DOCUMENT CLUSTERING AND RECONSTRUCTION

• US 20140280167A1
• Filed: 02/28/2014
• Published: 09/18/2014
• Est. Priority Date: 03/14/2013
• Status: Active Grant

• Current Assignee: Digitech Systems Private Reserve, LLC
• Original Assignee: Digitech Systems Private Reserve, LLC
• Inventors: Ghessassi, Karim
• Granted Patent: US 11,126,839 B2
• Time in Patent Office: Days
• US Class Current: 707/738
• CPC Class Codes: G06F 16/285 Clustering or classification
• CPC Class Codes: G06F 16/93 Document management systems
• CPC Class Codes: G06F 40/186 Templates
• CPC Class Codes: G06V 30/418 Document matching, e.g. of document images

Abstract

A scanner scans a group of documents. For example, the documents can be a group of invoices. The documents are received and processed. Objects (e.g., a text object, such as a word) and their locations are identified in each of the documents. Occurrences of similar objects in the identified locations between the documents are determined. A document sorting algorithm is applied to generate a score for each of the documents. The score for each of the documents is generated based on a number of occurrences of similar objects between the documents. The generated score of each of the documents is used to identify a template document. The template document is then used to cluster the documents.

Claims

1. A method comprising:

receiving a plurality of documents;

for each of the plurality of documents, identifying a plurality of objects and locations of each of the plurality of objects;

determining occurrences of similar objects in the identified locations of the plurality of objects between the plurality of documents;

applying a document sorting algorithm to generate a score for each of the plurality of documents, wherein the score for each of the plurality of documents is generated based on a number of occurrences of similar objects between the plurality of documents; and

comparing the generated score of each of the plurality of documents to identify a template document.

2. The method of claim 1, further comprising:

determining an amount of certainty for an occurrence of similar objects in a common object document location between the plurality of documents;

identifying the common object document location based on a minimum certainty threshold value;

determining that the template document contains an error for an individual object in the common object document location in the template document; and

in response to determining that the template document contains an error for the individual object in the common object document location in the template document, replacing the individual object in the common object document location in the template document with a second object, wherein the second object is from the common object location in a second one of the plurality of documents that has been determined to be correct.

3. The method of claim 2, wherein the error for the individual object in the template document is determined based on a number of occurrences of the similar objects in the common document location in the plurality of documents.

4. The method of claim 2, wherein each of the plurality of documents comprises at least two separate similar objects that are at different common locations between the plurality of documents and wherein the amount of certainty is determined based on one of:

the two separate objects;

or

a single one of the two separate objects.

5. The method of claim 1, wherein the similar objects comprises at least one of:

a text object in an unknown language, an object that is part of a computer programming language, a phrase, a number, a punctuation mark, a text object, a graphical object, a logo, and a picture.

6. The method of claim 1, wherein the common locations are determined based on at one or more of a distance, a relative distance, a relative angle, a character distance, a word distance, and line distance.

7. The method of claim 1, wherein the similar objects are text objects and wherein the plurality of documents are received from a scanner.

8. The method of claim 1, wherein the document sorting algorithm generates the score for each of the plurality of documents by at least one of:

summing the number of occurrences of similar objects between the plurality of documents; and

multiplying the number of occurrences of similar objects between the plurality of documents.

9. The method of claim 1, wherein the template document is used to cluster the plurality of documents based on the template document.

10. The method of claim 1, wherein determining occurrences of the similar objects in the identified locations of the plurality of objects between the documents further comprises recalculating the identified locations based on a misalignment of the identified locations due to a use of at least one of a different font and a different font size.

11. A system comprising:

a document processor configured to receive a plurality of documents, for each of the plurality of documents, identify a plurality of objects and locations of each of the plurality of objects, determining occurrences of similar objects in the identified locations of the plurality of objects between the plurality of documents, and apply a document sorting algorithm to generate a score for each of the plurality of documents, wherein the score for each of the plurality of documents is generated based on a number of occurrences of similar objects between the plurality of documents; and

a document classifier configured to compare the generated score of each of the plurality of documents to identify a template document.

12. The system of claim 11, wherein the document processor is further configured to determine an amount of certainty for an occurrence of similar objects in a common object document location between the plurality of documents, identify the common object document location based on a minimum certainty threshold value, determine that the template document contains an error for an individual object in the common object document location in the template document, and replace the individual object in the common object document location in the template document with a second object in response to determining that the template document contains an error for the individual object in the common object document location in the template document, wherein the second object is from the common object location in a second one of the plurality of documents that has been determined to be correct.

13. The system of claim 12, wherein the error for the individual object in the template document is determined based on a number of occurrences of the similar objects in the common document location in the plurality of documents.

14. The system of claim 12, wherein each of the plurality of documents comprises at least two separate similar objects that are at different common locations between the plurality of documents and wherein the amount of certainty is determined based on one of:

the two separate objects;

or

a single one of the two separate objects.

15. The system of claim 11, wherein the similar objects comprises at least one of:

a text object in an unknown language, an object that is part of a computer programming language, a phrase, a number, a punctuation mark, a text object, a graphical object, a logo, and a picture.

16. The method of claim 11, further comprising a scanner that generates the plurality of documents and wherein the similar objects are text objects.

17. The system of claim 11, wherein the document sorting algorithm generates the score for each of the plurality of documents by at least one of:

summing the number of occurrences of similar objects between the plurality of documents; and

multiplying the number of occurrences of similar objects between the plurality of documents.

18. The system of claim 11, wherein the template document is used to cluster the plurality of documents based on the template document.

19. The system of claim 11, wherein document processor is further configured to recalculate the identified locations based on a misalignment of the identified locations due to a use of at least one of a different font and a different font size.

20. A system comprising:

a scanner configured to scan a plurality of documents;

a document processor configured to receive the plurality of documents,

for each of the plurality of documents, identify a plurality of objects and locations of each of the plurality of objects, determine occurrences of similar objects in the identified locations of the plurality of objects between the plurality of documents, apply a document sorting algorithm to generate a score for each of the plurality of documents, wherein the score for each of the plurality of documents is generated based on a number of occurrences of similar objects between the plurality of documents, determine an amount of certainty for an occurrence of similar objects in a common object document location between the plurality of documents, identify the common object document location based on a minimum certainty threshold value, determine that the template document contains an error for an individual object in the common object document location in the template document, and replace the individual object in the common object document location in the template document with a second object in response to determining that the template document contains an error for the individual object in the common object document location in the template document, wherein the second object is from the common object location in a second one of the plurality of documents that has been determined to be correct; and

a document classifier configured to compare the generated score of each of the plurality of documents to identify a template document and cluster the plurality of documents based on the template document.