System and method for classifying electronically posted documents

A system for classifying electronically posted documents, the system comprising:

• a memory;
• a processor communicatively coupled to the memory;
• a metadata parser module coupled to receive electronically posted documents, the metadata parser configured to output a metadata summary for each of the posted documents, wherein each of the metadata summaries comprises a plurality of sub-trees and each of the sub-trees includes a plurality of nodes, wherein each of the sub-trees in the pluralities of sub-trees comprises a structural format defined by a plurality of hierarchical structural constructs, and wherein subsets of the metadata summaries are grouped into a set of different summary groups based on a mime-type designation associated with each subset of the metadata summaries;
• a summary repository coupled to receive and store the metadata summaries;
• and
• a summary consolidator coupled to the summary repository, the summary consolidator configured to:
• select a first metadata summary and a second metadata summary from one of the summary groups in the set of different summary groups;
• compare the first metadata summary and the second metadata summary on a structural level by comparing the entire structural format defined by the plurality of hierarchical structural constructs of the sub-trees of the first metadata summary with the entire structural format defined by the plurality of hierarchical structural constructs of the sub-trees of the second metadata summary;
• identify the first and second documents corresponding to the first and second metadata summaries as distinct if the entire structural formats of the sub-trees of the first and second metadata summaries are not equivalent;
• and
• if the entire structural formats of the sub-trees of the first and second metadata summaries are equivalent, further compare the first and second metadata summaries,
• wherein the further comparison of the first and second metadata summaries includes:
• comparing the first and second metadata summaries on a textual level by comparing textual content from the first document that is contained in the sub-trees of the first metadata summary with textual content from the second document that is contained in the sub-trees of the second metadata summary;
• identifying the first and second documents corresponding to the first and second metadata summaries as distinct if the textual content within the sub-trees of the first and second metadata summaries are not equivalent;
• before comparing the first and second metadata summaries on a textual level, comparing the first and second metadata summaries on an attribute level by comparing attribute values within the sub-trees of the first metadata summary with attribute values within the sub-trees of the second metadata summary;
• and
• identifying the first and second documents corresponding to the first and second metadata summaries as distinct if the attribute values within the sub-trees of the first and second metadata summaries are not equivalent.

A program product for use in a computer system that executes program steps recorded in a computer-readable media to perform a method for classifying electronically posted documents, the program product comprising:

• a record-able media;
• a program of computer-readable instructions executable by the computer system to perform processes comprising the steps of:
• receiving a first document and a second document;
• generating a first metadata summary for the first document and a second metadata summary for the second document, wherein the first metadata summary includes a first plurality of sub-trees and the second metadata summary includes a second plurality of sub-trees, and wherein each of the sub-trees includes a plurality of nodes, wherein each of the sub-trees in the first and second pluralities of sub-trees comprises a structural format defined by a plurality of hierarchical structural constructs, and wherein subsets of the metadata summaries are grouped into a set of different summary groups based on a mime-type designation associated with each subset of the metadata summaries;
• comparing the first and second metadata summaries on a structural level by comparing the entire structural format defined by the plurality of hierarchical structural constructs of the sub-trees of the first metadata summary with the entire structural format defined by the plurality of hierarchical structural constructs of the sub-trees of the second metadata summary;
• identifying the first and second documents as distinct if the entire structural formats of the sub-trees of the first and second metadata summaries are not equivalent;
• if the entire structural formats of the sub-trees of the first and second metadata summaries are equivalent, performing a further comparison of the first and second metadata summaries,
• wherein the further comparison of the first and second metadata summaries includes the sub-steps of:
• comparing the first and second metadata summaries on a textual level by comparing textual content from the first document that is contained in the sub-trees of the first metadata summary with textual content from the second document that is contained in the sub-trees of the second metadata summary;
• identifying the first and second documents as distinct if the textual content within the sub-trees of the first and second metadata summaries are not equivalent;
• before comparing the first and second metadata summaries on a textual level, comparing the first and second metadata summaries on an attribute level by comparing attribute values within the sub-trees of the first metadata summary with attribute values within the sub-trees of the second metadata summary;
• and
• identifying the first and second documents as distinct if the attribute values within the sub-trees of the first and second metadata summaries are not equivalent.