Method and system for personal information extraction and modeling with fully generalized extraction contexts

• US 7,949,629 B2
• Filed: 10/29/2007
• Issued: 05/24/2011
• Est. Priority Date: 10/30/2006
• Status: Expired due to Fees

• Current Assignee: Noblis, Inc.
• Original Assignee: Noblis, Inc.
• Inventors: Pollara, Victor J.
• Primary Examiner(s): Corrielus; Jean M
• Assistant Examiner(s): Ruiz; Angelica
• Application Number:
  US11/976,818
• Publication Number:
  US 20080133213A1
• Time in Patent Office: 1,303 Days
• Field of Search: None
• US Class Current: 707/602
• CPC Class Codes: G06F 16/36 Creation of semantic tools, e.g. ontology or thesauri

Abstract

Systems and methods for modeling information from a set of documents are disclosed. A tool allows a user to model concepts of interest and extract information from a set of documents in an editable format. The extracted information includes a list of instances of a document from the set of documents that contains the selected concept. The user may modify the extracted information to create subsets of information, add new concepts to the model, and share the model with others.

Claims

1. A method for modeling information implemented using a computer having a processor and a display from a set of documents, comprising:

importing a set of concepts;

creating a model including the concepts;

based on a user selection of a first concept, extracting first information from the set of documents using an extractor corresponding to the first concept, the first information comprising a list of documents containing the first concept;

based on the first information, defining a context of the first concept comprising a contiguous block of text in proximity of the first concept in each document in the list of documents containing the first concept;

based on a user selection of a second concept, extracting second information from the set of documents, the second information comprising a list of documents containing the second concept in the defined context of the first concept;

adding a new additional concept to the model wherein the new additional concept represents the second concept in the defined context of the first concept; and

extracting information related to the new additional concept from the set of documents using an extractor corresponding to the new additional concept.

2. The method of claim 1, wherein the list of documents containing the first concept comprises a document identifier for each of the documents.

3. The method of claim 1, wherein

the extracted information related to the new additional concept comprises a list of documents containing the second concept in the defined context of the first concept.

4. The method of claim 1, wherein the first information further comprises an excerpt from each of the documents containing the first concept.

5. The method of claim 1, further comprising:

based on a user selection of a third concept, extracting third information from the set of documents, the third information comprising a list of documents containing the third concept and the second concept in the defined context of the first concept.

6. The method of claim 1, further comprising:

automatically generating a database customized to the model.

7. The method of claim 1, further comprising:

selecting the set of concepts based on the set of documents.

8. The method of claim 1, further comprising:

determining a trigger phrase related to the set of documents; and

creating a second new additional concept based on the trigger phrase.

9. The method of claim 1, further comprising:

exporting the model including the new additional concept to another computer.

10. A method for personalizing a model of information implemented using a computer having a processor and a display, comprising:

obtaining a set of documents;

receiving the model that contains concepts;

extracting first information from the set of documents using an extractor corresponding to a first concept, the first information comprising a list of documents containing the first concept;

based on the first information, defining a context of the first concept comprising a contiguous block of text in proximity of the first concept in each document in the list of documents containing the first concept;

selecting a second concept to update;

updating the second concept;

extracting second information from the set of documents, the second information comprising a list of documents containing the updated second concept in the defined context of the first concept;

adding a new additional concept to the model wherein the new additional concept represents the updated second concept in the defined context of the first concept; and

extracting information related to the new additional concept from the set of documents using an extractor corresponding to the new additional concept.

11. The method of claim 10, wherein the defined context of the first concept further comprises each document in the list of documents containing the first concept.

12. The method of claim 10, wherein the defined context of the first concept further comprises a subset of the documents containing the first concept.

13. The method of claim 10, wherein the defined context of the first concept further comprises a section of a document in the list of documents containing the first concept.

14. The method of claim 10, wherein the defined context of the first concept further comprises a status of a document in the list of documents containing the first concept.

15. The method of claim 10, further comprising:

automatically assigning an extractor to the new additional concept.

16. The method of claim 10, further comprising:

exporting the model including the new additional concept.

17. A method for annotating a set of documents in a model of information implemented using a computer having a processor and a display, comprising:

downloading a set of documents;

receiving the model, wherein the model contains concepts;

extracting first information from the set of documents using an extractor corresponding to a first concept, the first information comprising a list of documents containing the first concept;

calculating a status of a document in the set of documents;

after extracting the first information, automatically annotating a document in the set of documents based on the calculation;

based on the first information, defining a context of the first concept comprising a contiguous block of text in proximity of the first concept in each document in the list of documents containing the first concept;

extracting second information from the set of documents, the second information comprising a list of documents containing a second concept in the defined context of the first concept;

adding a new additional concept to the model wherein the new additional concept represents the second concept in the defined context of the first concept; and

extracting information related to the new additional concept from the set of documents using an extractor corresponding to the new additional concept.

18. The method of claim 17, further comprising:

displaying, on the display device, the model using an entity-relationship diagram, wherein the concepts are depicted as entities and relationships between the concepts are depicted as relations between entities.

19. The method of claim 17, wherein calculating is based on a function provided by a user.

20. The method of claim 17, further comprising:

receiving a definition for the function from the user.

21. A method for modeling information implemented using a computer having a processor and a display from a set of documents, comprising:

importing a set of concepts and corresponding extractors;

creating a model including a first concept and a second concept;

establishing a definitional dependency between the first concept and the second concept;

based on a user selection of the first concept, extracting first information from the set of documents using an extractor corresponding to the first concept, the first information comprising a list of documents containing the first concept;

based on the first information, defining a context of the first concept comprising a contiguous block of text in proximity of the first concept in each document in the list of documents containing the first concept;

after extracting information related to the first concept, receiving a modification of the second concept;

after receiving the modification of the second concept, updating the first information using the definitional dependency between the first concept and the second concept;

adding a new additional concept to the model wherein the new additional concept represents the updated first information using the definitional dependency between the first concept and the second concept; and

based on a user selection of the new additional concept, extracting information related to the new additional concept from the set of documents using an extractor corresponding to the new additional concept.

22. The method of claim 21, further comprising:

receiving an additional document for the set of documents after extracting information related to the first concept; and

updating the first information based on the additional document.

23. The method of claim 21, further comprising:

notifying a user of the modification of the second concept; and

receiving a request from the user to update the first information.

24. A system for modeling information from a set of documents, comprising:

a processor;

an extraction engine configured in the processor to extract information from the set of documents;

software components comprising;

an importing component configured to import a set of concepts,

a modeling component configured to create a model including the concepts,

a first extraction component configured to cause the extraction engine to extract first information from the set of documents using an extractor corresponding to the first concept, the first information comprising a list of documents containing a first concept,

a context component for, based on the first information, defining a context of the first concept comprising a contiguous block of text in proximity of the first concept in each document in the list of documents containing the first concept,

a second extraction component configured to cause the extraction engine to extract second information from the set of documents, the second information comprising a list of documents containing a second concept in the defined context of the first concept,

a context concept component configured to add a new additional concept to the model wherein the new additional concept represents the second concept in the defined context of the first concept, and

a context extraction component configured to cause the extraction engine to extract information related to the new additional concept from the set of documents using an extractor corresponding to the new additional concept; and

a database configured to store the model.