Method and system for managing exemplar terms database for business-oriented metadata content

• US 7,873,670 B2
• Filed: 07/28/2006
• Issued: 01/18/2011
• Est. Priority Date: 07/29/2005
• Status: Expired due to Fees

• Current Assignee: Ginegar LLC (K.Mizra LLC)
• Original Assignee: International Business Machines Corporation
• Inventors: Statchuk, Craig
• Primary Examiner(s): Vo; Tim T.
• Assistant Examiner(s): TRAN, ANHTAI V
• Application Number:
  US11/494,936
• Publication Number:
  US 20070055691A1
• Time in Patent Office: 1,635 Days
• Field of Search: 707/796
• US Class Current: 707/796
• CPC Class Codes: G06F 16/31 Indexing; Data structures therefor; Storage structures

Abstract

An example management system comprises an indexing engine, an index store and an example engine. The indexing engine indexes content of source business oriented metadata. The indexing engine has a content scanner for reading the business oriented metadata, and building a content index of the business oriented metadata. The index store stores the content index of the business oriented metadata. The example engine manages logical associations of terms in a query using the content index.

Claims

1. An example management system, comprising:

at least one computer;

an indexing engine executable by the at least one computer to index content of source business oriented metadata, the source business oriented metadata including an authored topic hierarchy, the indexing engine having a content scanner that reads the source business oriented metadata included in source metadata documents containing terms to be indexed, and builds a content index of the source business oriented metadata, wherein the content scanner builds, as the content index, one or more knowledge base documents, wherein, for each term in the source metadata documents, the content scanner stores, in the one or more knowledge base documents, a knowledge base representation of the term along with one or more references to content of the source business oriented metadata that uses the term, and wherein the one or more knowledge base documents include a representation of a structure of the authored topic hierarchy;

an index store that stores the content index of the source business oriented metadata; and

an example engine executable by the at least one computer to use the representation of the structure of the authored topic hierarchy to determine logical associations among the terms in the source metadata documents, store the logical associations in the one or more knowledge base documents, and manage logical associations of terms in a query using the one or more knowledge base documents of the content index, wherein the example engine determines an example association between two terms when a term is an example of another term, and wherein the example engine combines terms into phrases and determines, based on example associations between terms in the phrases, an example association between two phrases when a phrase is an example of another phrase.

2. The example management system as claimed in claim 1, wherein the example engine locates relevant knowledge base documents relevant to the terms in the query, and determines the logical associations between terms in the query based on the relevant knowledge base documents.

3. The example management system as claimed in claim 1, wherein the example engine dynamically determines the logical associations of terms in the query upon receipt of the query from a search component, and provides the logical associations to the search component.

4. The example management system as claimed in claim 1, wherein the example engine interacts with a word stemming component that stems one or more terms in the query to manage logical associations of the terms based on the stemmed terms.

5. The example management system as claimed in claim 2, wherein the example engine locates the relevant knowledge base documents by searching for knowledge base documents containing each term in the query, each of the one or more knowledge base documents containing a list of source metadata documents using the term in the query, and reads the relevant knowledge base documents to find term isomorphisms for the terms in the query to retrieve examples from the relevant knowledge base documents.

6. The example management system as claimed in claim 2, wherein

the content scanner includes, in the one or more knowledge base documents, reporting elements included in the source metadata documents, and

the example engine determines example associations of terms using the reporting elements included in the relevant knowledge base documents.

7. The example management system as claimed in claim 2, wherein

the content scanner builds the one or more knowledge base documents to contain one or more term-to-document associations indicating one or more source business oriented metadata documents that contain the term,

the example engine determines example associations of terms in the query using the term-to-document associations in the relevant knowledge base documents.

8. The example management system as claimed in claim 2, wherein

the content scanner uses a structured hierarchy of reporting elements of the source business oriented metadata to include, in the one or more knowledge base documents, one or more references indicating parent/child relations between the term and the content of the source business oriented metadata, and

the example engine determines example associations of terms using the references indicating the parent/child relations included in the relevant knowledge base documents.

9. The example management system as claimed in claim 2, wherein

the content scanner includes, in the one or more knowledge base documents, one or more links to other content of the source business oriented metadata, and

the example engine determines example associations of terms using the links included in the relevant knowledge base documents.

10. The example management system as claimed in claim 3, wherein the example management system includes the word stemming component.

11. The example management system as claimed in claim 2, wherein

the content scanner includes, in the one or more knowledge base documents, a prompt value pick-list, and

the example engine determines example associations of terms using the prompt value pick-list included in the relevant knowledge base documents.

12. A method of managing associations of terms for a search component, the method comprising:

reading source business oriented metadata, wherein the source business oriented metadata is included in source metadata documents containing terms to be indexed, and wherein the source business oriented metadata includes an authored topic hierarchy;

indexing content of the source business oriented metadata;

building, with a computer, a content index of the source business oriented metadata, wherein building the content index comprises, for each term in the source metadata documents, generating a knowledge base representation of the term, and storing, in one or more knowledge base documents, the knowledge base representation of the term along with one or more references to content of the source business oriented metadata that uses the term, and wherein the one or more knowledge base documents include a representation of a structure of the authored topic hierarchy;

using the representation of the structure of the authored topic hierarchy to determine logical associations among the terms in the source metadata documents;

storing the logical associations in the one or more knowledge base documents; and

managing logical associations of terms in a query using the one or more knowledge base documents of the content index, wherein managing the logical associations comprises determining an example association between two terms when a term is an example of another term, combining terms into phrases and determining, based on example associations between terms in the phrases, an example association between two phrases when a phrase is an example of another phrase.

13. The method as claimed in claim 12, wherein reading the source business oriented metadata comprises reading the source business oriented metadata that includes at least one of Online Analytical Processing (OLAP) hypercubes, report query specifications and reports.

14. The method as claimed in claim 12,

wherein managing the logical associations comprises:

locating relevant knowledge base documents relevant to the terms in the query, and

determining the logical associations between terms in the query based on the relevant knowledge base documents.

15. The method as claimed in claim 12, wherein managing the logical associations comprises:

receiving the query from a search component;

dynamically determining the logical associations of terms in the query; and

providing the logical associations to the search component.

16. The method as claimed in claim 12, wherein managing the logical associations further comprises stemming one or more terms in the query to manage logical associations of the terms based on the stemmed terms.

17. The method as claimed in claim 14, wherein managing the logical associations comprises:

locating the relevant knowledge base documents by searching for knowledge base documents containing each term in the query, each of the one or more knowledge base documents containing a list of source metadata documents using the term in the query;

reading the relevant knowledge base documents;

finding term isomorphisms for the terms in the query; and

retrieving examples from the relevant knowledge base documents.

18. The method as claimed in claim 14, wherein

building the content index comprises including, in the one or more knowledge base documents, reporting elements included in the source metadata documents, and

managing the logical associations comprises determining example associations of terms using the reporting elements included in the relevant knowledge base documents.

19. The method as claimed in claim 14, wherein

building the content index comprises building the one or more knowledge base documents to contain one or more term-to-document associations indicating one or more source business oriented metadata documents that contain the term, and

managing the logical associations comprises determining example associations of terms in the query using the term-to-document associations in the relevant knowledge base documents.

20. The method as claimed in claim 14, wherein

building the content index comprises using a structured hierarchy of reporting elements of the source business oriented metadata to include, in the one or more knowledge base documents, one or more references indicating parent/child relations between the term and the content of the source business oriented metadata, and

managing the logical associations comprises determining example associations of terms using the references indicating the parent/child relations included in the relevant knowledge base documents.

21. The method as claimed in claim 14, wherein

building the content index comprises including, in the one or more knowledge base documents, one or more links to other content of the source business oriented metadata, and

managing the logical associations comprises determining example associations of terms using the links included in the relevant knowledge base documents.

22. The method as claimed in claim 14, wherein

building the content index comprises including, in the one or more knowledge base documents, a prompt value pick-list, and

managing the logical associations comprises determining example associations of terms using the prompt value pick-list included in the relevant knowledge base documents.

23. A computer readable medium storing instructions that cause a computer to:

read source business oriented metadata, wherein the source business oriented metadata is included in source metadata documents containing terms to be indexed, and wherein the source business oriented metadata includes an authored topic hierarchy;

index content of the source business oriented metadata;

build a content index of the source business oriented metadata, wherein the instructions that cause the computer to build the content index comprise instructions that cause the computer to, for each term in the source metadata documents, generate a knowledge base representation of the term, and store, in one or more knowledge base documents, the knowledge base representation of the term along with one or more references to content of the source business oriented metadata that uses the term, and wherein the one or more knowledge base documents include a representation of a structure of the authored topic hierarchy;

use the representation of the structure of the authored topic hierarchy to determine logical associations among the terms in the source metadata documents;

store the logical associations in the one or more knowledge base documents; and

manage logical associations of terms in a query using the one or more knowledge base documents of the content index, wherein the instructions that cause the computer to manage the logical associations comprise instructions that cause the computer to determine an example association between two terms when a term is an example of another term, combine terms into phrases and determine, based on example associations between terms in the phrases, an example association between two phrases when a phrase is an example of another phrase.