UK Web Focus (Brian Kelly)

Innovation and best practices for the Web

Linked Data for Events: the IWMW Case Study

Posted by Brian Kelly on 21 Sep 2010

Linked Data and Events

In a post entitled “Getting information about UK HE from Wikipedia” published in July on the Ancient Geek’s blog Martin Poulter commented that “At IWMW 2010, last week, a lot of discussion centred around how, in an increasingly austere atmosphere, we can make more use of free stuff. One category of free stuff is linked data. In particular, I was intrigued by Thom Bunting (UKOLN)‘s presentation about extracting information from Wikipedia.

Martin’s comment related to a Linked Data application developed by my colleague Thom Bunting which he demonstrated in the final session at the IWMW 2010 event.  In this post I would like to summarise this work.

Thom, along with UKOLN colleagues Adrian Stevenson and Mark Dewey, were facilitators for a workshop session at the IWMW 2010 event. In the run-up to the event  I suggested to Thom that it would be useful to exploit the historical  data for UKOLN’s annual Institutional Web Management Workshop (IWMW) series of events.  This event was launched in 1997 and has been held annually ever since, with this year’s event, IWMW 2010, being the 14th in the series.

The Web sites for all 14 events continue to be hosted on the UKOLN Web site and care has been taken to ensure that the URLs for the Web sites have remained persistent.   In the past five years or so as well as providing a series of HTML pages we have also provided a series of RSS files for each event which can be used not only to provide news for the events but also to enable key information sources, including speaker biographies and abstracts of the talks and workshop sessions to be syndicated and reused by other applications. We have recently ensured that such RSS files are available for all of the workshops.

Our intention was to make use of this information in order to develop a Linked Data application which would demonstrate the potential of Linked Data in an application area (organising events) which is of relevance to all higher educational institutions.

Thom has written a post on Consuming and producing linked data in a content management system in which he describes the technical details of how he used Drupal to produce his Linked data application, having first processed the various RSS feeds (which were available in the RSS 2.0 format which is not suitable for Linked Data applications without further processing). In this post I want to highlight some of the benefits to end users which his application provides:

Information on the host institution of participants, speakers and facilitators at events: The RSS 2.0 files had previously been used to provide a Google Map showing the location of speakers and facilitators.  This information had been extended to include the host institution of participants at recent events.  But in addition to the map, clicking on an icon will display information about the numbers of participants from the institution together with information about the institution. The important thing to note is that the institutional information is not held locally; rather this information is gathered from the institution’s entry in DBpedia.

The screen image below (taken from the Locations map area of the IWMW Linked Data application) shows this information, providing details of the speakers and facilitators from the University of Southampton at recent events (this being public information) and a summary  of the total numbers of participants from the institution.   The image also shows the information taken from DBpedia, which includes information that the institution is part of the Russell group, together with details of, for example, the student numbers.

This illustrates access via a specific institution. It is also possible to view such information for all institutions which have participated at recent event events. Illustrated below are the results of such a query sorted by the total number of registrations.

It should be noted that what is referred to as the ‘Loyalty rate’ is calculated as the total registrations / total persons registering. This gives a general indication of how many annual IWMW events, on average, each person registering from a specified institution (or company) has attended.

What Next?

This work has provided UKOLN with a better understanding of the impact the IWMW series of events has had across the HE sector.  We can now see the institutions which have contributed to or attended significant numbers of events and, conversely, those which haven’t. And although such information could be obtained through use of a internal database the integration with the institutional data could not realistically have been achieved without use of a Linked Data approach.

Our next step will be to include information about the speakers and participants at the IWMW events and the topics of their sessions.  As described previously such information is available at stable URIs. However we are waiting for an upgrade to the Drupal software before we begin on this next step.

We hope this summary illustrates some of the benefits which use of Linked Data can provide.

4 Responses to “Linked Data for Events: the IWMW Case Study”

  1. Les Carr said

    Mashing up around individual identities and institutions seems a good idea. Of course lanyrd and crowdvine do some of that in the social domain, but it would be great to develop some community best practice around conference chairs facilitating open and linked data from their access to the administrative information.

    What’s the best forum for doing this? There’s the royal society web science meeting next week (v high profile) and I’d like to make sure we do that right!

  2. I already mailed laynrd about linked data but didn’t get a response.

    Don’t forget our programme schema:

    I’m slowly working on tools for visualising it nicely
    (Still a work in progress)

    I could add a RDF to RSS converter if you are dead set on providing RSS of event streams.

Leave a Reply

Please log in using one of these methods to post your comment: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: