Method and apparatus for textual exploration discovery

• US 8,265,925 B2
• Filed: 11/15/2002
• Issued: 09/11/2012
• Est. Priority Date: 11/15/2001
• Status: Expired due to Fees

• Current Assignee: Texturgy AS (Solid Media Group AS)
• Original Assignee: Texturgy AS (Solid Media Group AS)
• Inventors: Aarskog, Brit Helle
• Primary Examiner(s): Hudspeth, David R
• Assistant Examiner(s): Spooner, Lamont
• Application Number:
  US10/495,727
• Publication Number:
  US 20050108001A1
• Time in Patent Office: 3,588 Days
• Field of Search: 704/9, 704/1, 704/10, 707 2- 6, 707/706, 707/711, 707/736, 707/741, 707/752, 707/755, 707/778
• US Class Current: 704/9
• CPC Class Codes: G06F 16/248 Presentation of query results
• CPC Class Codes: G06F 40/211 Syntactic parsing, e.g. based on context-free grammar [CFG] or unification grammars
• CPC Class Codes: G06F 40/253 Grammatical analysis; Style critique
• CPC Class Codes: G06F 40/289 Phrasal analysis, e.g. finite state techniques or chunking
• CPC Class Codes: G06F 40/30 Semantic analysis

Abstract

The present invention relates to a method and system for textual exploration and discovery. More specifically, the method and system provide a text-driven and grammar based tool for textual exploration and textual navigation. The facilities for textual exploration and textual navigation are based on a system of index entries that are connected to the underlying text segments from which the index entries are derived. Text units with particular grammatical, semantic, and/or pragmatic features constitute bundles of sentences or text zones.

Claims

1. A method for textual exploration and discovery comprising:

annotating, with a processor, Subject-Verb-Object Structures (SVOS) in a grammatically encoded electronic text; and

wherein said SVOS are used to identify semantic facets termed “

Agent”

, “

Process” and

“

Object”

, i.e. APOS, in a text span, and wherein said semantic facets “

Agent”

, “

Process” and

“

Object”

are provided as index entries in respective window panes on a display unit as contacts point to said electronic text.

2. A method in accordance with claim 1, wherein three separate window panes are provided on the display unit.

3. A method in accordance with claim 2, wherein said triplets are based on a grammatical founded design aiming at supporting exploration and discovery.

4. A method in accordance with claim 3, wherein the grammatical design is based on grammatical annotation.

5. A method in accordance with claim 4, wherein the grammatical annotation is based on part-of-speech tagging (POS-tagging).

6. A method in accordance with claim 4, wherein the grammatical annotation is based on constraint grammars.

7. A method in accordance with claim 1, wherein said triplets are dynamically extracted from a grammatically encoded text.

8. A method in accordance with claim 1, wherein the user after having evaluated a set of contacts can open, and see directly into the text segment from which these contacts are extracted.

9. A method in accordance with claim 1, wherein the SVOS are organized in triplets.

10. A method in accordance with claim 1, wherein the APOS are organized in triplets, termed an “

APO triplet”

.

11. A method in accordance with claim 1, wherein the user can explore the contacts through various options for filtering and sorting.

12. A method in accordance with claim 1, wherein the contacts relates to facets organized according to at least one of grammatical, semantic or pragmatic features in the underlying text.

13. A method in accordance with claim 1, wherein classes of nominal expressions provide semantic roles for subsets of “

Agents” and

“

Objects”

.

14. A method in accordance with claim 1, wherein classes of verbal phrases provide semantic roles for a subset of “

Process”

.

15. A method according to claim 1, wherein the contacts are displayed in an interface in the form of windows arranged for instance side-by-side, each window with options for expansion or reduction, and options for displaying the underlying words as they appear in the text.

16. A method for textual exploration and discovery according to claim 1, wherein;

i) a first window pane contains index entries of Agent structures containing noun units, and

ii) a second window pane contains index entries of Process structures containing verb units, and wherein this window pane is constrained to verb units that follow the noun units displayed in the text pane, and

iii) a third window pane contains index entries of Object structures containing noun units constrained to noun units and verb units displayed in the text pane.

17. Apparatus for textual exploration and discovery, wherein Subject-Verb-Object Structures (SVOS) are annotated in a grammatically encoded electronic text, and wherein said SVOS are used to identify semantic facets termed “

Agent”

, “

Process” and

“

Object”

, i.e. APOS, in a text span, wherein the system comprises;

a processor;

memory; and

a) an acquisition module for collection of documents, and capable of formatting the documents to at least one common format,

b) a segmentation module for the generation of Annotated Text Files (ATF), thus forming the Annotated Text Corpus, and

c) a Disambiguation Module for text disambiguation, and

d) a display unit for presenting said semantic facets “

Agent”

, “

Process” and

“

Object”

as index entries in respective window panes as contacts to said electronic text.

18. Apparatus according to claim 17, wherein the acquisition module are capable of administering, indexing and querying large text corpora.

19. Apparatus according to claim 17, wherein the documents can be annotated with structural information and grammatical information.

20. Apparatus according to claim 17, wherein the module provides a documental link structure, for instance, a group of peripheral documents are linked to a central document, the central documents can be linked to each other, or peripheral documents associated with a central document may also be linked to another central document.

21. Apparatus according to claim 17, wherein the module enables recording of various types of information about the texts, such as document source, collection date, person responsible for collecting it, language, copyright status, format information and version information.

22. Apparatus according to claim 17, wherein the segmentation process includes metadata assignment.

23. Apparatus according to claim 17, wherein the segmentation process applies the Dublin Core Metadata Element Set.

24. Apparatus according to claim 17, wherein a multileveled annotation file system is constructed.

25. Apparatus according to claim 24, wherein the Disambiguation module deals with techniques for converting output from Constraint Grammar taggers (CG-tagger) into an annotation format in compliance with the structure/architecture specified for the Multileveled Annotation File System (MAPS).

26. Apparatus according to claim 25, wherein extracted subsets of grammatical tags (codes) are combined with a selected set of semantic codes.

27. Apparatus according to claim 25, wherein special codes describing different linguistic features are assigned to the words in the texts.

28. Apparatus according to claim 17, wherein the module provides a framework based on triplets in the basic form Subject Verb Object Structures (SVOS).

29. An apparatus according to claim 17, wherein said display unit provides three separate window panes.

30. The apparatus according to claim 29, wherein;

a first window pane contains index entries of Agent structures containing noun units, and

a second window pane contains index entries of Process structures containing verb units, and wherein this window pane is constrained to verb units that follow the noun units displayed in the text pane, and

a third window pane contains index entries of Object structures containing noun units constrained to noun units and verb units displayed in the text pane.