Method and system for guided cluster based processing on prototypes

1. A method comprising:

receiving, by a computer system, a first hierarchy including at least a first category and a second category that is a child of the first category;

receiving, by the computer system, a plurality of items, each of the plurality of items including information, at least one of the items being associated with one of the categories in the first hierarchy;

processing, by the computer system, the plurality of items in a clustering process based upon at least the first hierarchy and the information in each of the items, wherein the processing comprises:

forming a tentative cluster from at least two of the items;

determining whether the tentative cluster is consistent with an extension of the first hierarchy; and

rejecting the tentative cluster in the event that the tentative cluster is inconsistent with an extension of the first hierarchy;

automatically determining, by the computer system, a second hierarchy based upon the processing, the second hierarchy including a portion of the first hierarchy and one or more additional categories coupled to the first hierarchy, at least one of the one or more additional categories being a sibling of the second category of the first hierarchy;

storing the second hierarchy in memory; and

assigning each of the plurality of items to a category of the second hierarchy.

2. The method of claim 1 wherein the clustering process guarantees that the second hierarchy is an extension of the first hierarchy.

3. The method of claim 1 wherein the first hierarchy further includes a third category and a fourth category that is a child of the third category, and wherein the processing includes interposing a new category between the third category and the fourth category.

4. A computer-readable storage medium having stored thereon program code executable by a processor, the program code comprising:

code that causes the processor to receive a first hierarchy including at least a first category and a second category that is child of the first category;

code that causes the processor to receive a plurality of items, each of the plurality of items including information, at least one of the items being associated with one of the categories in the first hierarchy;

code that causes the processor to process the plurality of items in a clustering process based upon at least the first hierarchy and the information in each of the items, wherein the code that causes the processor to process the plurality of items in a clustering process includes:

code that causes the processor to form a tentative cluster from at least two of the items;

code that causes the processor to determine whether the tentative cluster is consistent with an extension of the first hierarchy; and

code that causes the processor to reject the tentative cluster in the event that the tentative cluster is inconsistent with an extension of the first hierarchy;

code that causes the processor to automatically determine a second hierarchy based upon the processing, the second hierarchy including a portion of the first hierarchy and one or more additional categories coupled to the first hierarchy, at least one of the one or more additional categories being a sibling of the second category of the first hierarchy;

code that causes the processor to store the second hierarchy in memory; and

code that causes the processor to assign each of the plurality of items to a category of the second hierarchy.

5. The computer-readable storage medium of claim 4 wherein the clustering process guarantees that the second hierarchy is an extension of the first hierarchy.

6. The computer-readable medium of claim 4 wherein the first hierarchy further includes a third category and a fourth category that is a child of the third category, and wherein the processing includes interposing a new category between the third category and the fourth category such that the new category is a child of the third category and a parent of the fourth category in the second hierarchy.

7. A computer system comprising:

a processor configured to:

receive a first hierarchy including at least a first category and a second category that is a child of the first category;

receive a plurality of items, each of the plurality of items including information, at least one of the items being associated with one of the categories in the first hierarchy;

process the plurality of items in a clustering process based upon at least the first hierarchy and the information in each of the items, wherein the processing comprises:

forming a tentative cluster from at least two of the items;

determining whether the tentative cluster is consistent with an extension of the first hierarchy; and

rejecting the tentative cluster in the event that the tentative cluster is inconsistent with an extension of the first hierarchy;

automatically determine a second hierarchy based upon the processing, the second hierarchy including a portion of the first hierarchy and one or more additional categories coupled to the first hierarchy, at least one of the one or more additional categories being a sibling of the second category of the first hierarchy;

storing the second hierarchy in memory; and

assigning each of the plurality of items to a category of the second hierarchy.

8. The computer system of claim 7 wherein the clustering process guarantees that the second hierarchy is an extension of the first hierarchy.

9. The computer system of claim 7 wherein the first hierarchy further includes a third category and a fourth category that is a child of the third category, and wherein the processing includes interposing a new category between the third category and the fourth category.