Document-centric system with auto-completion and auto-correction

• US 20030033288A1
• Filed: 12/05/2001
• Published: 02/13/2003
• Est. Priority Date: 08/13/2001
• Status: Active Grant

• Current Assignee: III Holdings 6, LLC
• Original Assignee: Xerox Corporation (Xerox Holdings Corp.)
• Inventors: Shanahan, James G., Grefenstette, Gregory T.
• Granted Patent: US 6,820,075 B2
• Time in Patent Office: Days
• US Class Current: 707/3
• CPC Class Codes: G06F 16/3322 using system suggestions G06F16/3325 takes precedence
• CPC Class Codes: G06F 16/3344 using natural language analysis
• CPC Class Codes: G06F 16/38 Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
• CPC Class Codes: Y10S 707/917 Text
• CPC Class Codes: Y10S 707/955 Object-oriented
• CPC Class Codes: Y10S 707/99932 Access augmentation or optimizing
• CPC Class Codes: Y10S 707/99933 Query processing, i.e. searching
• CPC Class Codes: Y10S 707/99942 Manipulating data structure, e.g. compression, compaction, compilation

Abstract

An information space is created using a document. Entities from the document and its information space are used to create a database of entities. An auto-completion system uses contextual information surrounding a fragment from the document to formulate a query. The query is used to identify a set of entities in the database of entities that complete the fragment. An auto-correction system uses contextual information from identified errors in the document to formulate a query. The query is used to identify a set of entities in the database of entities that correct the error.

Claims

1. A method for auto-completing document content, comprising:

receiving a signal specifying an auto-completion request;

the auto-completion request including an entity fragment of a target document;

analyzing content surrounding the entity fragment in the target document to provide context information for identifying a first document attribute;

defining a query using the entity fragment and the first document attribute;

accessing a database of entities using the query to identify a set of entities that satisfy the auto-completion request;

the database of entities including entities and entity context information;

the entity context information identifying a second document attribute;

wherein said accessing compares the first document attribute and the second document attribute to determine a degree of match between the entity fragment and the entities in the database of entities.

2. The method according to claim 1, wherein the auto-completion request is satisfied when the degree of match between the first document attribute and the second document attribute is above a predefined threshold.

3. The method according to claim 1, wherein the document attribute is one of a part of speech and a class of document.

4. The method according to claim 1, further comprising:

initializing the database with entities using the target document; and

augmenting the database with entities in an information space of the target document.

5. The method according to claim 4, wherein the information space of the target document is defined using a personality.

6. The method according to claim 4, further comprising identifying those entities in the database determined to have a greater degree of match than others.

7. The method according to claim 4, wherein the database of entities is defined using a meta-document server.

8. The method according to claim 1, wherein the signal originates from one of a user, an OCR system, and a spell checker.

9. The method according to claim 1, wherein the database further comprises part of speech information regarding the expanded fragments.

10. The method according to claim 1, wherein the set of entities defines a generic object.

11. A method for auto-completing document content, comprising:

defining an information space for target document content;

creating a database of entities using the information space for the target document content;

said creating adding entities to the database of entities using the target document content;

receiving an auto-completion request that includes an entity fragment of the target document;

analyzing content surrounding the entity fragment in the target document to provide associated context information;

formulating a query using both the entity fragment of the target document and its associated context information;

using the query to identify a set of entities in the database of entities that satisfy the auto-completion request.

12. The method according to claim 11, further comprising:

updating the information space of the target document; and

propagating changes of the information space to the database of entities.

13. The method according to claim 11, further comprising initializing the database of entities using identified entities in the target document.

14. The method according to claim 11, wherein the associated context information is one of a document classification and a part of speech.

15. The method according to claim 11, wherein the query is formulated using n-grams of the entity fragment of the target document and wherein the set of entities identified using the query defines a generic object.

16. A method for auto-correcting document content, comprising:

(a) defining an information space using the document content;

(b) creating a database of entities using the information space;

said creating adding entities to the database of entities using the document content;

(c) identifying errors in the document content;

(d) formulating a query using the identified errors;

(e) identifying a set of entities in the database of entities that satisfy the query;

(f) correcting the document content using the identified set of entities;

(g) updating the information space with the corrected document content.

17. The method according to claim 16, repeating (c) (g) until said identifying identifies less than a threshold number of errors.

18. The method according to claim 17, wherein said correcting further comprises:

receiving a request to correct the document content; and

suggesting corrections to the document content using the identified set of entities;

wherein said suggesting ranks the set of entities by highest probability of occurrence.

19. The method according to claim 18, further comprising converting the document content to text data.

20. The method according to claim 16, wherein the identified set of entities defines a generic object.