Our Sponsors

Hosting Sponsors

 

Linking Sponsors

 

Supporting Sponsors

 

 

 

 

Media Sponsors

Challenge Entry: Canvas – Exploring Pathways through Collections

Posted by
|

Voting closed 9 May, 2013. 7 Liked

Title: Canvas РExploring Pathways through Collections

Team: Tim Wray

Short Description

Canvas is a re-imagining of the way we experience cultural heritage collections online : navigating pathways, encountering divergences, and finding connections. The project visualises linked data for the curious, wandering explorer.

Long Description

Imagine an online collection as a sprawling canvas of content : works – annotated with gallery-style text-labels – are laid out as pathways, linking thematically groups of works that create a vast landscape of the collection. Over time these pathways unravel and diverge – revealing common themes, artists, styles and movements. The pathways branch and diverge as they reveal semantically meaningful, but sometimes unexpected connections.

The intent of the project is to use museum APIs and data sets to present new and compelling experiential experiences for online collections. The idea heavily embodies the information seeking metaphors of wandering and exploration.

Canvas uses specialised algorithms to create and link these pathways. These algorithms are entirely data-driven: rich and heavily iconographical data-sets generate more intricate and meaningful pathways. Linked Open Data can be used to add layers to existing museum data-sets that can be represented by these pathways. For example, by enriching data about the artists who have produced the works, we can link them together, by their country of origin, or even by the styles and movements that they have worked in. Since the pathway algorithms can handle both structured and unstructured data sources, we can turn to alternate sources, such as social tagging to enrich these data-sets and provide more pathways.

As an investigation, the project looks at ways of using Linked Open Data to enrich existing museum meta-data from a number of public APIs, such as the Brooklyn Museum, the Powerhouse Museum and the Rijksmuseum APIs. I intend link the objects to the people, events and the places that surround them, and visualise these connections as a compelling, engaging browsing experience.

I intend the project to provide a powerful use case in the visual and aesthetic experiences that Linked Open Data can bring to museum collections. If my project is successful and I win the LODLAM challenge, I will undertake the prospect of linking my work to  external data-sets and APIs, and consider new ways of representing connections and browsing experiences that are built from the these enriched data sets.

Canvas is the result of work produced from my own PhD research. The IP for my software framework and raw results derived from my research are retained under myself (Tim Wray) and the University of Wollongong. However, I may choose to open-source part or all of my codebase once I am satisfied with its completeness.

A prototypical demonstrator URL for Canvas can be found at : http://collectionweb.cs.uow.edu.au/brooklynmuseum/vis/canvas/interact/. Please note that the current URL is a demonstrator only, and we recommend that you view it in a recent version of Safari or Chrome.

Add a comment

You must be logged in to comment.

Photo Credits

Montreal skyline photo CC BY from Flickr Manu_H
BAnQ elevator/stairs CC BY-NC-SA from Flickr 917Press