System for folder classification based on folder content similarity and dissimilarity

• US 7,630,946 B2
• Filed: 07/12/2006
• Issued: 12/08/2009
• Est. Priority Date: 05/16/2006
• Status: Expired due to Fees

• Current Assignee: Sony Corporation (Sony Group Corp.), Sony Electronics Inc. (Sony Group Corp.)
• Original Assignee: Sony Corporation (Sony Group Corp.), Sony Electronics Inc. (Sony Group Corp.)
• Inventors: Acharya, Chiranjit
• Primary Examiner(s): Vincent; David R
• Assistant Examiner(s): Kennedy; Adrian L
• Application Number:
  US11/457,107
• Publication Number:
  US 20070271291A1
• Time in Patent Office: 1,245 Days
• Field of Search: None
• US Class Current: 706/45
• CPC Class Codes: G06F 16/285 Clustering or classification
• CPC Class Codes: G06F 18/231 Hierarchical techniques, i.e. dividing or merging pattern sets so as to obtain a dendrogram
• CPC Class Codes: G06F 7/00 Methods or arrangements for processing data by operating upon the order or content of the data handled logic circuits H03K19/00
• CPC Class Codes: G06N 5/022 Knowledge engineering; Knowledge acquisition
• CPC Class Codes: G06N 7/01 Probabilistic graphical models, e.g. probabilistic networks

Abstract

A computerized method of representing a dataset with a taxonomy includes obtaining a dataset containing a plurality of records; initializing a folder-set containing a plurality of folders; assigning labels to folders within the folder set; and classifying the plurality of records into the plurality of folders according to a predetermined entropic similarity condition.

Claims

1. A computerized method of representing a dataset with a taxonomy, comprising:

obtaining a dataset containing a plurality of records;

initializing a folder-set containing a plurality of folders;

assigning labels to folders within the folder set;

classifying the plurality of records into the plurality of folders according to a predetermined entropic similarity condition; and

merging a plurality of folders when it is determined that a similarity between the plurality of folders is greater than a dissimilarity between the folders;

wherein the computerized method of representing a dataset with a taxonomy occurs within a physical computer.

2. The method of claim 1, wherein initializing comprises receiving a user-defined folder set.

3. The method of claim 1, wherein initializing comprises:

obtaining an exemplar, the exemplar establishing a classification of a record into a folder; and

automatically creating a folder-set formed of visible folders using the obtained exemplar.

4. The method of claim 1, wherein the exemplar is a previously classified record establishing a classification of a record into a folder.

5. The method of claim 4, wherein the exemplar is a user-defined exemplar establishing a user-defined classification of a record into a folder.

6. The method of claim 1, wherein classifying comprises:

characterizing combinations of records and folders according to a predetermined entropic similarity metric; and

classifying each record into a folder based on a value characterizing the combination of each record and folder.

7. The method of claim 6, wherein classifying comprises classifying each record into a plurality of folders based on values characterizing the combination of each record and folder.

8. The method of claim 1, further comprising splitting a folder within the folder set into a plurality of folders when it is determined that an intra-folder similarity of the folder is less than an intra-folder dissimilarity of the folder.

9. The method of claim 1, further comprising:

receiving feedback from the user, the feedback establishing that a record should not be classified within a folder;

creating a hidden folder using the feedback received; and

classifying the plurality of records into the hidden folder according to the predetermined entropic similarity condition.

10. A computer program product comprising a physical computer usable medium having computer readable code embodied therein for causing a physical computer to effect:

obtaining a dataset containing a plurality of records;

initializing a folder-set containing a plurality of folders;

assigning labels to folders within the folder set;

classifying the plurality of records into the plurality of folders according to a predetermined entropic similarity condition; and

a computer usable medium having computer readable code embodied therein for causing a computer to effect merging a plurality of folders when it is determined that a similarity between the plurality of folders is greater than a dissimilarity between the folders.

11. The computer program product of claim 10, further comprising a computer usable medium having computer readable code embodied therein for causing a computer to effect initializing by receiving a user-defined folder set.

12. The computer program product of claim 10, further comprising a computer usable medium having computer readable code embodied therein for causing a computer to effect initializing by:

obtaining at least one exemplar, the at least one exemplar establishing a classification of a record into a folder; and

automatically creating a folder-set formed of visible folders using the at least one received exemplar.

13. The computer program product of claim 10, wherein the exemplar is a previously classified record establishing a classification of a record into a folder.

14. The computer program product of claim 13, wherein the exemplar is a user-defined exemplar establishing a user-defined classification of a record into a folder.

15. The computer program product of claim 10, further comprising a computer usable medium having computer readable code embodied therein for causing a computer to effect classifying by:

characterizing combinations of records and folders according to a predetermined entropic similarity metric; and

classifying each record into a folder based on a value characterizing the combination of each record and folder.

16. The computer program product of claim 15, further comprising a computer usable medium having computer readable code embodied therein for causing a computer to effect classifying by classifying each record into a plurality of folders based on values characterizing the combination of each record and folder.

17. The computer program product of claim 10, further comprising a computer usable medium having computer readable code embodied therein for causing a computer to effect splitting a folder within the folder set into a plurality of folders when it is determined that an intra-folder similarity of the folder is less than an intra-folder dissimilarity of the folder.

18. The computer program product of claim 10, further comprising a computer usable medium having computer readable code embodied therein for causing a computer to effect:

receiving feedback from the user, the feedback establishing that a record should not be classified within a folder;

creating at least one hidden folder using the feedback received; and

classifying the plurality of records into the at least one folder according to the predetermined entropic similarity condition.