically correct language structure from at least one structure interpretation of parsed web text,

wherein the at least one structure interpretation comprises a summation of parsed web textual structures based, in part, on underlying grammatical components and definitions of the parsed web text, and

wherein the at least one grammatically correct language structure is created by incorporating the at least one structure interpretation into one or more grammatical templates that include basic grammar confines.

14. The module of claim 13, wherein the module uses predefined textual structures to construct the at least one grammatically correct language structure.

15. The module of claim 13, wherein the module uses a statistical analysis methodology to construct the at least one grammatically correct language structure.