Data classification methods and apparatus for use with data fusion

• US 7,792,771 B2
• Filed: 02/19/2009
• Issued: 09/07/2010
• Est. Priority Date: 09/28/2004
• Status: Active Grant

• Current Assignee: Nielsen Company (US), LLC
• Original Assignee: The Nielsen Company LLC (MOOD Media Corporation)
• Inventors: McMillan, Francis Gavin, Samson, Jerome
• Primary Examiner(s): Holmes; Michael
• Application Number:
  US12/389,162
• Publication Number:
  US 20090157581A1
• Time in Patent Office: 565 Days
• Field of Search: 706/45
• US Class Current: 706/45
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising

Abstract

Methods and apparatus for classifying data for use in data fusion processes are disclosed. An example method of classifying data selectively groups nodes of a classification tree so that each node is assigned to only one of a plurality of groups and so that at least one of the groups includes at least two of the nodes. Data is classified based on the classification tree and the selective grouping of the nodes, and the results displayed.

Claims

1. A computer implemented method of classifying data, comprising:

selectively grouping, with a programmed processor, nodes of a classification tree so that each node is assigned to only one of a plurality of groups and so that at least one of the groups includes at least two of the nodes;

classifying the data based on the classification tree and the selective grouping of the nodes; and

displaying the classified data results.

2. A method as defined in claim 1, wherein selectively grouping the nodes includes assigning each of the nodes to one of the groups based on values associated with the nodes, wherein each of the values is representative of a relationship between its respective node and one of the groups.

3. A method as defined in claim 2, wherein the each of the values is associated with a frequency distribution, a probability, or an index value.

4. A method as defined in claim 1, wherein selectively grouping the nodes includes comparing values associated with the nodes and assigning each of the nodes to one of the groups based on the comparisons.

5. A method as defined in claim 4, wherein comparing the values associated with the nodes and assigning each of the nodes to one of the groups based on the comparisons includes re-assigning at least one of the nodes from a first one of the groups to a second one of the groups.

6. A method as defined in claim 1, wherein selectively grouping the nodes includes selectively grouping the nodes in accordance with an overall population frequency distribution of the groups.

7. A method as defined in claim 1, wherein classifying the data based on the selective grouping of the nodes includes assigning group identifying information to each of a plurality of data records comprising the data so that each of the plurality of the data records is associated with only one of the groups.

8. A method as defined in claim 7, further comprising fusing the data records based on the group identifying information.

9. An apparatus for classifying data, comprising:

a memory; and

a processor coupled to a memory and configured to;

selectively group nodes of a classification tree so that each node is assigned to only one of a plurality of groups and so that at least one of the groups includes at least two of the nodes;

classify the data based on the classification tree and the selective grouping of the nodes; and

display the classified data results.

10. An apparatus as defined in claim 9, wherein the processor is configured to selectively group the nodes by assigning each of the nodes to one of the groups based on values associated with the nodes, wherein each of the values is representative of a relationship between its respective node and one of the groups.

11. An apparatus as defined in claim 10, wherein the each of the values is associated with a frequency distribution, a probability, or an index value.

12. An apparatus as defined in claim 9, wherein the processor is configured to selectively group the nodes by comparing values associated with the nodes and assigning each of the nodes to one of the groups based on the comparisons.

13. An apparatus as defined in claim 12, wherein the processor is configured to compare the values associated with the nodes and assign each of the nodes to one of the groups based on the comparisons by re-assigning at least one of the nodes from a first one of the groups to a second one of the groups.

14. An apparatus as defined in claim 9, wherein the processor is configured to selectively group the nodes by selectively grouping the nodes in accordance with an overall population frequency distribution of the groups.

15. An apparatus as defined in claim 9, wherein the processor is configured to classify the data based on the selective grouping of the nodes by assigning group identifying information to each of a plurality of data records comprising the data so that each of the plurality of the data records is associated with only one of the groups.

16. An apparatus as defined in claim 15, wherein the processor is configured to fuse the data records based on the group identifying information.

17. A tangible machine accessible medium having instructions stored thereon that, when executed, cause the machine to:

selectively group nodes of a classification tree so that each node is assigned to only one of a plurality of groups and so that at least one of the groups includes at least two of the nodes;

classify data based on the classification tree and the selective grouping of the nodes; and

display the classified data results.

18. A machine accessible medium as defined in claim 17, wherein the instructions, when executed, cause the machine to selectively group the nodes by assigning each of the nodes to one of the groups based on values associated with the nodes, wherein each of the values is representative of a relationship between its respective node and one of the groups.

19. A machine accessible medium as defined in claim 18, wherein the each of the values is associated with frequency distribution, a probability, or an index value.

20. A machine accessible medium as defined in claim 17, wherein the instructions, when executed, cause the machine to selectively group the nodes by comparing values associated with the nodes and assigning each of the nodes to one of the groups based on the comparisons.

21. A machine accessible medium as defined in claim 20, wherein the instructions, when executed, cause the machine to compare the values associated with the nodes and assign each of the nodes to one of the groups based on the comparisons by re-assigning at least one of the nodes from a first one of the groups to a second one of the groups.

22. A machine accessible medium as defined in claim 17, wherein the instructions, when executed, cause the machine to selectively group the nodes by selectively grouping the nodes in accordance with an overall population frequency distribution of the groups.

23. A machine accessible medium as defined in claim 17, wherein the instructions, when executed, cause the machine to classify the data based on the selective grouping of the nodes by assigning group identifying information to each of a plurality of data records comprising the data so that each of the plurality of the data records is associated with only one of the groups.

24. A machine accessible medium as defined in claim 23, wherein the instructions, when executed, cause the machine to fuse the data records based on the group identifying information.

25. An apparatus for classifying data, comprising:

a node analyzer configured to generate a plurality of values for each of a plurality of nodes of a classification tree, wherein each of the plurality of values represents a relationship between one of a plurality of classes and the characteristics of a population of data;

a node grouper configured to group the classification tree nodes so that each of a plurality of groups corresponds to only one of the classes;

a class code assignor configured to use the groups to assign class codes to the data; and

a display to display the classified data results.

26. An apparatus as defined in claim 25, wherein the node analyzer is configured to generate a plurality of index values, each of which represents a ratio of the likelihood of occurrence of a class within a node and the likelihood of occurrence of a class within the population of data.

27. An apparatus as defined in claim 25, wherein the node grouper is configured re-assign at least one of the classification tree nodes to one of the groups.

28. An apparatus as defined in claim 25, wherein the node grouper is configured to assign each classification tree node to only one of the classes based on comparisons of the values.

29. An apparatus as defined in claim 25, wherein the node grouper is configured to assign the assigning the classification tree nodes to the classes so that at least two nodes are assigned to the same class.