Systems and methods for pairing of a semantic network and a natural language processing information extraction system

• US 9,298,702 B1
• Filed: 04/04/2013
• Issued: 03/29/2016
• Est. Priority Date: 11/18/2008
• Status: Active Grant

• Current Assignee: Semantic AI, Inc.
• Original Assignee: Semantic Research, Inc. (ESW Capital, LLC)
• Inventors: Faletti, Joseph, Stockley, Hilary Anne, Wells, Harold Jason, Murphy, Timothy Andrew
• Primary Examiner(s): Neway, Samuel G
• Application Number:
  US13/856,726
• Time in Patent Office: 1,090 Days
• Field of Search: 704/9, 707/610
• US Class Current: 1/1
• CPC Class Codes: G06F 16/367 Ontology
• CPC Class Codes: G06F 40/30 Semantic analysis
• CPC Class Codes: G06F 40/40 Processing or translation of natural language natural language analysis G06F40/20; semantic analysis G06F40/30

Abstract

Systems and methods for coupling a semantic network editing tool and a natural language processing (NLP) system are disclosed. In one embodiment, a network editing tool includes or is connected to an application module configured to facilitate queries to the NLP system and extraction of data from a collection of document by the NLP system for integration of contents with the semantic network.

Claims

1. A computer implemented method for processing natural language information in conjunction with a semantic network editing tool, the method comprising:

providing a user interface in communication with the semantic network editing tool disposed to facilitate processing of information in a natural language processing (NLP) system from a collection of documents stored in a document repository so as to provide a set of processed information;

communicatively coupling the semantic network editing tool and the NLP system so as to facilitate transfer of data or information between the semantic network editing tool and the NLP system, the communicatively coupling including automatically converting one or more semantic network entity types and associated semantic network templates into NLP extractor definitions wherein the extractor definitions are trained using sample documents prior to being utilized to facilitate the processing of the information in the NLP system;

mapping, using the semantic network editing tool, entity types and relation types of a first ontology of the NLP system to element types and relation types, respectively, of a second ontology of a semantic network, the first ontology being different from the second ontology wherein ones of the element types are associated with ones of the semantic network templates, each of the semantic network templates including a plurality of roles corresponding to a plurality of additional element types;

responsive to a user input provided at the user interface, extracting the set of processed information from the NLP system for use in the semantic network; and

using the semantic network editing tool to view and edit entities and triplets included within the semantic network.

2. The method of claim 1 wherein the coupling includes sharing data describing the second ontology of the semantic network and the first ontology of the NLP system.

3. The method of claim 1 wherein the coupling includes sharing data describing the contents of the semantic network and the NLP system.

4. The method of claim 2 wherein the second ontology of the semantic network and the first ontology of the NLP system are automatically mapped responsive to a change in the content of either the semantic network or the NLP system.

5. The method of claim 2 wherein the second ontology of the semantic network and the first ontology of the NLP system are transformed into corresponding formats in response to a user request received at the user interface.

6. The method of claim 1 wherein the user interface is incorporated in a user interface of the semantic network editing tool.

7. The method of claim 1 wherein the user interface is provided in an application external to the semantic network editing tool.

8. The method of claim 7 wherein the user interface is configured to interface with the semantic network editing tool through one or more application programming interfaces (APIs).

9. The method of claim 1 wherein the user interface is configured to provide one or more options for manual selection of one or more documents of the document repository for information processing in conjunction with the NLP system.

10. The method of claim 1 wherein one or more documents of the document repository are selected automatically for processing in conjunction with the NLP system in response to a user request received at the user interface.

11. The method of claim 1 wherein the coupling includes automatic conversion of one or more NLP extractors into semantic network entity types and their corresponding templates.

12. The method of claim 1 wherein the NLP system includes the functionality of entity recognition and entity type assignment, and further wherein these are provided by a wiki that includes the ability to receive requested pages for word sequences and provide corresponding response data.

13. The method of claim 12 wherein the wiki includes the ability to receive a request for a list of categories or types for any pages found responsive to the page request and provide corresponding response data.

14. The method of claim 1 further comprising providing a second user interface configured to graphically display a graph corresponding to-the semantic network responsive to said extracting.

33. The method of claim 1 wherein the processed information includes a set of triplets associating the entities with each other and with other information within the documents.

34. The method of claim 33 wherein the set of triplets includes one or more groups of triplets relating to one or more user-specified target entities.

35. The method of claim 34 wherein the one or more groups of triplets are selected for inclusion based on one or more relation types prescribed for an element type by the second ontology.

36. The method of claim 33 wherein the set of triplets includes one or more groups of triplets relating to an event.

37. The method of claim 1 wherein the processed information from the NLP system includes at least one of a set of triplets or frames identifying attributes or characteristics of the entities.

38. The method of claim 1 wherein the processed information from the NLP system includes at least one of a judgment, opinion, perspective, or point of view of one of the entities relative to another of the entities.

39. The method of claim 1 wherein the processed information from the NLP system includes at least one of a judgment, opinion, perspective, or point of view of one of the entities relative to an event.

40. The method of claim 1 wherein the processed information includes provenance information concerning one or more of the entities, attachments, or triplets.

41. The method of claim 1 wherein the processed information includes one or more text components from which one of the entities was extracted, the one or more text components being included within the collection of documents.

42. The method of claim 1 wherein the processed information includes one or more text components from which one of the entities was extracted, the one or more text components being included as attachments within the semantic network.

43. The method of claim 1 wherein the processed information includes a summary of the entities and associated triplets.

44. The method of claim 42 wherein the summary of the entities and associated triplets is included as an attachment within the semantic network.

45. The method of claim 43 wherein the summary of the entities and associated triplets is included as a new document in the repository.

46. The method of claim 1, further including:

transforming a representation of a knowledge structure within the semantic network into a document, the knowledge structure containing information relevant to one of the entities;

storing the document within the document repository.

47. The method of claim 46 wherein the document is formatted in a structured text format.

48. The method of claim 46 wherein the document is a tabular formatted file or a graphics formatted file.

49. The method of claim 1, further including:

identifying one or more gaps in relations or entities in the triplets in the semantic network;

submitting a query to the NLP system relating to the gaps.

50. The method of claim 49 wherein the query is accompanied by one or more restrictions on the gaps.

51. The method of claim 50 wherein the one or more gaps comprise one or more triplets with gap entities.

52. The method of claim 1 further including extracting events from the NLP system and mapping the events to elements of the semantic network.

54. The method of claim 1 wherein the coupling includes automatically converting one or more semantic network entity types into NLP extractor definitions wherein at least one of the NLP extractor definitions includes a plurality of role names respectively associated with semantic network relation types and semantic network entity types.

15. A non-transitory machine readable medium including executable instructions for execution on a processor to:

provide a user interface in communication with a semantic network editing tool disposed to facilitate processing of information in a natural language processing (NLP) system from a collection of documents stored in a document repository so as to provide a set of processed information;

communicatively couple the semantic network editing tool and the NLP system so as to facilitate transfer of data or information between the semantic network editing tool and the NLP system;

automatically convert one or more semantic network entity types and associated semantic network templates into NLP extractor definitions wherein the extractor definitions are trained using sample documents prior to being utilized to facilitate the processing of the information in the NLP system;

map, using the semantic network editing tool, entity types and relation types of a first ontology of the NLP system to element types and relation types, respectively, of a second ontology of a semantic network, the first ontology being different from the second ontology wherein ones of the element types are associated with ones of the semantic network templates, each of the semantic network templates including a plurality of roles corresponding to a plurality of additional element types;

responsive to said coupling, extract the set of processed information from the NLP system for use in the semantic network; and

use the semantic network editing tool to view and edit entities and triplets included within the semantic network.

16. The medium of claim 15 wherein the instructions for coupling include instructions for sharing data describing the second ontology of the semantic network and the first ontology of the NLP system.

17. The medium of claim 15 wherein the instructions for coupling include instructions for sharing data describing the contents of the semantic network and the NLP system.

18. The medium of claim 16 wherein the instructions including instructions for automatically synchronizing the second ontology of the semantic network and the first ontology of the NLP system responsive to a change in the content of either the semantic network or the NLP system.

19. The medium of claim 16 wherein the instructions include instructions for transforming the second ontology of the semantic network and the first ontology of the NLP system into corresponding formats in response to a user request received at the user interface.

20. The medium of claim 15 wherein the instructions include instructions for incorporating the user interface into a user interface of the semantic network editing tool.

21. The medium of claim 15 wherein the instructions include instructions to provide the user interface in an application external to the semantic network editing tool.

22. The medium of claim 21 wherein the instructions include instructions to interface the user interface with the semantic network editing tool through one or more application programming interfaces (APIs).

23. The medium of claim 15 wherein the instructions include instructions to provide, in the user interface, one or more options for manual selection of one or more documents of the document repository for information processing.

24. The medium of claim 15 wherein the instructions include instructions to automatically select one or more documents of the document repository for processing in conjunction with the NLP system in response to a user request received at the user interface.

25. The medium of claim 15 wherein instructions for coupling include instructions for automatic conversion of one or more NLP extractors into semantic network entity types and their corresponding templates.

26. The medium of claim 15 wherein the NLP system includes the functionality of entity recognition and entity type assignment, and further wherein these are provided by a wiki that includes the ability to receive requested pages for word sequences and provide corresponding response data.

27. The medium of claim 26 wherein the wiki includes the ability to receive a request for a list of categories or types for any pages found responsive to the page request and provide corresponding response data.

28. The medium of claim 15 further comprising instructions for providing a second user interface configured to graphically display information in the semantic network responsive to said extracting.

53. The non-transitory machine readable medium of claim 15 wherein the instructions include instructions for execution on a processor to extract events from the NLP system and map the events to elements of the semantic network.

29. A system for processing natural language information, comprising:

a processor;

a memory; and

a semantic network editing tool, said semantic network editing tool including one or more modules communicatively coupled with a natural language processing (NLP) system and a document repository to facilitate extraction of processed information from the NLP system for use in a semantic network wherein the semantic network editing tool is disposed to be used to view and edit entities and triplets included within the semantic network and to map entity types and relation types of a first ontology of the NLP system to element types and relation types, respectively, of a second ontology of the semantic network, the first ontology being different from the second ontology;

wherein communicatively coupling the one or more modules with the NLP system includes automatically converting one or more semantic network entity types and associated semantic network templates into NLP extractor definitions and wherein the extractor definitions are trained using sample documents prior to being utilized to facilitate the extraction of the processed information from the NLP system,

wherein ones of the element types are associated with ones of the semantic network templates, each of the semantic network templates including a plurality of roles corresponding to a plurality of additional element types.

30. The system of claim 29 further comprising the document repository.

31. The system of claim 29 further comprising the NLP system.

32. The system of claim 29 further comprising the document repository and the NLP system.