Enhancing cluster analysis using document metadata

• US 8,849,811 B2
• Filed: 06/29/2011
• Issued: 09/30/2014
• Est. Priority Date: 06/29/2011
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Bonanni, Cristina, Di Balsamo, Arcangelo, Donatelli, Alessandro, Lazzaro, Luca, Dauri, Francesco, Falchetti, Giovanni
• Primary Examiner(s): Arjomandi, Noosha
• Application Number:
  US13/172,186
• Publication Number:
  US 20130006979A1
• Time in Patent Office: 1,189 Days
• Field of Search: 707/728, 707/731, 707/772
• US Class Current: 707/728
• CPC Class Codes: G06F 16/35 Clustering; Classification

Abstract

A search query including search criteria can be received. The search criteria can be a text string. An enhanced search against an enhanced index can be executed. The enhanced index can be metadata associated with an enhanced cluster. The enhanced cluster can be a document cluster associated with the metadata. The enhanced cluster can be aggregated into a merged document. The merged document can be a document including the enhanced cluster contents. The ranking algorithm can be executed on the merged document to obtain a final ranking of content within the single document.

Claims

1. A system comprising:

one or more processors;

one or more memories storing program instructions that the one or more processors execute;

computer usable program code stored in at least one of the memories, which when the computer usable program code is executed by the one or more processors the system is operable to receive a search query comprising of at least one search criteria, wherein the at least one search criteria is a text string;

computer usable program code stored in at least one of the memories, which when the computer usable program code is executed by the one or more processors the system is operable to perform a keyword search against an standard index, wherein the standard index is comprised of keywords obtained from a document within a standard cluster, wherein the standard cluster is a document cluster sharing a plurality of keywords;

computer usable program code stored in at least one of the memories, which when the computer usable program code is executed by the one or more processors the system is operable to execute an enhanced search against an enhanced index, wherein the enhanced index is comprised of metadata associated with an enhanced cluster, wherein the enhanced cluster is a document cluster associated with the metadata, wherein the metadata is a user defined metadata;

computer usable program code stored in at least one of the memories, which when the computer usable program code is executed by the one or more processors the system is operable to aggregate the standard cluster and the enhanced cluster into separate merged documents, wherein the separate merged documents are documents comprising of the standard cluster and enhanced cluster contents; and

computer usable program code stored in at least one of the memories, which when the computer usable program code is executed by the one or more processors the system is operable to execute a ranking algorithm on each separate merged document to obtain a final ranking of content within the single document, wherein the ranking algorithm applies a ranking formula, wherein the ranking formula comprises sum of product of a tag line presence value with a hit weight and product of a cluster rank with the hit weight subtracted from one.

2. The system of claim 1, wherein the metadata is a keyword tag associated with the document.

3. The system of claim 1, wherein the search result conforms to an Extensible Markup Language format.

4. The system of claim 1, wherein the merged document comprises a concatenated document comprising contents of documents of the enhanced cluster.

5. A computer program product comprising a non-transitory computer readable storage medium having computer usable program code embodied therewith, the computer usable program code comprising:

computer usable program code stored in a non-transitory storage medium, when said computer usable program code is executed by a processor it is operable to receive a search query comprising of at least one search criteria, wherein the at least one search criteria is a text string;

computer usable program code stored in a non-transitory storage medium, when said computer usable program code is executed by a processor it is operable to perform a keyword search against an standard index, wherein the standard index is comprised of keywords obtained from a document within a standard cluster, wherein the standard cluster is a document cluster sharing a plurality of keywords;

computer usable program code stored in a non-transitory storage medium, when said computer usable program code is executed by a processor it is operable to execute an enhanced search against an enhanced index, wherein the enhanced index is comprised of metadata associated with an enhanced cluster, wherein the enhanced cluster is a document cluster associated with the metadata, wherein the metadata is a user defined metadata;

computer usable program code stored in a non-transitory storage medium, when said computer usable program code is executed by a processor it is operable to aggregate the standard cluster and the enhanced cluster into separate merged documents, wherein the separate merged documents are documents comprising of the standard cluster and enhanced cluster contents; and

computer usable program code stored in a non-transitory storage medium, when said computer usable program code is executed by a processor it is operable to execute a ranking algorithm on each separate merged document to obtain a final ranking of content within the single document, wherein the ranking algorithm applies a ranking formula, wherein the ranking formula comprises sum of product of a tag line presence value with a hit weight and product of a cluster rank with the hit weight subtracted from one.

6. The computer program product of claim 5, wherein the metadata is descriptive metadata, wherein the descriptive metadata is provided by a document author.

7. The computer program product of claim 5, wherein the metadata is a keyword tag associated with the document.

8. The computer program product of claim 5, wherein the computer program product is operable to populate a database with a tag, a link, and a table of contents associated with a document, wherein the document is associated with the enhanced cluster.

9. The computer program product of claim 8, wherein the table of contents comprises a tree structure, wherein documents are associated with each tree leaf, wherein children of a root node of the tree structure are classified as a main topic.

10. The computer program product of claim 5, wherein the search query is performed upon a product documentation repository.

11. A system comprising hardware comprising a processer and software stored on a non-transitory storage medium, wherein programmatic instructions of the software are executable on the hardware causing the system to:

receive a search query comprising of at least one search criteria, wherein the at least one search criteria is a text string;

perform a keyword search against an standard index, wherein the standard index is comprised of keywords obtained from a document within a standard cluster, wherein the standard cluster is a document cluster sharing a plurality of keywords;

execute an enhanced search against an enhanced index, wherein the enhanced index is comprised of metadata associated with an enhanced cluster, wherein the enhanced cluster is a document cluster associated with the metadata, wherein the metadata is a user defined metadata;

aggregate the enhanced cluster each into a merged document, wherein the merged document is a document comprising of the enhanced cluster contents; and

execute a ranking algorithm on the merged document to obtain a final ranking of content within the single document, wherein the ranking algorithm applies a ranking formula, wherein the ranking formula comprises sum of product of a tag line presence value with a hit weight and product of a cluster rank with the hit weight subtracted from one.

12. The system of claim 11, wherein the metadata is descriptive metadata, wherein the descriptive metadata is provided by a document author.

13. The system of claim 11, wherein the metadata is a keyword tag associated with the document.

14. The system of claim 11, wherein the computer program product is operable to populate a database with a tag, a link, and a table of contents associated with a document, wherein the document is associated with the enhanced cluster.

15. The system of claim 14, wherein the table of contents comprises a tree structure, wherein documents are associated with each tree leaf, wherein children of a root node of the tree structure are classified as a main topic.

16. The system of claim 11, wherein the search query is performed upon a product documentation repository.

17. The system of claim 11, wherein the merged document comprises a concatenated document comprising contents of documents of the enhanced cluster.