Method and apparatus for textual exploration discovery

• US 20050108001A1
• Filed: 11/15/2002
• Published: 05/19/2005
• Est. Priority Date: 11/15/2001
• Status: Active Grant

• Current Assignee: Texturgy AS (Solid Media Group AS)
• Original Assignee: Texturgy AS (Solid Media Group AS)
• Inventors: Aarskog, Brit Helle
• Granted Patent: US 8,265,925 B2
• Time in Patent Office: Days
• US Class Current: 704/10
• 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

1-108. -108. (canceled)

109. A method for textual exploration and discovery, wherein a set of interconnected terms with particular grammatical roles is extracted from a grammatically encoded electronic text, wherein the set of interconnected terms is provided on a display unit as contacts to the underlying text.

110. The method of claim 109, wherein the set of interconnected terms contains three terms establishing and displaying three interconnected contacts as a triplet.

111. The method of claim 109, wherein said interconnected terms are based on a grammatical founded design aiming at supporting exploration and discovery.

112. The method of claim 111, wherein the grammatical design is based on grammatical annotation.

113. The method of claim 112, wherein the grammatical annotation is based on part-of-speech tagging (POS-tagging).

114. The method of claim 112, wherein the grammatical annotation is based on constraint grammars.

115. The method of claim 109, wherein said triplets are dynamically extracted from a grammatically encoded text.

116. The method of claim 109, 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.

117. The method of claim 109., wherein the contacts relates to facets termed Subject-Verb-Object Structures (SVOS).

118. The method of claim 117, wherein the SVOS are organized in triplets.

119. The method of claim 109, wherein the contacts relate to facets termed “

Agent”

, “

Process” and

“

Object”

(APOS), and wherein these triplets are derived from the SVOS.

120. The method of claim 119, wherein the APOS are organized in triplets, termed an “

APO triplet”

.

121. The method of claim 109, wherein the user can explore the contacts through various options for filtering and sorting.

122. An apparatus for textual exploration and discovery, wherein a set of interconnected terms with particular grammatical roles is extracted from a grammatically encoded electronic text, wherein the system comprises:

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 where the set of interconnected terms is provided as contacts to the underlying text.

123. The apparatus according to claim 122, wherein the acquisition module are capable of administering, indexing and querying large text corpora.

124. The apparatus according to claim 122, wherein the documents can be annotated with structural information (documental logical object types as title, header, paragraphs, sentence, etc.) and grammatical information (Part-of-speech tags and CG tags).

125. The apparatus according to claim 122, wherein the module provides a documental link structure, for instance, a group of peripheral documents are linked to a central document (first order peripheral-to-central links), the central documents can be linked to each other (hierarchically or in networks), or peripheral documents associated with a central document may also be linked to another central document (second order peripheral-to-central links)

126. The apparatus according to claim 122, 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 (permission is obtained, permission denied or restricted), format information and version information.

127. The apparatus according to claim 122, wherein the segmentation process includes metadata assignment.

128. The apparatus according to claim 122, wherein the segmentation process applies the Dublin Core Metadata Element Set.

129. The apparatus according to claim 122, wherein a multileveled annotation file system is constructed.

130. the apparatus according to claim 122, 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).

131. The apparatus according to claim 122, wherein extracted subsets of grammatical tags (codes) are combined with a selected set of semantic codes.

132. The apparatus according to claim 122, wherein special codes describing different linguistic features are assigned to the words in the texts.

134. The apparatus according to claim 122, wherein the module provides a framework based on triplets in the basic form Subject Verb Object Structures (SVOS).