Generating a conceptual association graph from large-scale loosely-grouped content
First Claim
1. A computer implemented method comprising:
- grouping content nodes into one or more topically biased clusters, the content nodes comprising structured digital content and unstructured digital content, the grouping based at least in part on the connectedness of each content node member to other content node members in the same cluster;
tagging the grouped content nodes in each of the one or more topically biased clusters with one or more concepts after grouping the content nodes into one or more topically biased clusters, wherein tagging the content nodes comprises;
analyzing content of the content nodes in each group and extracting a plurality of collective token and potential concept statistics for the grouped content nodes in one of the one or more topically biased clusters;
scoring and filtering the statistics based on a measure of relevance;
generating a view of the content nodes within the topically biased cluster by aggregating the scored and filtered statistics; and
selecting one or more descriptive concepts and keywords for each content node in the topically biased cluster based on the generated view, wherein the one or more concepts comprise the one or more descriptive concepts and keywords;
finding and scoring one or more conceptual association based on the tagged, grouped content nodes and patterns of co-occurrence of the one or more concepts in the tagged content nodes, the one or more conceptual associations indicating a relevance of the one or more associations; and
generating a conceptual association graph across the topically-biased clusters based on the one or more associations between the one or more concepts.
6 Assignments
0 Petitions
Accused Products
Abstract
A method for generating a conceptual association graph from structured content includes grouping content nodes into one or more topically biased clusters, the content nodes comprising structured digital content and unstructured digital content, the grouping based at least in part on the connectedness of each content node member to other content node members in the same cluster. The method also includes, responsive to the grouping, tagging the content nodes with one or more descriptive concepts. The method also includes, responsive to the tagging, establishing one or more associations between the one or more concepts, the one or more associations indicating a relevance of the one or more associations, the indicating based at least in part on patterns of co-occurrence of concepts in the tagged content nodes.
163 Citations
22 Claims
-
1. A computer implemented method comprising:
-
grouping content nodes into one or more topically biased clusters, the content nodes comprising structured digital content and unstructured digital content, the grouping based at least in part on the connectedness of each content node member to other content node members in the same cluster; tagging the grouped content nodes in each of the one or more topically biased clusters with one or more concepts after grouping the content nodes into one or more topically biased clusters, wherein tagging the content nodes comprises; analyzing content of the content nodes in each group and extracting a plurality of collective token and potential concept statistics for the grouped content nodes in one of the one or more topically biased clusters; scoring and filtering the statistics based on a measure of relevance; generating a view of the content nodes within the topically biased cluster by aggregating the scored and filtered statistics; and selecting one or more descriptive concepts and keywords for each content node in the topically biased cluster based on the generated view, wherein the one or more concepts comprise the one or more descriptive concepts and keywords; finding and scoring one or more conceptual association based on the tagged, grouped content nodes and patterns of co-occurrence of the one or more concepts in the tagged content nodes, the one or more conceptual associations indicating a relevance of the one or more associations; and generating a conceptual association graph across the topically-biased clusters based on the one or more associations between the one or more concepts. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16)
-
-
17. A computer system comprising:
-
memory; and a processor coupled to the memory, the processor comprising; a content grouping module configured to group content nodes into one or more topically biased clusters, the content nodes comprising structured digital content and unstructured digital content, the grouping based at least in part on the connectedness of each content node member to other content node members in the same cluster; a concept extraction module configured to tag the content nodes in each of the one or more topically biased clusters with one or more concepts after the content nodes are grouped into one or more topically biased clusters, wherein the concept extraction module is configured to tag the content nodes by analyzing content of the content nodes in each group and extracting a plurality of collective token and potential concept statistics for the grouped content nodes in one of the one or more topically biased clusters, scoring and filtering the statistics based on a measure of relevance, generating a view of the content nodes within the topically biased cluster by aggregating the scored and filtered statistics, and selecting one or more descriptive concepts and keywords for each content node in the topically biased cluster based on the generated view, wherein the one or more concepts comprise the one or more descriptive concepts and keywords; and a conceptual map builder configured to find and score one or more conceptual associations based on the tagged, grouped content nodes and based on patterns of co-occurrence of the one or more concepts in the tagged content nodes, the one or more conceptual associations indicating a relevance of the one or more associations, and wherein the conceptual map builder is further configured to generate a conceptual association graph across the topically biased clusters based on the one or more associations between the one or more concepts. - View Dependent Claims (18, 19)
-
-
20. A non-transitory computer-readable media having computer executable instructions stored thereon which cause a computer system to carry out a method when executed, the method comprising
grouping content nodes into one or more topically biased clusters, the content nodes comprising structured digital content and unstructured digital content, the grouping based at least in part on the connectedness of each content node member to other content node members in the same cluster; -
tagging the grouped content nodes in each of the one or more topically biased clusters with one or more concepts after grouping the content nodes into one or more topically biased clusters, wherein tagging the content nodes comprises; analyzing content of the content nodes in each group and extracting a plurality of collective token and potential concept statistics for the grouped content nodes in one of the one or more topically biased clusters; scoring and filtering the statistics based on a measure of relevance; generating a view of the content nodes within the topically biased cluster by aggregating the scored and filtered statistics; and selecting one or more descriptive concepts and keywords for each content node in the topically biased cluster based on the generated view, wherein the one or more concepts comprise the one or more descriptive concepts and key words; finding and scoring one or more conceptual associations based on the tagged, grouped content nodes and based on patterns of co-occurrence of the one or more concepts in the tagged content nodes, the one or more conceptual associations indicating a relevance of the one or more associations; and generating a conceptual association graph across the topically biased clusters based on the one or more associations between the one or more concepts. - View Dependent Claims (21, 22)
-
Specification