Navigating electronic documents using domain discourse trees

Patent metadata

Field	Value
Publication number	US-10853574-B2
Application number	US-201816145644-A
Country	US
Kind code	B2
Filing date	Sep 28, 2018
Priority date	Sep 28, 2017
Publication date	Dec 1, 2020
Grant date	Dec 1, 2020

Abstract

Systems, devices, and methods of the present invention relate to extended discourse trees and using extended discourse trees to navigate text. In an example, a discourse navigation application creates a first discourse tree for a first paragraph of a first document and a second discourse tree for a second paragraph of a second document. The application determines an entity and a corresponding first elementary discourse unit from the first discourse tree. The application determines, in the second discourse tree, a second elementary discourse unit that matches the first elementary discourse unit that exists. The application determines a rhetorical relationship between the two elementary discourse units and creates a navigable link between the two discourse trees.

First claim

What is claimed is:

1. A computer-implemented method for determining a rhetorical relationship between one or more documents, the method comprising:
creating a first discourse tree for a first paragraph of a first document, the first paragraph comprising a first plurality of elementary discourse units, and the first discourse tree representing a first rhetorical relationship between at least two of the first plurality of elementary discourse units;
creating a second discourse tree for a second paragraph of a second document, the second paragraph comprising a second plurality of elementary discourse units, and the second discourse tree representing a second rhetorical relationship between at least two of the second plurality of elementary discourse units;
determining that a first elementary discourse unit of the first discourse tree includes an entity by:
extracting a noun phrase from the first elementary discourse unit; and
classifying the noun phrase as including an entity;

determining, in the second discourse tree, a second elementary discourse unit that matches the first elementary discourse unit;
responsive to determining a rhetorical relationship between the first elementary discourse unit in the first discourse tree and the second elementary discourse unit in the second discourse tree, creating an extended discourse tree by linking the first discourse tree and the second discourse tree via the rhetorical relationship;
presenting text associated with the first elementary discourse unit and the rhetorical relationship to a user device; and
responsive to receiving, from a user device, a selection of the rhetorical relationship, presenting text associated with the second elementary discourse unit to the user device.

2. The computer-implemented method of claim 1 , wherein the first plurality of elementary discourse units and the second plurality of elementary discourse units each comprise a verb, and wherein
each of the first discourse tree and the second discourse tree comprises a plurality of nodes, each nonterminal node representing a rhetorical relationship, and each terminal node of the nodes of the respective discourse tree is associated with an elementary discourse unit.

3. The computer-implemented method of claim 1 , wherein the classifying comprises applying a trained machine-learning model to the noun phrase.

4. The computer-implemented method of claim 1 , wherein an entity refers to one of (i) a person, (ii) a company, (iii) a location, (iv) a name of a document, or (v) a date or time.

5. The computer-implemented method of claim 1 , further comprising responsive to not determining a rhetorical relationship, creating a default rhetorical relationship of type elaboration between the first elementary discourse unit and the second elementary discourse unit and linking the first discourse tree and the second discourse tree with the default rhetorical relationship, thereby creating an extended discourse tree.

6. The computer-implemented method of claim 1 , wherein determining the rhetorical relationship further comprises:
combining the first elementary discourse unit and the second elementary discourse unit into a temporary paragraph; and
determining the rhetorical relationship within the temporary paragraph by applying discourse parsing to the temporary paragraph.

7. The computer-implemented method of claim 1 , wherein the entity is represented by either (i) one or more phrases or (ii) one or more elementary discourse units.

8. The computer-implemented method of claim 1 , wherein accessing the first document and the second document comprises determining that a difference between (i) a first content score for the first document and (ii) a second content score for the second document are within a threshold.

9. The computer-implemented method of claim 1 , wherein the first document and the second document are obtained by executing a user query of one or more documents.

10. The computer-implemented method of claim 1 , wherein first document and the second document include text based on a particular topic.

11. The computer-implemented method of claim 1 , further comprising determining, in the first document and the second document, a common subject, topic, or entity.

12. The computer-implemented method of claim 1 , wherein the classifying comprises using a list of keywords or searching an internet resource.

13. A non-transitory computer-readable medium storing computer-executable program instructions, wherein when executed by a processing device, the computer-executable program instructions cause the processing device to perform operations comprising:
creating a first discourse tree for a first paragraph of a first document, the first paragraph comprising a first plurality of elementary discourse units, and the first discourse tree representing a second rhetorical relationship between at least two of the first plurality of elementary discourse units;
creating a second discourse tree for a second paragraph of a second document, the second paragraph comprising a second plurality of elementary discourse units, and the second discourse tree representing a second rhetorical relationship between at least two of the second plurality of elementary discourse units;
determining that a first elementary discourse unit of the first discourse tree includes an entity by:
extracting a noun phrase from the first elementary discourse unit; and
classifying the noun phrase as including an entity;

determining, in the second discourse tree, a second elementary discourse unit that matches the first elementary discourse unit;
responsive to determining a rhetorical relationship between the first elementary discourse unit in the first discourse tree and the second elementary discourse unit in the second discourse tree, creating an extended discourse tree by linking the first discourse tree and the second discourse tree via the rhetorical relationship;
presenting text associated with the first elementary discourse unit and the rhetorical relationship to a user device; and
responsive to receiving, from a user device, a selection of the rhetorical relationship, presenting text associated with the second elementary discourse unit to the user device.

14. The non-transitory computer-readable medium of claim 13 , wherein the first plurality of elementary discourse units and the second plurality of elementary discourse units each comprise a verb, and wherein each of the first discourse tree and the second discourse tree comprises a plurality of nodes, each nonterminal node representing a rhetorical relationship, each terminal node of the nodes of the respective discourse tree is associated with an elementary discourse unit.

15. The non-transitory computer-readable medium of claim 13 , wherein an entity refers to one of (i) a person, (ii) a company, (iii) a location, (iv) a name of a document, or (v) a date or time.

16. The non-transitory computer-readable medium of claim 13 , further comprising responsive to not determining a rhetorical relationship, creating a default rhetorical relationship of type elaboration between the first elementary discourse unit and the second elementary discourse unit and linking the first discourse tree and the second discourse tree with the default rhetorical relationship, thereby creating an extended discourse tree.

17. The non-transitory computer-readable medium of claim 13 , wherein determining the rhetorical relationship further comprises:
combining the first elementary discourse unit and the second elementary discourse unit into a temporary paragraph; and

Assignees

• Oracle Int Corp

Inventors

• Galitsky Boris

Classifications

• G06F40/211 Primary

  Syntactic parsing, e.g. based on context-free grammar [CFG] or unification grammars · CPC title

• G06F16/9014

  hash tables · CPC title

• G06F40/253

  Grammatical analysis; Style critique · CPC title

• G06F16/00

  Information retrieval; Database structures therefor; File system structures therefor · CPC title

• G06N20/00

  Machine learning · CPC title