IP Library Granted Patent US 12,093,271
Granted Patent B2
US 12,093,271 · App. 17/489,533 · Granted Sep 17, 2024

Method and system for text understanding in an ontology driven platform

Inventor: Parsa Mirhaji (Pearland, TX)
Assignee: Board of Regents, The University of Texas System
G06F16/2465G06F16/22G06F16/248G06F16/3329G06F16/3344G06F16/367G06F16/9024G06F40/211G06F40/253G06F40/284G06F40/289G06N5/02
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,093,271
App. No.
17/489,533
Granted
Sep 17, 2024
Kind
B2
Abstract

Embodiments of methods and systems for informatics systems are disclosed. Such informatics systems may utilize a unifying format to represent text to facilitate linking between data from the text and one or more ontologies, and the commensurate ability to mine such data.

Claims (47)

1. A system, comprising:

a processor; and

a non-transitory computer readable medium comprising instructions for:

receiving text as input from a data source;

creating a parse graph of the text, wherein creating the parse graph comprises:

parsing the text by segmenting the text into a set of evidence spaces based on one or more identifiers within the text;

ordering the set of evidence spaces;

chunking each of the set of evidence spaces into one or more chunks, each chunk comprising a permutation of tokens of an associated evidence space within a predetermined distance of one another in the associated evidence space, wherein chunking the associated evidence space into the one or more chunks utilizes a moving window of the predetermined distance to iterate over tokens of the associated evidence space, and an iteration over a token of the associated evidence space comprises generating one or more chunks based on the permutation of a set of the tokens of the associated evidence space within the predetermined distance of the token; and

based on the ordering and chunking, creating nodes and relationships of the parse graph representing a sequence of the evidence spaces and chunks within each evidence space.

2. The system of claim 1 , wherein the nodes of and relationships of the parse graph represent dependencies of the evidence spaces and chunks within each evidence space.

3. The system of claim 2 , wherein the parse graph is a directed graphs with a non-hierarchical structure.

4. The system of claim 3 , wherein the parse graph maintains an index of tokens and positional information from the text.

5. The system of claim 1 , wherein the instructions are further for:

obtaining a representation of a domain ontology, wherein the domain ontology comprises a set of concepts and a set of relationships; and

mapping the parse graph to the representation of the domain ontology to create a unified graph comprising the graph representation of the text and the representation of the domain ontology.

6. The system of claim 5 , wherein mapping the parse graph to the representation of the domain ontology comprises matching a concept associated with the parse graph to one or more of the set of concepts of the domain ontology.

7. The system of claim 1 , wherein the chunking is based on a based on a position of syntactic concepts associated with the evidence space.

8. A method, comprising:

receiving text as input from a data source;

creating a parse graph of the text, wherein creating the parse graph comprises:

parsing the text by segmenting the text into a set of evidence spaces based on one or more identifiers within the text;

ordering the set of evidence spaces;

chunking each of the set of evidence spaces into one or more chunks, each chunk comprising a permutation of tokens of an associated evidence space within a predetermined distance of one another in the associated evidence space, wherein chunking the associated evidence space into the one or more chunks utilizes a moving window of the predetermined distance to iterate over tokens of the associated evidence space, and an iteration over a token of the associated evidence space comprises generating one or more chunks based on the permutation of a set of the tokens of the associated evidence space within the predetermined distance of the token; and

based on the ordering and chunking, creating nodes and relationships of the parse graph representing a sequence of the evidence spaces and chunks within each evidence space.

9. The method of claim 8 , wherein the nodes of and relationships of the parse graph represent dependencies of the evidence spaces and chunks within each evidence space.

10. The method of claim 9 , wherein the parse graph is a directed graphs with a non-hierarchical structure.

11. The method of claim 10 , wherein the parse graph maintains an index of tokens and positional information from the text.

12. The method of claim 8 , further comprising:

obtaining a representation of a domain ontology, wherein the domain ontology comprises a set of concepts and a set of relationships; and

mapping the parse graph to the representation of the domain ontology to create a unified graph comprising the graph representation of the text and the representation of the domain ontology.

13. The method of claim 12 , wherein mapping the parse graph to the representation of the domain ontology comprises matching a concept associated with the parse graph to one or more of the set of concepts of the domain ontology.

14. The method of claim 8 , wherein the chunking is based on a based on a position of syntactic concepts associated with the evidence space.

15. A non-transitory computer readable medium, comprising instructions for:

receiving text as input from a data source;

creating a parse graph of the text, wherein creating the parse graph comprises:

parsing the text by segmenting the text into a set of evidence spaces based on one or more identifiers within the text;

ordering the set of evidence spaces;

chunking each of the set of evidence spaces into one or more chunks, each chunk comprising a permutation of tokens of an associated evidence space within a predetermined distance of one another in the associated evidence space, wherein chunking the associated evidence space into the one or more chunks utilizes a moving window of the predetermined distance to iterate over tokens of the associated evidence space, and an iteration over a token of the associated evidence space comprises generating one or more chunks based on the permutation of a set of the tokens of the associated evidence space within the predetermined distance of the token; and

based on the ordering and chunking, creating nodes and relationships of the parse graph representing a sequence of the evidence spaces and chunks within each evidence space.

16. The non-transitory computer readable medium of claim 15 , wherein the nodes of and relationships of the parse graph represent dependencies of the evidence spaces and chunks within each evidence space.

17. The non-transitory computer readable medium of claim 16 , wherein the parse graph is a directed graphs with a non-hierarchical structure.

18. The non-transitory computer readable medium of claim 16 , wherein the parse graph maintains an index of tokens and positional information from the text.

19. The non-transitory computer readable medium of claim 15 , further comprising instructions for:

obtaining a representation of a domain ontology, wherein the domain ontology comprises a set of concepts and a set of relationships; and

mapping the parse graph to the representation of the domain ontology to create a unified graph comprising the graph representation of the text and the representation of the domain ontology.

20. The non-transitory computer readable medium of claim 19 , wherein mapping the parse graph to the representation of the domain ontology comprises matching a concept associated with the parse graph to one or more of the set of concepts of the domain ontology.

21. The non-transitory computer readable medium of claim 15 , wherein the chunking is based on a based on a position of syntactic concepts associated with the evidence space.

Assignments (2)
CONFIRMATORY LICENSE Recorded Mar 7, 2023
From: UNIVERSITY OF TEXAS HEALTH SCI CTR HOUSTON
To: UNITED STATES GOVERNMENT
Reel/Frame 062967/0401 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Oct 12, 2021
From: MIRHAJI, PARSA
To: THE BOARD OF REGENTS OF THE UNIVERSITY OF TEXAS SYSTEM
Reel/Frame 057762/0420 →
Continuity (8)
Continuation 16540823 · Aug 14, 2019
Continuation 15335205 · Oct 26, 2016
Continuation 13777393 · Feb 26, 2013
Continuation 12928489 · Dec 13, 2010
Provisional Application 61284330 · Dec 16, 2009
Provisional Application 61284331 · Dec 16, 2009
Provisional Application 61284332 · Dec 16, 2009
Related Publication 20220019580A1 · Jan 20, 2022