Method and system for using OCR data for grouping and classifying documents
Document classes for classifying documents are created by comparing the spatial relations of words between a first and second document. If the spatial relations are the same, a document class may be created to classify documents similar to the first and second document. If the spatial relations are different, a first document class may be created to classify documents similar to the first document, and a second document class may be created to classify documents similar to the second document.
1. A system for creating classes for classifying digitized documents, the system comprising:
a processor-based document management system comprising at least one processor coupled to a memory executed on a computer system and configured to:
generate a plurality of word pairs, each word pair comprising a word from a first digitized document, and a corresponding word from a second digitized document;
compute for each word pair first location information for the word that indicates a location of the word in the first digitized document relative to other words in the first digitized document;
compute for each word pair second location information for the corresponding word that indicates a location of the corresponding word in the second digitized document relative to other words in the second digitized document;
compare the first and second location information; and
create one or more classes responsive to the comparison to classify digitized documents similar to the first digitized document, and to classify digitized documents similar to the second digitized document.
2. The system of claim 1 wherein the processor-based document management system is further configured to:
if the comparison indicates locations of words in the first digitized document are the same as locations of corresponding words in the second digitized document, determine that the first and second digitized documents should be in a same class; and
upon the determination that the first and second digitized documents should be in the same class, create a class to classify documents similar to the first and second digitized documents.
3. The system of claim 1 wherein the processor-based document management system is further configured to:
if the comparison indicates the locations of words in the first digitized document are different from the locations of the corresponding words in the second digitized document, determine that the first and second digitized documents should be in different classes; and
upon the determination that the first and second digitized documents should be in different classes, create a first class to classify digitized documents similar to the first digitized document, and
create a second class, different from the first class, to classify digitized documents similar to the second digitized document.
4. The system of claim 1 wherein the first location information comprises top location information, and bottom location information,
the top location information is associated with a top portion of the first digitized document, and comprises a location of a word in the top portion of the first digitized document relative to other words in the top portion, and
the bottom location information is associated with a bottom portion of the first digitized document and comprises a location of a word in the bottom portion of the first digitized document relative to other words in the bottom portion.
5. The system of claim 1 wherein the processor-based document management system is further configured to:
calculate a first vector from the word in the first digitized document to a first other word in the first digitized document, the first vector thereby indicating a location of the word relative to the first other word; and
calculate a second vector from the word in the first digitized document to a second other word in the first digitized document, the second vector thereby indicating a location of the word relative to the second other word.
6. The system of claim 1 wherein the processor-based document management system is further configured to:
calculate a Levenshtein distance between a word from the first digitized document and a word from the second digitized document;
determine that the Levenshtein distance is below a threshold value; and
based on the determination, identify the word from the first digitized document as being a word for a word pair, and identify the word from the second digitized document as being a corresponding word for the word pair.
7. The system of claim 1 wherein the processor-based document management system is further configured to:
calculate a first value that indicates an area occupied by a word in the first digitized document;
calculate a second value that indicates an area occupied by a word in the second digitized document;
determine that a difference between the first and second values is below a threshold value; and
based on the determination, identify the word in the first digitized document as being a word for a word pair, and identifying the word in the second digitized document as being a corresponding word for the word pair.
8. The system of claim 1 wherein the processor-based document management system is further configured to:
map each numerical digit of a first number in the first digitized document to a predefined value to alter the first number to a first altered number; and
map each numerical digit of a second number, different from the first number, in the second digitized document to the predefined value to alter the second number to a second altered, number, the same as the first altered number,
thereby permitting the first altered number to be considered as a word for a word pair, and the second altered number to be considered as a corresponding word for the word pair.
9. A method implemented by a computer comprising at least one processor for creating classes for classifying digitized documents comprising:
generating a plurality of word pairs, each word pair comprising a word from a first digitized document, and a corresponding word from a second digitized document;
computing for each word pair first location information for the word that indicates a location of the word in the first digitized document relative to other words in the first digitized document;
computing for each word pair second location information for the corresponding word that indicates a location of the corresponding word in the second digitized document relative to other words in the second digitized document;
comparing the first and second location information; and
creating one or more classes responsive to the comparison to classify digitized documents similar to the first digitized document, and to classify digitized documents similar to the second digitized document.
10. The method of claim 9 comprising:
if the comparison indicates locations of words in the first digitized document are the same as locations of corresponding words in the second digitized document, determining that the first and second digitized documents should be in a same class; and
upon the determination that the first and second digitized documents should be in the same class, creating a class to classify documents similar to the first and second digitized documents.
11. The method of claim 9 comprising:
if the comparison indicates the locations of words in the first digitized document are different from the locations of the corresponding words in the second digitized document, determining that the first and second digitized documents should be in different classes; and
upon the determination that the first and second digitized documents should be in different classes, creating a first class to classify digitized documents similar to the first digitized document, and
creating a second class, different from the first class, to classify digitized documents similar to the second digitized document.
12. The method of claim 9 wherein the first location information comprises top location information, and bottom location information,
the top location information is associated with a top portion of the first digitized document, and comprises a location of a word in the top portion of the first digitized document relative to other words in the top portion, and
the bottom location information is associated with a bottom portion of the first digitized document and comprises a location of a word in the bottom portion of the first digitized document relative to other words in the bottom portion.
13. The method of claim 9 wherein the computing first location information for the word that indicates a location of the word in the first digitized document relative to other words in the first digitized document comprises:
calculating a first vector from the word in the first digitized document to a first other word in the first digitized document, the first vector thereby indicating a location of the word relative to the first other word; and
calculating a second vector from the word in the first digitized document to a second other word in the first digitized document, the second vector thereby indicating a location of the word relative to the second other word.
14. The method of claim 9 wherein the generating a plurality of word pairs comprises:
calculating a Levenshtein distance between a word from the first digitized document and a word from the second digitized document;
determining that the Levenshtein distance is below a threshold value; and
based on the determination, identifying the word from the first digitized document as being a word for a word pair, and identifying the word from the second digitized document as being a corresponding word for the word pair.
15. The method of claim 9 wherein the generating a plurality of word pairs comprises:
calculating a first value that indicates an area occupied by a word in the first digitized document;
calculating a second value that indicates an area occupied by a word in the second digitized document;
determining that a difference between the first and second values is below a threshold value; and
based on the determination, identifying the word in the first digitized document as being a word for a word pair, and identifying the word in the second digitized document as being a corresponding word for the word pair.
16. The method of claim 9 wherein the generating a plurality of word pairs comprises:
mapping each numerical digit of a first number in the first digitized document to a predefined value to alter the first number to a first altered number; and
mapping each numerical digit of a second number, different from the first number, in the second digitized document to the predefined value to alter the second number to a second altered number, the same as the first altered number,
thereby permitting the first altered number to be considered as a word for a word pair, and the second altered number to be considered as a corresponding word for the word pair.
17. A computer program product, comprising a non-transitory computer-readable medium having a computer-readable program code embodied therein, the computer-readable program code adapted to be executed by one or more processors to implement a method comprising:
generating a plurality of word pairs, each word pair comprising a word from a first digitized document, and a corresponding word from a second digitized document;
computing for each word pair first location information for the word that indicates a location of the word in the first digitized document relative to other words in the first digitized document;
computing for each word pair second location information for the corresponding word that indicates a location of the corresponding word in the second digitized document relative to other words in the second digitized document;
comparing the first and second location information; and
creating one or more classes responsive to the comparison to classify digitized documents similar to the first digitized document, and to classify digitized documents similar to the second digitized document.
18. The computer program product of claim 17 wherein the method comprises:
if the comparison indicates locations of words in the first digitized document are the same as locations of corresponding words in the second digitized document, determining that the first and second digitized documents should be in a same class; and
upon the determination that the first and second digitized documents should be in the same class, creating a class to classify documents similar to the first and second digitized documents.
19. The computer program product of claim 17 wherein the method comprises:
if the comparison indicates the locations of words in the first digitized document are different from the locations of the corresponding words in the second digitized document, determining that the first and second digitized documents should be in different classes; and
upon the determination that the first and second digitized documents should be in different classes, creating a first class to classify digitized documents similar to the first digitized document, and
creating a second class, different from the first class, to classify digitized documents similar to the second digitized document.
20. The computer program product of claim 17 wherein the first location information comprises top location information, and bottom location information,
the top location information is associated with a top portion of the first digitized document, and comprises a location of a word in the top portion of the first digitized document relative to other words in the top portion, and
the bottom location information is associated with a bottom portion of the first digitized document and comprises a location of a word in the bottom portion of the first digitized document relative to other words in the bottom portion.