Browse Prior Art Database

Populate Glossary Content for Business Intelligence Reports

IP.com Disclosure Number: IPCOM000200980D
Publication Date: 2010-Nov-02
Document File: 9 page(s) / 100K

Publishing Venue

The IP.com Prior Art Database

Abstract

This article introduces a bottom-up approach for populating a glossary repository with a complete set of terms from Business Intelligence (BI) reports. These terms are automatically determined by their usage in BI reports and after several processing routines (syntax check, already available in glossary, assigning context to terms) stored in the glossary repository. This solution ensures completeness and accuracy for terms being looked up from BI reports as well as periodical updates.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 52% of the total text.

Page 01 of 9

Populate Glossary Content for Business Intelligence Reports

Usually a complete set of business or technical terms is loaded into a glossary repository and a data steward is responsible for defining and maintaining those term. Input terms can either be entered manually or via bulk insert methods e.g. from flat files or spreadsheets. However gathering the input for loading a glossary is a large effort and time consuming. This article introduces a different bottom-up approach for populating the glossary repository with a complete set of terms for use with Business Intelligence (BI) reports:

BI report objects like column titles, data item names, data item labels, series names and query items like names, nomenclatures, group names and categories are typical candidates for being looked up in a glossary.

The terms to be made available in the glossary repository are now automatically determined by their usage in a BI report and searched within the glossary (already existing entries), organizational internal sources or the world wide web. The data steward will decide on which definition from the offered results to use and confirm to store it in the glossary repository.

The advantage of this solution is that a customized glossary content will be available and the completeness for terms being looked up from BI reports can be assured and periodically updated.

1. The data steward provides a list of BI report URLs.

2. The glossary population tool reads data objects from the presentation layer and from the query layer of the BI report definitions.

3a. In case of a presentation layer object it verifies whether the object type is applicable to the glossary. For example column titles, data item names, data item labels and series names are exercisable for the use with a glossary. If the object type is applicable then the term will be extracted from that object.

3b. In case of a query layer object (query items) it verifies whether the results from the query are applicable to the glossary. For example a measure is a numeric field and not exercisable but names, nomenclatures, group names, categories, etc. are exercisable for the use with a glossary. For that reason the query item properties (identifier, fact, attribute) are detected. If the query item property is attribute then it is verified whether the data type is character. If both criteria match an SQL query is executed returning the values for that specific column (e.g. tariff names, product names, product type names, region names or currency keys). This result set now contains the terms for the glossary.

4. A syntax check is done for t...