PATHS at Royal Melbourne Institute of Technology

Personalised Access to Cultural
Heritage Spaces using Pathways

Mark Stevenson
University of Sheffield

Overview

•  Background

•  Information access in cultural heritage
•  PATHS project
–  Processing Cultural Heritage data

Information access in cultural heritage

•  Significant amounts of Cultural Heritage material
available online
–  Web portals, digital libraries, Wikipedia …

•  Users find it difficult to navigate and interpret the
wealth of information
–  users are normally not subject experts
–  systems offer limited support for knowledge exploration and
discovery

PATHS
http://www.paths-project.eu
•  Supporting user’s knowledge
discovery and exploration
•  Use of pathways/trails to
navigate and explore the
information space
•  Personalisation to adapt views/
paths to specific users or
groups of users
•  Links to items within the
information space and
externally to contextualise and
aid interpretation

PATHS: Basic facts

•  STREP funded under the European Commission's
Seventh Framework Programme
•  36 months
–  1st January 2011 to 31st December 2013
•  Budget – 3,199,299 euros in total
–  2,300,000 euros EU grant
•  6 partners in 5 countries

PATHS consortium
•  Universities
–  Sheffield University (co-ordinator)
–  Universidad del Pais Vasco
•  Technology enterprises
–  i-sieve technologies Ltd
–  Asplan Viak Internet Ltd
•  Cultural heritage enterprises
–  MDR Partners
–  Alinari 24 Ore Spa
•  With an additional content provider
–  Europeana

Project objectives
•  Analysis of users’ requirements for discovering knowledge in
Cultural Heritage collections and construction of pathways/trails
•  Automated organisation and enrichment of Cultural Heritage content
for use within a navigation system
•  Implementation of a system for navigating Cultural Heritage
resources
•  Techniques for providing personalised access to Cultural Heritage
content (e.g. recommender systems)
•  Porting the navigation system for use on mobile devices and
Facebook
•  Evaluation with user groups and in field trials

Research areas
•  Information Access
–  user-driven navigation through collections
–  knowledge of users’ requirements for access to cultural heritage
collections
–  modeling of user preferences and context
•  Educational Informatics
–  adapting to individual learners in relation to being directed and
being allowed the freedom to explore autonomously
•  Content Interpretation and Enrichment
–  representation and sharing of information about items in Digital
Libraries
–  identifying background information related to the items in cultural
heritage collections (e.g. links to Wikipedia pages)

Pathways for navigation
•  Navigation through a collection via metaphor of “pathways”

•  A path is a ‘route’ through an information space
–  defined as collections of cultural heritage resources
–  consists of items, links connecting them and narrative

Presentation at Glasgow University, 14th March 2011

Guided and user generated paths
•  Users can follow pre-defined “guided paths”
–  created by domain experts, such as scholars or teachers
•  Provide an easily accessible entry point to the collection
–  can be followed in their entirety or left at any point
•  Users can also create and share their own paths

•  Paths can be based around any theme
–  artist and media (“paintings by Picasso”)
–  historic periods (“the Cold War”)
–  places (“Venice”)
–  famous people (“Muhammed Ali”)
–  or any other topic (e.g. “Europe”, “food”)

Paths and trails have been studied in many fields

•  Trails (Memex, 1945)
–  Associative trails explicitly created by users forming links
between stored materials to help others navigate
•  Destinations (search engines and web analytics)
–  Origin/landing page (from query), intermediate pages and
destination page
•  Search strategies (information seeking)
–  Users moving between information sources, perhaps due to
changes in their information needs
•  Guided tours (hypertext)
–  authors create sequence of pages useful to others (manual)
–  automatically generated trails to assist with web navigation
–  used in educational informatics and cultural heritage

Learning and knowledge discovery
•  A particular area of focus in PATHS
–  Aims to help people to learn and discover new knowledge as
they use cultural heritage resources
•  People learn and solve problems differently
–  some people require a lot of guiding; others are self-directed
–  some people welcome irrelevant material; others are intolerant
–  some people creatively explore and come up with new ideas;
others want to answer a set problem
•  Users may perform information seeking
–  must navigate through information spaces
–  different people may require different levels of assistance

Local (analytic) Global
Learning/problem-solving goals
Convergent goals. Divergent goals.
“Find an answer”. Creatively explore.
Learn pre-defined content. Come up with new ideas.
Process goals
Concerned with procedures Concerned with conceptual overview Adopting a navigation path that
and vertical deep detail and horizontal broad inter- matches one’s predominant style
(procedure building). relationships (description building).
can influence the effectiveness of
Navigation styles
the resultant learning.
Serialist navigation style Holist navigation style
Narrow focus. Broad global focus.
One thing at a time. Many things on the go at the same
Short logical links between time. Autonomous
nodes. Rich links between nodes.
Intolerance of strictly Welcoming of enrichment (but strictly
irrelevant material. irrelevant) material.
Finish with one topic before Layered approach returning to nodes
going on to the next. at different level of detail. Local Global
(analytic)
Positive learning outcomes
Good grasp of detailed Well developed conceptual overview.
evidence. Broad inter-relationship of ideas.
Deep understanding of Good grasp of the “big picture”. Dependent
individual topics.
In-depth understanding of the Key cognitive dimensions (Pask and Witkin)
parts.
Characteristic learning pathologies
Poor appreciation of topic Poor grasp of detail.
inter-relationships. Over-generalisation.
Failing to see the “big
picture”.

Europeana
http://www.europeana.eu
•  Europe’s Digital Library, Museum and Archive
•  Vision of six European heads of state (2005)
•  Launched in November 2008
•  1,500 contributing institutions
•  Over 15 million items

6/22/11 © The University of Sheffield

Europeana data
<record>
<dc:creator>Davies, J O</dc:creator>
<dc:date>[2001]</dc:date>
<dc:title>Stembridge Windmill, High Ham, Somerset</dc:title>
<dc:description>This is a random-coursed blue lias stone tower mill, with a unique thatched cap. Built 1822 to replace an earlier mill
which was sited a few hundred metres to the north east. It ceased work in 1908 and was willed to the National Trust in 1969, since
when quite extensive repairs have been carried out.</dc:description>
<dc:identifier>http://viewfinder.english-heritage.org.uk/search/detail.asp?calledFrom=oai&imageUID=8</dc:identifier>
<dc:language>en</dc:language>
<dc:rights>Copyright English Heritage.NMR</dc:rights>
<dc:subject>Agriculture</dc:subject>
<dc:subject>Windmill</dc:subject>
<dc:subject>Tower Mill</dc:subject>
<dc:type>Image</dc:type>
<dcterms:isPartOf>English Heritage</dcterms:isPartOf>
<europeana:country>uk</europeana:country>
<europeana:dataProvider>English Heritage - Viewfinder</europeana:dataProvider>
<europeana:isShownAt>http://viewfinder.english-heritage.org.uk/search/detail.asp?calledFrom=oai&imageUID=8</
europeana:isShownAt>
<europeana:language>en</europeana:language>
<europeana:object>http://www.culturegrid.org.uk/dpp/resource/1512084/stream/thumbnail_image_jpeg</europeana:object>
<europeana:provider>CultureGrid</europeana:provider>
<europeana:rights>http://www.europeana.eu/rights/rr-f/</europeana:rights>
<europeana:type>IMAGE</europeana:type>
<europeana:uri>http://www.europeana.eu/resolve/record/09405o/AC721E03934115003DEA60494AC9C88441A255ED</
europeana:uri>
<europeana:year>2001</europeana:year>
</record>

Europeana data
•  Positives
–  Consistent format

•  Negatives
–  Field values not standardised
•  Different vocabularies
•  Different levels of detail

–  Very frequent field values

–  URIs are not stable!

Content Processing and Enrichment
•  Linguistic analysis
–  Named entity identification

•  Identify similar and related items

•  Linking to Wikipedia (or other resources)
–  Link entire Europeana entries
–  Link items with entries (eg. named entities)
–  Background context

Determining Similarity

•  Online survey to obtain human similarity
judgements
•  30 pairs of items randomly selected from
Europeana
•  Users asked to rate pair on scale of 0 to 4
–  0 completely unrelated
–  4 almost identical

•  http://compare.net78.net/
•  Over 30 participants so far

Computing Similarity
•  Significant amount of work on computing
lexical similarity in NLP
–  “dog” and “hound” are similar, “cat” and “cap”
are not

•  Approaches include
–  comparing dictionary definitions
–  measuring distance in hierarchy (eg WordNet)
–  mapping to another resource (eg Wikipedia)

Computing Similarity in Europeana
•  Word overlap
–  <dc:title> and <dc:description> fields
•  Map items to Wikipedia
•  Apply graph-based measures

Graph creation

Item:
930075 925673 929829

title: title:
title:
title: necklace woman head
title:
reclining title:
nude
thin
title:
title:
figure
seated

title:
930638 neck 981684

Graph creation

Item:
930075 925673 929829

title: title:
title:
title: necklace woman head
title:
reclining title:
nude
thin
title: artist:
Pablo title:
figure Picasso seated
artist:
Henry
Moore
title:
930638 neck 981684

Conclusions
Paths project:
•  Aim to improve access to large Cultural Heritage
collections
•  Research areas:
–  Information Access
–  Educational Informatics
–  Content Processing and Enrichment
•  Processing Europeana data
–  Determining similarity
–  Mapping to external resources

Contact

Any questions?

info@paths-project.eu
http://www.paths-project.eu/

m.stevenson@dcs.shef.ac.uk

PATHS at Royal Melbourne Institute of Technology

More Related Content

What's hot (19)

Viewers also liked (13)

Similar to PATHS at Royal Melbourne Institute of Technology (20)

More from pathsproject (20)

Recently uploaded (20)

PATHS at Royal Melbourne Institute of Technology