Method and system for classifying documents that have different scales

• US 8,832,108 B1
• Filed: 04/18/2013
• Issued: 09/09/2014
• Est. Priority Date: 03/28/2012
• Status: Active Grant

• Current Assignee: Open Text Corporation
• Original Assignee: EMC Corporation (Dell Technologies Inc.)
• Inventors: Sampson, Steven
• Primary Examiner(s): Nguyen, Phong
• Application Number:
  US13/865,350
• Time in Patent Office: 509 Days
• Field of Search: 707/737, 707/749, 707/758
• US Class Current: 707/737
• CPC Class Codes: G06V 30/196 using sequential comparisons of the image signals with a plurality of references
• CPC Class Codes: G06V 30/414 Extracting the geometrical structure, e.g. layout tree; Block segmentation, e.g. bounding boxes for graphics or text
• CPC Class Codes: G06V 30/418 Document matching, e.g. of document images

Abstract

Classifying documents that have different scales is described. Instances are counted for each character size in documents. Character sizes for the first document and the second document are selected based on the instance count for each character size. Scales are calculated based on ratios of each first character size relative to each second character size. Scale products are calculated based on each instance count for each character size range for the first character sizes multiplied by each instance count for each corresponding character size range for the second character sizes. The corresponding character size range is based on a corresponding scale. Scale scores are calculated based on summing each of the scale products for each scale. A scale is selected based a highest scale score. The second document may be classified with the first document based on a comparison of first document location information and second document location information. The second document location information is based on the scale.

Claims

1. A system for classifying documents that have different scales, the system comprising:

one or more processors; and

a non-transitory computer readable medium storing a plurality of instructions, which when executed, cause the one or more processors to;

count instances for each character size in a first document and instances for each character size in a second document;

select a first plurality of character sizes for the first document and a second plurality of character sizes for the second document, based on a corresponding count of instances associated with each corresponding character size;

calculate a plurality of scales, wherein each scale of the plurality of scales is based on a corresponding ratio of a corresponding one of the first plurality of character sizes relative to a corresponding one of the second plurality of character sizes;

calculate a plurality of scale products based on each corresponding count of instances for each character size range associated with the first plurality of character sizes multiplied by each corresponding count of instances for each corresponding character size range associated with the second plurality of character sizes, wherein the corresponding character size range is based on a corresponding one of the plurality of scales;

calculate a plurality of scale scores based on summing each of the plurality of scale products associated with each corresponding one of the plurality of scales;

select a scale of the plurality of scales based a highest one of the plurality of scale scores associated with a corresponding one the plurality of scales;

determine whether the second document is in a class associated with the first document based on a comparison of location information associated with the first document and location information associated with the second document, wherein the location information associated with second document is based on the scale; and

classify the second document in the class associated with the first document in response to a determination that the second document is in the class associated with the first document.

2. The system of claim 1, wherein the first document and the second document comprise digitized optical character recognition data.

3. The system of claim 1, further comprising instructions, which when executed, cause the one or more processors to classify the second document in a class associated with another document in response to a determination that the second document is not in the class associated with the first document.

4. The system of claim 1, wherein the comparison of the location information associated with the first document and the location information associated with the second document comprises:

generating a plurality of word pairs, wherein each word pair comprises a word from the first document and a corresponding word from the second document;

computing, for each word pair, first location information for the word that indicates a location of the word in the first document relative to other words in the first document;

computing, for each word pair, second location information for the corresponding word that indicates a location of the corresponding word in the second document relative to other words in the second document; and

comparing the first location information and the second location information.

5. The system of claim 4, wherein the word pairs comprise keywords associated with the second document based on a comparison of the second document with at least one of a class and a template.

6. The system of claim 1, wherein the second document location information is also based on at least one of a scale associated with a next highest one of the plurality of scale scores and a scale of one.

7. The system of claim 1, wherein the first document is associated with a template in response to a comparison to classify documents similar to a document associated with the template.

8. A computer-implemented method for classifying documents that have different scales, the method comprising:

counting, by a server computer, instances for each character size in a first document and instances for each character size in a second document;

selecting, by the server computer, a first plurality of character sizes for the first document and a second plurality of character sizes for the second document, based on a corresponding count of instances associated with each corresponding character size;

calculating, by the server computer, a plurality of scales, wherein each scale of the plurality of scales is based on a corresponding ratio of a corresponding one of the first plurality of character sizes relative to a corresponding one of the second plurality of character sizes;

calculating, by the server computer, a plurality of scale products based on each corresponding count of instances for each character size range associated with the first plurality of character sizes multiplied by each corresponding count of instances for each corresponding character size range associated with the second plurality of character sizes, wherein the corresponding character size range is based on a corresponding one of the plurality of scales;

calculating, by the server computer, a plurality of scale scores based on summing each of the plurality of scale products associated with each corresponding one of the plurality of scales;

selecting, by the server computer, a scale of the plurality of scales based a highest one of the plurality of scale scores associated with a corresponding one the plurality of scales;

determining, by the server computer, whether the second document is in a class associated with the first document based on a comparison of location information associated with the first document and location information associated with the second document, wherein the location information associated with second document is based on the scale; and

classifying, by the server computer, the second document in the class associated with the first document in response to a determination that the second document is in the class associated with the first document.

9. The computer-implemented method of claim 8, wherein the first document and the second document comprise digitized optical character recognition data.

10. The computer-implemented method of claim 8, wherein the further comprises classifying the second document in a class associated with another document in response to a determination that the second document is not in the class associated with the first document.

11. The computer-implemented method of claim 8, wherein the comparison of the location information associated with the first document and the location information associated with the second document comprises:

generating a plurality of word pairs, wherein each word pair comprises a word from the first document and a corresponding word from the second document;

computing, for each word pair, first location information for the word that indicates a location of the word in the first document relative to other words in the first document;

computing, for each word pair, second location information for the corresponding word that indicates a location of the corresponding word in the second document relative to other words in the second document; and

comparing the first location information and the second location information.

12. The computer-implemented method of claim 11, wherein the word pairs comprise keywords associated with the second document based on a comparison of the second document with at least one of a class and a template.

13. The computer-implemented method of claim 8, wherein the second document location information is also based on at least one of a scale associated with a next highest one of the plurality of scale scores and a scale of one.

14. The computer-implemented method of claim 8, wherein the first document is associated with a template in response to a comparison to classify documents similar to a document associated with the template.

15. A computer program product, comprising a non-transitory computer-readable medium having a computer-readable program code embodied therein to be executed by one or more processors, the program code including instructions to:

count instances for each character size in a first document and instances for each character size in a second document;

select a first plurality of character sizes for the first document and a second plurality of character sizes for the second document, based on a corresponding count of instances associated with each corresponding character size;

calculate a plurality of scales, wherein each scale of the plurality of scales is based on a corresponding ratio of a corresponding one of the first plurality of character sizes relative to a corresponding one of the second plurality of character sizes;

calculate a plurality of scale products based on each corresponding count of instances for each character size range associated with the first plurality of character sizes multiplied by each corresponding count of instances for each corresponding character size range associated with the second plurality of character sizes, wherein the corresponding character size range is based on a corresponding one of the plurality of scales;

calculate a plurality of scale scores based on summing each of the plurality of scale products associated with each corresponding one of the plurality of scales;

select a scale of the plurality of scales based a highest one of the plurality of scale scores associated with a corresponding one the plurality of scales;

determine whether the second document is in a class associated with the first document based on a comparison of location information associated with the first document and location information associated with the second document, wherein the location information associated with second document is based on the scale; and

classify the second document in the class associated with the first document in response to a determination that the second document is in the class associated with the first document.

16. The computer program product of claim 15, wherein the first document and the second document comprise digitized optical character recognition data.

17. The computer program product of claim 15, wherein the further comprises classifying the second document in a class associated with another document in response to a determination that the second document is not in the class associated with the first document.

18. The computer program product of claim 15, wherein the comparison of the location information associated with the first document and the location information associated with the second document comprises:

generating a plurality of word pairs, wherein each word pair comprises a word from the first document and a corresponding word from the second document;

computing, for each word pair, first location information for the word that indicates a location of the word in the first document relative to other words in the first document;

computing, for each word pair, second location information for the corresponding word that indicates a location of the corresponding word in the second document relative to other words in the second document; and

comparing the first location information and the second location information.

19. The computer program product of claim 18, wherein the word pairs comprise keywords associated with the second document based on a comparison of the second document with at least one of a class and a template.

20. The computer program product of claim 15, wherein the second document location information is also based on at least one of a scale associated with a next highest one of the plurality of scale scores and a scale of one.