Category: Team

Team: Data Converters

Country: United States

Team Leader: Sammy Davidson

Team Members: Laurence Skirvin, Jeff Mixter

The LOD-LAM KSU Research Group wants to help libraries, archives, and museums integrate linked data into their information systems. The project submitted to the LOD-LAM challenge focuses on the needs of small to medium sized libraries. Smaller libraries may not have the resources, knowledge, or access to participate in and benefit from linked data. In order to do this, we need to create a recipe for these libraries to publish their own linked data resources, generated from their MARC records. Our solution uses existing open source tools and proposes a prototype to automate the conversion of MARC fields to RDF links. This is one of the many activities of the Research Group. http://lod-lam.slis.kent.edu

Team: NTNU University Library

Country: Norway

Team Leader: Rurik Greenall

Team Members: Stein Olle Johansen, Lene Bertheussen, Ove Wolden, Ellen Alm, Ingunn Østgaard, Eva Sauvage

Team working with linked data at NTNU University Library; creating data-driven frameworks for information delivery. Specific areas: acquisitions, cataloguing, content-to-Web. Use cases: archives, manuscripts, photography.

Team: WWI Linked Open Data (WWI LOD) Project

Country: International

Team Leader: Juha Törnroos (Aalto University)

Team Members: Thea Lindquist (University of Colorado Boulder), Eetu Mäkelä (Aalto University), Eero Hyvönen (Aalto University), Rami Aamulehto (Aalto University), Holley Long (University of Colorado Boulder), Michael Ortiz (University of Colorado Boulder), Michael Dulock (University of Colorado Boulder), Martha Hanna (University of Colorado Boulder)

Computer scientists from Aalto University and librarians and historians from the University of Colorado Boulder are collaborating to enhance access to and context for the people, places, events and topics buried in World War I (WWI) primary sources using Linked Open Data. 

Team: DataModelers

Country: United States

Team Leader: Jeff Mixter

The goal of the project was to convert the existing VRA 4 restricted XML schema in a Linked Data data model that incorporated popular vocabularies to the greatest extent. Once the model was completed, an XSLT stylesheet was used to demonstrate how existing XML data could be converted into RDF. For the study, the Notre Dame Lantern Slide collection (which consists of 4,150 records) was used to test the stylesheet. For more information regarding the study and to download the tools developed, please visit http://purl.org/jmixter/thesis

Team: Pundit

Country: International

Team Leader: Simone Fonda

Team Members: Francesca Di Donato, Christian Morbidoni, Sam Leon, Joris Pekel

The team is composed by people from Net7 (a small italian company) and OKFN (Open Knowledge Foundation). Together, with other partners, we are now developing Pundit to be used inside the DM2E (Digital Manuscript to Europeana) project.

Team: LODLAM Patterns

Country: USA

Team Leader: Richard J. Urban

The universe of methods for representing cultural heritage resources is growing rapidly, despite the multiple standards that already exist across the library, archive, and museum domain. These various standards may address common problems, but there is little explicit coordination among the solutions. As Linked Data Principles increasingly allow us to “mix and match” vocabularies, we need a new way to understand the available techniques that solve specific representation problems.

Design patterns are a common tool in software and ontology engineering circles that present a clear definition of common problems, identify available solutions, provide examples, and establish links to related patterns. By providing this clear organization, design patterns facilitate discussion about problems and solutions, rather than debates about the “right” standard to use.

The LODLAM Patterns website will provide a venue for identifying, publishing, and refining what I call “”representation patterns”” for cultural heritage resources. Initial patterns will emerge from an analysis of contemporary cultural heritage metadata standards, but these will only serve to prime the pump. At the LODLAM Summit I will invite interested members of the community to join me in identifying useful patterns and improving published patterns through comments and discussion.

Team: Digital “Terroirs”

Country: France

Team Leader: Adrien Di Mascio

Team Members: Romain Wenz, Vincent Michel, Agnès Simon

“Find digital resources about the French countryside”

Disclaimer: This “challenge team” also works on the data.bnf.fr project, but the problems solved here are parallel to our main project.

Problem: In France, anybody can go down the street and buy a Croissant or a Baguette of bread.
And you can find pictures of the Eiffel tower anywhere. But finding relevant digital resources is sometimes a bit more difficult. It is hard to find online documents about places. Tools have been made to describe historical monuments, but not yet to describe the diversity of geographic entities. Several datasets are available, and there is a wealth of available documents about all places of France. But few of them are properly indexed. This would meet both a strong demand from researchers, and an ongoing tourist interest.

Why LOD?: Linked Open data technologies could be used to map the digital resources from the French national Library (BnF) with geographic information, so as to provide both a proper indexing, and easy-to-use maps that would provide you with relevant digital documents about the French Terroirs. This service would have to address issues such as: named entity extraction, data management, and coordination of research applications with an easy-to-use end-user interface.
Any data available?
The first datasets would be:

– Works and Writers from data.bnf.fr, so as to rely on authority data.

– “Manifestations” from data.bnf.fr for bibliographic information.

– “Rameau” subject headings from data.bnf.fr for topics.

– Digital items from Gallica for the content provided to the end-user.

– Geonames and other geographic services for extracting coordinates.

– Other map services for implementing maps in an end-use interface.

Next steps: It would be necessary run some intelligence on the publication information, so as to extract towns and places. It would also be possible to integrate information from the performing arts, and to show the activities that have taken place throughout the time. Depending on the outcome, it could even be possible to use some OCR on the documents from the “Gallica” digital library, by extracting the named entities.”

Team: Reload

Country: Italy

Team Leader: Silvia Mazzini

Team Members: Brunella Argelli (IBC), Agostino Attanasio (ACS), Ilaria Barbanti (regesta.exe), Giovanni Bruno (regesta.exe), Silvia Mazzini (regesta.exe), Mirella Plazzi (IBC), Francesca Ricci (IBC), Chiara Veninata (ACS)

Partners
Archivio Centrale dello Stato ACS (State Central Archives, http://www.acs.beniculturali.it/) – a body of the Ministry of Cultural Heritage and activities endowed with special autonomy – is the archival depository Institute of the Unified State’s documentary heritage.
Istituto Beni Culturali Regione Emilia Romagna IBC (http://ibc.regione.emilia-romagna.it) is the scientific and technical instrument for the Emilia-Romagna regional planning in the field of artistic, cultural and environmental heritage. IBC develops the IT facilities that convey archives, libraries and museums data to institutions and the general public, promotes and coordinates the census and the description of archival, book and museum material, grants the readability of specific DBs on the web and at present IBC’s working on the standards for interoperability through the use of semantic web technologies.
Regesta.exe (http://www.regesta.com) provides to any type of cultural institution with services and tools, allowing to create, manage, retrieve and access online documentary, iconographic and audiovisual resources.

Goal
Goal of Reload project is to experiment with Semantic Web technical standards and methods relating to linked open data in order to further the sharing of data among a broad range of archives and other cultural institutes. In details, this project is designed to verify the possibility to create a “”web of archival data””, by exploring Semantic Web technologies to link differnt datasets of cultural heritage domain. This experience aims at applying Semantic Web technology to create Linked Open Data of archival descriptions and of entities related to them.

Team: Canvas

Country: Australia

Team Leader: Tim Wray

What ways of seeing can Linked Open Data bring to our collections? Offering unparalleled access to structured knowledge, we often envision it as a giant nebulous cloud, bringing together the silos of data into a conglomeration of nodes, links and connections.

But how can we envision these connections in a meaningful way? How can we represent these knowledge structures at the macro level? How can we create visualisations of Linked Open Data that convey meaning and entice curiousity?

As an independent researcher, I’m interested in creating immersive interactive experiences for museum collections within the digital medium – be that on the Web, the screen or the tablet device.

The physical museum exhibition is crafted by a careful selection of works that exploits the spatial properties of the physical building to present works thematically, chronologically, or by some other means that expresses a narrative. In my work, I rely on the spatial concepts of pathways and divergences as a framework for visualizing my interactive experiences – allowing visitors to explore and browse in ways that mimic a thematically induced exhibition built from concepts mined from data. I’ve built prototypes and concepts that have expressed this ideal, and I’m currently looking to incorporate Linked Open Data so that I can provide enriched data to seed and create more meaningful pathways. The work is intended both as an investigation into new conventions for expressing online collections, and as a case study on how Linked Open Data can be represented in novel and compelling ways.

My work is based on the idea that humans think in terms of concepts, associations and similarity, and that we are fundamentally curious creatures. It is based on over 15 years of research into conceptual clustering and I’m currently investigating how concept lattices can be used to mine, link and depict themes that create landscapes of pathways that are inviting to the curious explorer.