Using Pivots To Explore Heterogeneous Collections: A Case Study in Musicology


Smith, Daniel Alexander, Bretherton, David, schraefel, mc, Polfreman, Richard, Everist, Mark, Brooks, Jeanice and Lambert, Joe (2009) Using Pivots To Explore Heterogeneous Collections: A Case Study in Musicology. At All Hands Meeting 2009, Oxford, 07 - 09 Dec 2009. (Submitted).

Download

[img] PDF
Download (389Kb)
[img] Microsoft PowerPoint
Download (2130Kb)
[img] Video (QuickTime)
Download (4Mb)

Description/Abstract

In order to provide a better e-research environment for musicologists, the musicSpace project has partnered with musicology’s leading data publishers, aggregated and enriched their data, and developed a richly featured exploratory search interface to access the combined dataset. There have been several significant challenges to developing this service, and intensive collaboration between musicologists (the domain experts) and computer scientists (who developed the enabling technologies) was required. One challenge was the actual aggregation of the data itself, as this was supplied adhering to a wide variety of different schemas and vocabularies. Although the domain experts expended much time and effort in analysing commonalities in the data, as data sources of increasing complexity were added earlier decisions regarding the design of the aggregated schema, particularly decisions made with reference to simpler data sources, were often revisited to take account of unanticipated metadata types. Additionally, in many domains a single source may be considered to be definitive for certain types of information. In musicology, this is essentially the case with the “works lists” of composers’ musical compositions given in Grove Music Online (http://www.oxfordmusiconline.com/public/book/omo_gmo), and so for musicSpace, we have mapped all sources to the works lists from Grove for the purposes of exploration, specifically to exploit the accuracy of its metadata in respect to dates of publication, catalogue numbers, and so on. Therefore, rather than mapping all fields from Grove to a central model, it would be far quicker (in terms of development time) to create a system to “pull-in” data from other sources that are mapped directly to the Grove works lists.

Item Type: Conference or Workshop Item (Speech)
Additional Information: Event Dates: 7-9 December 2009
Divisions: Faculty of Physical Sciences and Engineering > Electronics and Computer Science > Agents, Interactions & Complexity
ePrint ID: 267998
Date Deposited: 05 Oct 2009 16:00
Last Modified: 27 Mar 2014 20:14
Further Information:Google Scholar
URI: http://eprints.soton.ac.uk/id/eprint/267998

Actions (login required)

View Item View Item