SYSTEM AND METHOD FOR ELECTRONIC DOCUMENT CLASSIFICATION

• US 20090110275A1
• Filed: 09/30/2008
• Published: 04/30/2009
• Est. Priority Date: 10/26/2007
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Ahmed, Abbas
• Granted Patent: US 8,428,367 B2
• Time in Patent Office: Days
• US Class Current: 382/170
• CPC Class Codes: G06F 16/353 into predefined classes
• CPC Class Codes: G06Q 10/107 Computer-aided management of electronic mailing [e-mailing]
• CPC Class Codes: H04L 51/212 using filtering or selective blocking

Abstract

A system and method for electronic document classification are provided. A method in accordance with an embodiment of the present invention includes: converting a candidate electronic document comprising character data to a candidate image; obtaining a representation of a degree of visual similarity of the candidate image to a reference image, the reference image having been obtained by identifying a reference electronic document containing character data representative of a specified classification; and converting the reference electronic document to a reference image.

Claims

1. A method of classifying electronic documents, comprising:

converting a candidate electronic document comprising character data to a candidate image;

obtaining a representation of a degree of visual similarity of the candidate image to a reference image, the reference image having been obtained by identifying a reference electronic document containing character data representative of a specified classification;

and

converting the reference electronic document to a reference image.

2. The method of claim 1, further comprising:

presenting the representation to a user.

3. The method of claim 1, further comprising:

in a case where the degree of visual similarity exceeds a predetermined threshold, classifying the candidate electronic document under the specified classification.

4. The method of claim 3, wherein the reference electronic document is one of a plurality of reference electronic documents, each representative of the specified classification, each of which having been converted to a respective reference image;

and wherein a representation of the degree of visual similarity of each of the respective reference images to the candidate image is obtained; and

in a case where the degree of visual similarity exceeds a predetermined threshold with respect to any one of the respective reference images, classifying the candidate electronic document under the specified classification.

5. The method of claim 1, wherein the candidate electronic document and the reference electronic document are email messages and wherein the specified classification corresponds to unsolicited email messages.

6. The method of claim 1, further comprising:

reducing a resolution of the candidate image to a point where the character data becomes illegible.

7. The method of claim 1, further comprising:

reducing a color depth of the candidate image to a grayscale image.

8. The method of claim 1, wherein obtaining a representation of the degree of visual similarity further comprises:

deriving shape or texture descriptors of the candidate image and comparing the shape or texture descriptors of the candidate image to respective shape or texture descriptors of the reference image.

9. The method of claim 8, wherein deriving shape or texture descriptors of the candidate image further comprises:

deriving an edge histogram of the candidate image, and wherein the comparing further comprises comparing the edge histogram of the candidate image with an edge histogram of the reference image.

10. A computer program product loaded on a computer readable medium, which when executed, classifies electronic documents, comprising program code for:

converting a candidate electronic document comprising character data to a candidate image;

obtaining a representation of a degree of visual similarity of the candidate image to a reference image, the reference image having been obtained by identifying a reference electronic document containing character data representative of a specified classification;

and

converting the reference electronic document to a reference image.

11. A system for classifying electronic documents, comprising:

a system for converting a candidate electronic document comprising character data to a candidate image;

a system for obtaining a representation of a degree of visual similarity of the candidate image to a reference image, the reference image having been obtained by identifying a reference electronic document containing character data representative of a specified classification; and

a system for converting the reference electronic document to a reference image.

12. The system of claim 11, further comprising:

a system for presenting the representation to a user.

13. The system of claim 11, further comprising:

a system for classifying the candidate electronic document under the specified classification, in a case where the degree of visual similarity exceeds a predetermined threshold.

14. The system of claim 13, wherein the reference electronic document is one of a plurality of reference electronic documents, each representative of the specified classification, each of which having been converted to a respective reference image;

and wherein a representation of the degree of visual similarity of each of the respective reference images to the candidate image is obtained; and

in a case where the degree of visual similarity exceeds a predetermined threshold with respect to any one of the respective reference images, classifying the candidate electronic document under the specified classification.

15. The system of claim 11, wherein the candidate electronic document and the reference electronic document are email messages and wherein the specified classification corresponds to unsolicited email messages.

16. The system of claim 11, further comprising:

a system for reducing a resolution of the candidate image to a point where the character data becomes illegible.

17. The system of claim 11, further comprising:

a system for reducing a color depth of the candidate image to a grayscale image.

18. The system of claim 11, wherein the system for obtaining a representation of the degree of visual similarity further comprises:

a system for deriving shape or texture descriptors of the candidate image and comparing the shape or texture descriptors of the candidate image to respective shape or texture descriptors of the reference image.

19. The system of claim 18, wherein the system for deriving shape or texture descriptors of the candidate image further comprises:

a system for deriving an edge histogram of the candidate image, and wherein the system for comparing further comprises a system for comparing the edge histogram of the candidate image with an edge histogram of the reference image.