Methods and apparatus for determining focal points of clusters in a tree structure

Claim 3

The method as set forth in claim 1 , wherein said base terms comprise themes from a corpus of documents.

Claim 6

The method as set forth in claim 1 , wherein the step of assigning quantitative values to nodes in said set of nodes from said weight values in said data set comprises the step of generating a raw weight for a node of said tree structure by summing weights for values of each attribute of said data set with a corresponding attribute.

Claim 7

The method as set forth in claim 1 , wherein the step of assigning quantitative values to nodes in said set of nodes from said weight values in said data set comprises the step of summing weight values for a node in said set of nodes for all nodes that descend from said node.

Claim 8

The method as set forth in claim 1 , wherein the step of assigning quantitative values to nodes in said set of nodes from said weight values in said data set comprises the step of dividing a weight value for a node in said set from a parent node proportionally among its child nodes based upon relative weight values of said child nodes.

Claim 9

The method as set forth in claim 1 , wherein the step of selecting one or more clusters of nodes from said set of nodes based on said quantitative values and said relationships in said tree structure comprises the step of applying a breadth cut-off percent to determine how broadly across said tree structure to select nodes as part of a cluster.

Claim 10

The method as set forth in claim 1 , wherein the step of selecting at least one focal point node for said cluster of nodes comprises the step of applying a depth cut-off percent to determine how far down in said hierarchy to select said focal point node.

Claim 11

A computer readable medium comprising a plurality of instructions, which when executed, cases the computer to determine at least one focal node on a hierarchically arranged tree structure of nodes for attributes of a data set, said instructions causing the computer to perform the steps of: storing a tree structure comprising a plurality of nodes of tree structure attributes, arranged in a hierarchy, to depict relationships among said tree structure attributes; receiving a data set comprising a plurality of data set attributes with associated weight values; selecting a set of nodes from said tree structure with tree structure attributes that correspond with said data set attributes; assigning quantitative values to nodes in said set of nodes from said weight values in said data set; selecting at least one cluster of nodes, based on close proximity of said nodes in said tree structure; and selecting a focal node on said tree structure for said cluster of nodes based on said data set attributes, said focal node comprising tree structure attributes most representative of said data set attributes, wherein selection of said focal node includes evaluating attributes of said nodes of said cluster starting from a node at the top of said hierarchy of said tree structure and analyzing downward to select said focal node based on said quantitative values and said relationships of said attributes in said tree structure.

Claim 12

A computer readable medium comprising a plurality of instructions, which when executed, cases the computer to determine at least one focal node on a hierarchically arranged tree structure of nodes for attributes of a data set, said instructions causing the computer to perform the steps of: storing a tree structure comprising a plurality of nodes of tree structure attributes, arranged in a hierarchy, to depict relationships among said tree structure attributes; receiving a data set comprising a plurality of data set attributes with associated weight values; selecting a set of nodes from said tree structure with tree structure attributes that correspond with said data set attributes; assigning quantitative values to nodes in said set of nodes from said weight values in said data set; selecting at least one cluster of nodes, based on close proximity of said nodes in said tree structure; and selecting a focal node on said tree structure for said cluster of nodes based on said data set attributes, said focal node comprising tree structure attributes most representative of said data set attributes, wherein selection of said focal node includes evaluating attributes of said nodes of said cluster starting from a node at the top of said hierarchy of said tree structure and analyzing downward to select said focal node based on said quantitative values and said relationships of said attributes in said tree structure.

Claim 13

The computer readable medium as set forth in claim 9 , wherein the step of assigning quantitative values to nodes in said set of nodes from said weight values in said data set comprises the step of generating a raw weight for a node of said tree structure by summing weights for values of each attribute of said data set with a corresponding attribute.

Claim 14

The computer readable medium as set forth in claim 9 , wherein the step of assigning quantitative values to nodes in said set of nodes from said weight values in said data set comprises the step of summing weight values for a node in said set of nodes for all nodes that descend from said node.

Claim 15

The computer readable medium as set forth in claim 9 , wherein the step of assigning quantitative values to nodes in said set of nodes from said weight values in said data set comprises the step of dividing a weight value for a node in said set from a parent node proportionally among its child nodes based upon relative weight values of said child nodes.

Claim 16

The computer readable medium as set forth in claim 9 , wherein the step of selecting one or more clusters of nodes from said set of nodes based on said quantitative values and said relationships in said tree structure comprises the step of applying a breadth cut-off percent to determine how broadly across said tree structure to select nodes as part of a cluster.

Claim 17

The computer readable medium as set forth in claim 9 , wherein the step of selecting at least one focal point node for said cluster of nodes comprises the step of applying a depth cut-off percent to determine how far down in said hierarchy to select said focal point node.