EarthCube's OceanLink - Project Overview and Presentation Updates (March 2014)

0
EarthCube Building Blocks:
OceanLink
Leveraging Semantics and
Linked Data for
Geoscience Data Sharing
and Discovery

1
Overview
Goal
•  Enable discovery of geoscience data and knowledge,
and ultimately, integration
Strategy
•  Publish content from existing network of repositories as
Linked Open Data (LOD)
•  Enable horizontal semantic integration
•  Provide tools + services useful to working scientists

2
Domain
Ocean Science
•  Research vessels collect
data from the solid earth,
water column, atmosphere
•  Many repositories already
interoperate
•  Approach is extensible to
other geo domains
U.S. academic oceanographic
research fleet (above), and
recent expedition tracks (left)

3
Project Team
Lamont-Doherty Earth Observatory
Robert Arko
Suzanne Carbotte
Marymount University
Tom Narock (lead)
University of Maryland, Baltimore County
Tim Finin
Woods Hole Oceanographic Institution
Cynthia Chandler
Lisa Raymond
Adam Shepherd
Peter Wiebe
Wright State University
Pascal Hitzler
Michelle Cheatham
Adila Krisnadhi

4
Collections
•  Biological & Chemical Oceanography Data Management
Office (BCO-DMO)
•  Rolling Deck to Repository (R2R)
•  cruise catalog +underway enviro sensor data
•  Marine Biological Laboratory / Woods Hole Oceanographic
Institution (MBLWHOI) Library
•  published articles, theses, tech reports, datasets
•  AGU meeting abstracts
•  NSF funding award abstracts

5
ODPs
Ontology Design Patterns
•  Core set of conceptual primitives from Ocean Science
Vessel
Cruise
Instrument
Dataset
Person
Organization
etc.
•  Reuse existing standard vocabularies where they exist
(DCAT, FOAF, PROV)
•  Maximize reusability, minimize commitment

6
ODPs
•  Patterns published as OWL files with embedded axioms
and local vocabularies
eg.
Cruise must have a Vessel
Cruise may have a Person in the Role of Chief Scientist
•  Leverage existing alignment among repositories that
use eg. NERC Vocabulary Server
•  Inference to find relationships among cruises, datasets,
people, publications, etc.
(cont.)

7
Work Plan
1.  Model, align, inference over existing LOD collections
(BCO-DMO + R2R)
•  Develop use cases eg. "find publications related to
cruises at the Bermuda Rise that produced
CTD profiles and/or seafloor mapping data”
•  Develop ODPs
•  Map existing collections to ODPs
2.  Publish LOD for other collections (Library, AGU, NSF)
and map to ODPs
3.  Prototype end-user tools and services
•  Search/browse across federated LOD collections
•  Edit ontologies
•  Annotate LOD resources incl. provenance

8
Initial Results
“An Ontology Pattern for Oceanographic Cruises” (Krisnadhi
et al.)
Technical Report and draft set of ODPs
Reuses existing patterns including
•  Semantic Trajectory (Janowicz et al.)
•  Information Object
•  Simple Event Model
to model a Cruise and ship’s track
R/V Atlantis cruise AT22
(Scotian Shelf Survey,
August 2012) Basemap:
GMRT

9
Lessons
•  Recurrent themes in EarthCube Workshop Reports eg.
•  Data are still difficult to discover and access
•  Data attribution and citation are critical
•  Reuse of data still hampered by need for implicit
understanding
•  Collaboration between Geo Science and Computer Science
works best with Use Cases
•  In-person working meetings are key to initial progress
Oct. 2013 Woods Hole
Nov. 2013 Baltimore
Jan. 2014 Washington
(probably more)

10
Acknowledgements
“EAGER: Collaborative Research: EarthCube Building Blocks,
Leveraging Semantics and Linked Data for Geoscience Data
Sharing and Discovery”
NSF Funding Awards:
ICER 13-54990 LDEO
ICER 13-54693 UMBC
ICER 13-54778 WSU
ICER 13-54107 WHOI
September 15, 2013 - August 31, 2014

11
Thank you.
www.oceanlink.org

March 2014 – Montana State University – Pascal Hitzler
Metadata Semantics: What Semantic Web
technologies can contribute to scientific
data and information sharing and
discovery
Pascal Hitzler
DaSe Lab for Data Semantics
http://www.pascal-hitzler.de

March 2014 – Montana State University – Pascal Hitzler 2
Distributing scientific information
• Since the rise of the World Wide Web, the role of publishing
houses and scientific libraries is changing.
• Scientific publishing houses are redefining their roles and are
investigating new revenue models.
• What exactly is the role of libraries?
• What will the role of libraries be in, say, 20 years?

Library discovery issues
• “I’m looking for an easy but introductory text on discrete
mathematics suitable for computer scientists, with high quality
in the mathematical formalization and notation, and including
(besides the usual stuff) at least brief treatments of Russel’s
paradox and of countable versus uncountable sets, e.g.
uncountability of the real numbers.”
• “I’m looking for a textbook for a second-year introductory class
on logic for computer scientists. Formal treatment of
mathematics, tableaux algorithms for propositional and
predicate logic, and preferably some coverage of datalog.”

Semantic Web journal
• EiCs: Pascal Hitzler
Krzysztof Janowicz
• Established 2010. Going strong.
• We very much welcome contributions at
the “rim” of traditional Semantic Web
research – e.g., work which is strongly
inspired by a different field.
• Non-standard (open & transparent)
review process.
• http://www.semantic-web-journal.net/

Summary Statistics

Citation Maps

Collaboration Networks

Topic Trends

Publications analysis
• Provide analysis of citations, topic trends, research networks,
etc., which can be obtained from (suitable!) metadata.
• Establish the social, economical and computational
infrastructure to provide such data: open access, legal
reusability of text and data, rich metadata (citations and
beyond)

Data Discovery

Scenario
Determine if a GMRT grid contains high-resolution data from a
ship’s multibeam sonar in the proximity of a specified
physiographic feature.
Return the list of ship expeditions that contributed high-resolution
data to those grid cells.
For these expeditions, determine which, if any, are found in the R2R
catalog and contain quality-controlled geophysical (gravity/
magnetics) profiles along the same ship track.
Further determine which investigators are linked to those
expeditions; which expeditions and investigators are linked to
journal publications and/or meeting abstracts that contain thematic
keywords pertaining to the physiographic feature; and what other
data are available from the same expeditions in other repositories
such as BCO-DMO.

“Inside” and beyond the publications
• Make paper contents available through rich metadata.
• Combine papers with data and datasets, and with information
from “outside” the publishing process proper, such as funding
awards, geographical information, affiliations, etc.
• More importantly, help in providing a social, economical and
technological infrastructure where such information is provided
to scientists and students.

EarthCube
EarthCube:
Developing a Community-Driven Data and Knowledge Environment
for the Geosciences
“concepts and approaches to create integrated data management
infrastructures across the Geosciences.”
“EarthCube aims to create a well-connected and facile environment
to share data and knowledge in an open, transparent, and inclusive
manner, thus accelerating our ability to understand and predict the
Earth system.”

OceanLink
• An EarthCube Building Block
• Integrating ocean science respositories BCO-DMO and R2R, as
well as datasets from the WHOI Library, AGU abstracts, NSF
projects.
• Demonstrable added value (faceted integrated search).
• Key: extensible architecture that has the potential to grow to
EarthCube size

Integration approach
• Well-established:
– using controlled vocabularies
– which are standardized through a social process
• How many vocabularies do you need to
– answer circumstantial queries?
– cover all scientific paper contents?
– even just to cover the earth sciences?
• What do you do if scientific notions or perspectives change?

E.g., “Event”
Event
xsd:dateTime xsd:string
occursAtPlaceoccursAtTime

Better Event (more general)
Event
<TemporalThing> <SpatialThing>
But what about events taking place in Second Life?

Perhaps even …
Event
<TemporalThing> <Place>
<Agent>
hasParticipant

Different representations
person/101396
“Smith, John”
name
R2R:
foaf:Person
type Person_752
name
foaf:Person
type
“John Smith”
familyName
“Smith”
givenName
“John”
BCO-DMO:
What about other countries?

Semantic Web
• Research field in computer science.
• Took off in ca. the year 2000.
• Significant funding, initially DARPA, then large-scale in the EU.
• In the meantime, large international effort, with significant
investment by funding agencies and companies.
• The Semantic Web vision is about seamless integration of data,
knowledge, and services. It is not restricted to the WWW.
• The Semantic Web approach has (whatever type of) formal
knowledge representation as a key ingredient.

Knowledge Representation
• Vocabularies on steroids.
– Complex relationships between notions are part of the formal
and machine-processable vocabulary definitions,
e.g. “Every cruise must have a chief scientist who is PI on
one of the research awards which pays for the expenses of
the cruise.”
• Standardization of languages for defining vocabularies.
E.g., the Web Ontology Language OWL.
– Rather than standardizing vocabularies themselves.
– Requires establishing best practices for defining and sharing
vocabularies.

Libraries?
• Libraries could again be at the forefront of being providers for
scientific information.
• Trends go towards integrated information spaces with a plethora
of differing and heterogeneous information sources.
• How to organize this information space conceptually,
technologically, and socially, is a key quest in the Big Data age.

Thanks!

OceanLink Collaborators
Robert Arko, Columbia University
Suzanne Carbotte, Columbia University
Cynthia Chandler, Woods Hole Oceanographic Institution
Michelle Cheatham, Wright State University
Timothy Finin, University of Maryland, Baltimore County
Pascal Hitzler, Wright State University
Krzysztof Janowicz, University of California, Santa Barbara
Adila Krisnadhi, Wright State University
Thomas Narock, Marymount University
Lisa Raymond, Woods Hole Oceanographic Institution
Adam Shepherd, Woods Hole Oceanographic Institution
Peter Wiebe, Woods Hole Oceanographic Institution
Some of the presented work is part of the NSF OceanLink project:
EarthCube Building Blocks, Leveraging Semantics and Linked Data
for Geoscience Data Sharing and Discovery

References
• Pascal Hitzler, Frank van Harmelen, A reasonable Semantic Web.
Semantic Web 1 (1-2), 39-44, 2010.
• Prateek Jain, Pascal Hitzler, Peter Z. Yeh, Kunal Verma, Amit P.
Sheth, Linked Data is Merely More Data. In: Dan Brickley, Vinay K.
Chaudhri, Harry Halpin, Deborah McGuinness: Linked Data Meets
Artificial Intelligence. Technical Report SS-10-07, AAAI Press,
Menlo Park, California, 2010, pp. 82-86. ISBN 978-1-57735-461-1.
Proceedings of LinkedAI at the AAAI Spring Symposium, March
2010.
• Pascal Hitzler, Krzysztof Janowicz, What’s Wrong with Linked
Data? http://blog.semantic-web.at/2012/08/09/whats-wrong-with-
linked-data/ , August 2012.
• Pascal Hitzler, Markus Krötzsch, Sebastian Rudolph, Foundations
of Semantic Web Technologies. Chapman and Hall/CRC Press,
2009.

References
• Pascal Hitzler, Krzysztof Janowicz, Linked Data, Big Data, and the 4th
Paradigm. Semantic Web 4 (3), 2013, 233-235.
• Krzysztof Janowicz, Pascal Hitzler, The Digital Earth as Knowledge
Engine. Semantic Web 3 (3), 213-221, 2012.
• Gary Berg-Cross, Isabel Cruz, Mike Dean, Tim Finin, Mark Gahegan,
Pascal Hitzler, Hook Hua, Krzysztof Janowicz, Naicong Li, Philip
Murphy, Bryce Nordgren, Leo Obrst, Mark Schildhauer, Amit Sheth,
Krishna Sinha, Anne Thessen, Nancy Wiegand, Ilya Zaslavsky,
Semantics and Ontologies for EarthCube. In: K. Janowicz, C. Kessler,
T. Kauppinen, D. Kolas, S. Scheider (eds.), Workshop on GIScience in
the Big Data Age, In conjunction with the seventh International
Conference on Geographic Information Science 2012 (GIScience 2012),
Columbus, Ohio, USA. September 18th, 2012. Proceedings.
• Krzysztof Janowicz, Pascal Hitzler, Thoughts on the Complex Relation
Between Linked Data, Semantic Annotations, and Ontologies. In: Paul
N. Bennett, Evgeniy Gabrilovich, Jaap Kamps, Jussi Karlgren (eds.),
Proceedings of the 6th International Workshop on Exploiting Semantic
Annotation in Information Retrieval, ESAIR 2013, ACM, San Francisco,
2013, pp. 41-44.

References
• Prateek Jain, Pascal Hitzler, Amit P. Sheth, Kunal Verma, Peter Z. Yeh,
Ontology Alignment for Linked Open Data. In P. Patel-Schneider, Y.
Pan, P. Hitzler, P. Mika, L. Zhang, J. Pan, I. Horrocks, B. Glimm (eds.),
The Semantic Web - ISWC 2010. 9th International Semantic Web
Conference, ISWC 2010, Shanghai, China, November 7-11, 2010,
Revised Selected Papers, Part I. Lecture Notes in Computer Science
Vol. 6496. Springer, Berlin, 2010, pp. 402-417.
• Amit Krishna Joshi, Prateek Jain, Pascal Hitzler, Peter Z. Yeh, Kunal
Verma, Amit P. Sheth, Mariana Damova, Alignment-based Querying of
Linked Open Data. In: Meersman, R.; Panetto, H.; Dillon, T.; Rinderle-
Ma, S.; Dadam, P.; Zhou, X.; Pearson, S.; Ferscha, A.; Bergamaschi, S.;
Cruz, I.F. (eds.), On the Move to Meaningful Internet Systems: OTM
2012, Confederated International Conferences: CoopIS, DOA-SVI, and
ODBASE 2012, Rome, Italy, September 10-14, 2012, Proceedings, Part
II. Lecture Notes in Computer Science Vol. 7566, Springer, Heidelberg,
2012, pp. 807-824.

References
• Yingjie Hu, Krzysztof Janowicz, David Carral, Simon Scheider, Werner
Kuhn, Gary Berg-Cross, Pascal Hitzler, Mike Dean, Dave Kolas, A Geo-
Ontology Design Pattern for Semantic Trajectories. In: Thora Tenbrink,
John G. Stell, Antony Galton, Zena Wood (Eds.): Spatial Information
Theory - 11th International Conference, COSIT 2013, Scarborough, UK,
September 2-6, 2013. Proceedings. Lecture Notes in Computer
Science Vol. 8116, Springer, 2013, pp. 438-456.
• Yingjie Hu, Krzysztof Janowicz, Grant McKenzie, Kunal Sengupta,
Pascal Hitzler, A Linked Data-driven Semantically-enabled Journal
Portal for Scientometrics. In: H. Alani, L. Kagal, A. Fokoue, P. Groth, C.
Biemann, J.X. Parreira, L. Aroyo, N. Noy, C. Welty, K. Janowicz (eds.),
The Semantic Web - ISWC 2013. 12th International Semantic Web
Conference, Sydney, NSW, Australia, October 21-25, 2013,
Proceedings, Part II. Lecture Notes in Computer Science Vol. 8219,
Springer, Heidelberg, 2013, pp. 114-129.

References
• Prateek Jain, Peter Z. Yeh, Kunal Verma, Reymonrod G. Vasquez,
Mariana Damova, Pascal Hitzler, Amit P. Sheth, Contextual
Ontology Alignment of LOD with an Upper Ontology: A Case
Study with Proton. In: Grigoris Antoniou, Marko Grobelnik, Elena
Paslaru Bontas Simperl, Bijan Parsia, Dimitris Plexousakis,
Pieter De Leenheer, Jeff Pan (Eds.): The Semantic Web:
Research and Applications - 8th Extended Semantic Web
Conference, ESWC 2011, Heraklion, Crete, Greece, May 29-June
2, 2011, Proceedings, Part I. Lecture Notes in Computer Science
6643, Springer, 2011, pp. 80-92.
• Prateek Jain, Pascal Hitzler, Kunal Verma, Peter Yeh, Amit Sheth,
Moving beyond sameAs with PLATO: Partonomy detection for
Linked Data. In: Ethan V. Munson, Markus Strohmaier (Eds.):
23rd ACM Conference on Hypertext and Social Media, HT '12,
Milwaukee, WI, USA, June 25-28, 2012. ACM, 2012, pp. 33-42.

References
• Sebastian Rudolph, Markus Krötzsch, Pascal Hitzler, Cheap Boolean
Role Constructors for Description Logics. In: Steffen Hölldobler and
Carsten Lutz and Heinrich Wansing (eds.), Proceedings of 11th
European Conference on Logics in Artificial Intelligence (JELIA),
volume 5293 of LNAI, pp. 362-374. Springer, September 2008.
• Adila Alfa Krisnadhi, Frederick Maier, Pascal Hitzler, OWL and Rules.
In: A. Polleres, C. d'Amato, M. Arenas, S. Handschuh, P. Kroner, S.
Ossowski, P.F. Patel-Schneider (eds.), Reasoning Web. Semantic
Technologies for the Web of Data. 7th International Summer School
2011, Galway, Ireland, August 23-27, 2011, Tutorial Lectures. Lecture
Notes in Computer Science Vol. 6848, Springer, Heidelberg, 2011,
pp. 382-415.
• Adila Krisnadhi, Robert Arko, Suzanne Carbotte, Cynchia Chandler,
Michelle Cheatham, Timothy Finin, Pascal Hitzler, Krzysztof
Janowicz, Thomas Narock, Lisa Raymond, Adam Shepherd, Peter
Wiebe, An Ontology Pattern for Oceanograhic Cruises: Towards an
Oceanograhper's Dream of Integrated Knowledge Discovery.
OceanLink Technical Report 2014.1.

January 2014 – Ontology Summit – Pascal Hitzler
Towards ontology patterns for ocean
science repository integration
Pascal Hitzler

January 2014 – Ontology Summit – Pascal Hitzler 2
Collaborators
The presented work is part of the NSF OceanLink project:

OceanLink and EarthCube
EarthCube:
for the Geosciences
Earth system.”

OceanLink
Bottom-up constructed project.
Currently first phase:
• Integrating ocean science respositories BCO-DMO and R2R, as
well as datasets from the WHOI Library, AGU abstracts, NSF
projects.
• Demonstrable added value (faceted integrated search).
• Key: extensible architecture that has the potential to grow to
EarthCube size

Logic
Many axioms / strong theory
Few axioms / weak theory
Few models
Many inferences
Many models
Few inferences

Ontologies
Strong / many ontological
commitments
Weak / few ontological
commitments
Few models
Many inferences
Not very reusable
Many models
Few inferences
More easily reusable

Strong / many ontological
commitments
Weak / few ontological
commitments
Few models
Many inferences
Not very reusable
Many models
Few inferences
More easily reusable

“An ontology design pattern is a reusable successful solution to a
recurrent modeling problem.”
So-called content patterns usually encode specific abstract
notions, such as process, event, agent, etc.

E.g., “Event”
Event
xsd:dateTime xsd:string

Better Event (more general)
Event
<TemporalThing> <SpatialThing>
This is a
pattern!
But what about events taking place in Second Life?

Perhaps even …
Event
<Agent>
hasParticipant

Event
<Place>
occursAtPlace
Event
xsd:string
occursAtPlace
Shortcuts / views
xsd:string
hasName
There are several things wrong here!

Event
<Place>
a:occursAtPlace
Event
xsd:string
b:occursAtPlace
Shortcuts / views
xsd:string
a:hasName
Better, but …

Event
<Place>
a:occursAtPlace
Event
xsd:string
b:occursAtPlace
Shortcuts / views
xsd:string
a:hasName
The latter is not in OWL!

Event
<Place>
a:occursAtPlace
Shortcuts / views
xsd:string
a:hasName
The latter is not in OWL!
b:occursAtPlace

Similar problem
Splitting a role:
hasParent
hasFather
hasMother

Cruise
For us: ocean science cruise.
A cruise is a type of event.
But what kind of place does it occur at?

Cruise
Cruise
<Agent>
hasParticipant

Semantic
Trajectories
[Hu, Janowicz, Carral, Scheider, Kuhn, Berg-Cross, Hitzler, Dean, COSIT2013]

Semantic Trajectories

Semantics in OWL

Ocean Science Cruise (draft)

Cruise trajectory (draft)

Cruise trajectory

Why ODPs?
Traditionally, ODPs are thought of as building blocks for ontology
modeling.
This idea is certainly valid in the context of special purpose
ontology-based systems.
However, it can be argued that ODPs can be much more than mere
building blocks.

Horizontal alignment
“Horizontal” alignment via patterns
Pattern1
Pattern1
Pattern2
Pattern2
Pattern2
Pattern3
Pattern3

OceanLink setup
OceanLink Patterns
R2R BCO-DMO MBLWHOI
Library
AGU NSF
UI Views
User Interface
mappings

Other added values of patterns
• Pattern-driven GUIs
• Pattern-driven mapping tools
• Pattern-driven query rewriting
• Pattern-driven reasoning modularization
• …

OceanLink setup
EarthCube Patterns
repository repository repository repository repository
UI Views
User Interface
mappings

Thanks!

References
• BCO-DMO: Biological & Chemical Oceanography Data
Management Office, http://www.bco-dmo.org/
• R2R: Rolling Deck to Repository, http://www.rvdata.us
• OceanLink website and publications are forthcoming
• Yingjie Hu, Krzysztof Janowicz, David Carral, Simon Scheider,
Werner Kuhn, Gary Berg-Cross, Pascal Hitzler, Mike Dean, Dave
Kolas, A Geo-Ontology Design Pattern for Semantic Trajectories.
In: Thora Tenbrink, John G. Stell, Antony Galton, Zena Wood
(Eds.): Spatial Information Theory - 11th International
Conference, COSIT 2013, Scarborough, UK, September 2-6, 2013.
Proceedings. Lecture Notes in Computer Science Vol. 8116,
Springer, 2013, pp. 438-456.
• http://ontologydesignpatterns.org

General References
Semantic Web 1 (1-2), 39-44, 2010.
2010.
2009.

General References
• Pascal Hitzler, Krzysztof Janowicz, Linked Data, Big Data, and the
4th Paradigm. Semantic Web 4 (3), 2013, 233-235.
Semantics and Ontologies for EarthCube. In: K. Janowicz, C.
Kessler, T. Kauppinen, D. Kolas, S. Scheider (eds.), Workshop on
GIScience in the Big Data Age, In conjunction with the seventh
International Conference on Geographic Information Science 2012
(GIScience 2012), Columbus, Ohio, USA. September 18th, 2012.
Proceedings.
• Krzysztof Janowicz, Pascal Hitzler, Thoughts on the Complex
Relation Between Linked Data, Semantic Annotations, and
Ontologies. In: Paul N. Bennett, Evgeniy Gabrilovich, Jaap Kamps,
Jussi Karlgren (eds.), Proceedings of the 6th International
Workshop on Exploiting Semantic Annotation in Information
Retrieval, ESAIR 2013, ACM, San Francisco, 2013, pp. 41-44.

General References
• Sebastian Rudolph, Markus Krötzsch, Pascal Hitzler, Cheap
Boolean Role Constructors for Description Logics. In: Steffen
Hölldobler and Carsten Lutz and Heinrich Wansing (eds.),
Proceedings of 11th European Conference on Logics in Artificial
Intelligence (JELIA), volume 5293 of LNAI, pp. 362-374. Springer,
September 2008.
• Adila Alfa Krisnadhi, Frederick Maier, Pascal Hitzler, OWL and
Rules. In: A. Polleres, C. d'Amato, M. Arenas, S. Handschuh, P.
Kroner, S. Ossowski, P.F. Patel-Schneider (eds.), Reasoning
Web. Semantic Technologies for the Web of Data. 7th
International Summer School 2011, Galway, Ireland, August 23-
27, 2011, Tutorial Lectures. Lecture Notes in Computer Science
Vol. 6848, Springer, Heidelberg, 2011, pp. 382-415.

March 2014 – GeoVoCampSB – Pascal Hitzler
OceanLink: Using Patterns for Discovery
in EarthCube
Pascal Hitzler

March 2014 – GeoVoCampSB – Pascal Hitzler 2
OceanLink Collaborators
The presented work is part of the NSF OceanLink project:

Classical ontology-based
integration
Query Upper level
ontology
LOD
IMDB
Dataset
LOD
Wikipedia
Dataset
(DBPedia)
Answer
[ODBASE 2012, JWS 2007]

Example querying LoD
“Identify congress members, who have voted “No” on pro
environmental legislation in the past four years, with
high-pollution industry in their congressional districts.”
In principle, all the knowledge is there:
• GovTrack
• GeoNames
• DBPedia
• US Census
But even with LoD we cannot answer this query.

Some missing puzzle pieces:
• Where is the data?
– GovTrack
GeoNames
US Census
requires intimate knowledge of the LoD data sets

(smart federation needed)
• Missing background (schema) knowledge.
(enhancements of the LoD cloud)
• Crucial info still hidden in texts.
(ontology learning from texts)
• Added reasoning capabilities (e.g., spatial).
(new ontology language features)

Linked Data: Variety
“Nancy Pelosi voted in favor of the Health Care Bill.”
Bills:h3962
H.R. 3962: Affordable
Health Care for America
Act
Votes:2009-887/+
people/P000197
Nancy Pelosi
On Passage: H R
3962 Affordable
Health Care for
America Act
Vote:
2009-887
vote:hasAction
vote:vote
dc:title
vote:hasOption
rdfs:label
Aye
dc:title
vote:votedBy
name

Querying approach
Works very well, but only in some very limited cases.
Cannot deal with graph representations of even
very minimal complexity.

Automated federation?
person/101396
“Smith, John”
name
R2R:
foaf:Person
type
Person_752
name
foaf:Person
type
“John Smith”
familyName
“Smith”
givenName
“John”
BCO-DMO:

Ways forward?
How to establish a flexible conceptual architecture using data and
ontological modeling?

Patterns provide modular, reusable, replaceable, pieces.
By agreeing on reuse of generic patterns (but leaving the
relationships between the patterns to a specific assembly for a
special purpose), we can have reuse while preserving
heterogeneity.

• Bottom-up homogenization of data representation.
• Avoidance of strong ontological commitments.
• Avoidance of standardization of specific modeling details.
• Well thought-out patterns can be very strong and versatile, thus
serve many needs.
We are currently establishing many geo-patterns in a series of
hands-on workshops, the GeoVoCamps, see http://vocamp.org/

Pattern1
Pattern1
Pattern2
Pattern2
Pattern2
Pattern3
Pattern3

EarthCube
EarthCube:
for the Geosciences
Earth system.”

OceanLink
NSF EarthCube project “OceanLink”:
• Integration of existing ocean science data repositories.
• For faceted browsing and semantic search.
• To be done in a flexible, extendable, modular way.
• With minimal effort for additional data providers to integrate
their content.
National Science Foundation award 1354778 "EAGER: Collaborative Research: EarthCube Building Blocks,
Leveraging Semantics and Linked Data for Geoscience Data Sharing and Discovery."

OceanLink setup
OceanLink Patterns
R2R BCO-DMO WHOI
Library
AGU NSF
additional application-specific modeling
User Interface
mappings

OceanLink patterns
Some central patterns:
• Cruise
• Trajectory
• Person
• Organization
• Roles of Agents
• Repository Object
• Data Set
• Document
We’re not starting from zero of course.

Cruise trajectory

Ways forward
• Establish a flexible conceptual architecture using data and
ontological modeling.
• A principled use of patterns, including
– the development of a theory of patterns and
– the provision of a critical amount of central patterns
may provide a primary path forward.

Thanks!

Semantic
Trajectories

Semantics in OWL

References
Semantic Web 1 (1-2), 39-44, 2010.
2010.
2009.

References
2013, pp. 41-44.

References
• Prateek Jain, Pascal Hitzler, Amit P. Sheth, Kunal Verma, Peter Z. Yeh, Ontology
Alignment for Linked Open Data. In P. Patel-Schneider, Y. Pan, P. Hitzler, P. Mika,
L. Zhang, J. Pan, I. Horrocks, B. Glimm (eds.), The Semantic Web - ISWC 2010. 9th
International Semantic Web Conference, ISWC 2010, Shanghai, China, November
7-11, 2010, Revised Selected Papers, Part I. Lecture Notes in Computer Science
• Amit Krishna Joshi, Prateek Jain, Pascal Hitzler, Peter Z. Yeh, Kunal Verma, Amit
P. Sheth, Mariana Damova, Alignment-based Querying of Linked Open Data. In:
Meersman, R.; Panetto, H.; Dillon, T.; Rinderle-Ma, S.; Dadam, P.; Zhou, X.;
Pearson, S.; Ferscha, A.; Bergamaschi, S.; Cruz, I.F. (eds.), On the Move to
Meaningful Internet Systems: OTM 2012, Confederated International Conferences:
CoopIS, DOA-SVI, and ODBASE 2012, Rome, Italy, September 10-14, 2012,
Proceedings, Part II. Lecture Notes in Computer Science Vol. 7566, Springer,
Heidelberg, 2012, pp. 807-824.
• Yingjie Hu, Krzysztof Janowicz, David Carral, Simon Scheider, Werner Kuhn, Gary
Berg-Cross, Pascal Hitzler, Mike Dean, Dave Kolas, A Geo-Ontology Design
Pattern for Semantic Trajectories. In: Thora Tenbrink, John G. Stell, Antony Galton,
Zena Wood (Eds.): Spatial Information Theory - 11th International Conference,
COSIT 2013, Scarborough, UK, September 2-6, 2013. Proceedings. Lecture Notes
in Computer Science Vol. 8116, Springer, 2013, pp. 438-456.

References
6643, Springer, 2011, pp. 80-92.

References
• D. Oberle, A. Ankolekar, P. Hitzler, P. Cimiano, M. Sintek, M. Kiesel,
B. Mougouie, S. Vembu, S. Baumann, M. Romanelli, P. Buitelaar, R.
Engel, D. Sonntag, N. Reithinger, B. Loos, R. Porzel, H.-P. Zorn, V.
Micelli, C. Schmidt, M. Weiten, F. Burkhardt, J. Zhou, DOLCE ergo
SUMO: On Foundational and Domain Models in the SmartWeb
Integrated Ontology (SWIntO). Journal of Web Semantics: Science,
Services and Agents on the World Wide Web 5 (3), 2007, 156-174.
• Adila Krisnadhi, Robert Arko, Suzanne Carbotte, Cynchia Chandler,
Michelle Cheatham, Timothy Finin, Pascal Hitzler, Krzysztof
Janowicz, Thomas Narock, Lisa Raymond, Adam Shepherd, Peter
Wiebe, An Ontology Pattern for Oceanograhic Cruises: Towards an
Oceanograhper's Dream of Integrated Knowledge Discovery.
OceanLink Technical Report 2014.1.

References
• Sebastian Rudolph, Markus Krötzsch, Pascal Hitzler, Cheap Boolean
Role Constructors for Description Logics. In: Steffen Hölldobler and
Carsten Lutz and Heinrich Wansing (eds.), Proceedings of 11th
European Conference on Logics in Artificial Intelligence (JELIA),
volume 5293 of LNAI, pp. 362-374. Springer, September 2008.
• Adila Alfa Krisnadhi, Frederick Maier, Pascal Hitzler, OWL and Rules.
In: A. Polleres, C. d'Amato, M. Arenas, S. Handschuh, P. Kroner, S.
Ossowski, P.F. Patel-Schneider (eds.), Reasoning Web. Semantic
Technologies for the Web of Data. 7th International Summer School
2011, Galway, Ireland, August 23-27, 2011, Tutorial Lectures. Lecture
Notes in Computer Science Vol. 6848, Springer, Heidelberg, 2011,
pp. 382-415.

January 2014 – IBM – Pascal Hitzler
Ontologies in a Data-driven World
Pascal Hitzler

January 2014 – IBM – Pascal Hitzler 2
Textbook
Pascal Hitzler, Markus Krötzsch,
Sebastian Rudolph
Foundations of Semantic Web
Technologies
Chapman & Hall/CRC, 2010
Choice Magazine Outstanding Academic
Title 2010 (one out of seven in Information
& Computer Science)
http://www.semantic-web-book.org

Textbook – Chinese translation
Pascal Hitzler, Markus Krötzsch, Sebastian Rudolph
语义Web技术基础
Tsinghua University Press (清华大学出版社), 2013.
Translators:
Yong Yu, Haofeng Wang, Guilin Qi (俞勇，王昊奋，漆桂林)
http://www.semantic-web-book.org

• EiCs: Pascal Hitzler
Krzysztof Janowicz
• New journal with significant uptake.
• We very much welcome contributions at
the “rim” of traditional Semantic Web
research – e.g., work which is strongly
inspired by a different field.
• Non-standard (open & transparent)
review process.
• http://www.semantic-web-journal.net/

Ontologies?

• ... Agent 1
Thing
Person 2
Ontology
description
Agent
2
exchange of symbols
‘‘Duck“
Concept
MA1
HA1 HA2
MA2
Symbol
Specific Domain,
e.g. Animals
agreementOntology
Semantics
Person
1
exchange of symbols
agreement
A Basic Idea of the Semantic Web

Ontology
represents
general domain knowledge
Reconciling OWL and Rules
Knorr, Hitzler, Maier
ECAI 2012
Data e.g. on
Websites
e.g. every publication has an author

ECAI 2012
Publication
Event
Title
Author

Ontology
represents
general domain knowledge
Data e.g. on
Websites
ECAI 2012

The ontology hype
• Large, well-thought-out ontologies (foundational/domain/etc).
• Networked, interlinked ontologies
• “You just have to get your formal definitions right, and a lot of
the rest will just fall into place.”

The ontology hype
• Large, well-thought-out ontologies (foundational/domain/etc).
• Networked, interlinked ontologies
• “You just have to get your formal definitions right, and a lot of
the rest will just fall into place.”
– This does not even work for
• scientists
• wanting to share and reuse scientific data
• through well-kept data repositories
– So how is this supposed to work for the web at large?

Multiple perspectives
• Try to find a universal definition for
– Forest
– Mountain
– City
– River
– Etc.
• The stronger our ontological commitments, the more we loose
reusability.
• We need to accept that conceptualizations are often very local,
resulting in “micro-ontologies”.

Multiple perspectives
Two ontologies.
Left: transportation domain
Right: agriculture domain
We cannot simply equate a:Canal and b:Canal !

The well-done ontologies
• Brittle
• Expensive
• Sometimes unintuitive
• Unwieldy
• Single-perspective
• Difficult to reuse
• Work in some contexts.
• Work if a lot of central control is imposed.
• Take a lot of manpower.

Pre-LOD Semantic Web
• Foundational ontologies
• Networked ontologies
• Sophisticated ontology languages
Scientific Hypothesis:
These will solve your data and information management problems
Remember that scientific progress is fundamentally about
falsification, not verification 

Linked Data?

The linked data counter-hype
• “Ontologies don’t work, let’s just link data”
• “Okay, with a little bit of ontologies on top.”
• “The Linked Data Web is the true Semantic Web.”

Linked Data 2011

DBpedia: LOTR page

Information as RDF graph
LOTR hasAuthor Tolkien .
Hobbit hasAuthor Tolkien .
LOTR hasCharacter Bilbo .
Hobbit hasCharacter Bilbo .
LOTR
Hobbit
Tolkien
Bilbo
hasAuthor
hasAuthor
hasCharacter
hasCharacter

Linked Data: Volume
Number of Datasets
2011-09-19 295
2010-09-22 203
2009-07-14 95
2008-09-18 45
2007-10-08 25
2007-05-01 12
Number of triples (Sept 2011)
31,634,213,770
with 503,998,829 out-links
From http://www4.wiwiss.fu-berlin.de/lodcloud/state/

Linked Data: Volume
Geoindexed Linked Data – courtesy of Krzysztof Janowicz
http://stko.geog.ucsb.edu/location_linked_data

Linked Data: Volume
October 2013:
Ca. 25,000,000,000 schema.org references on the web.
15% of all pages now have schema.org markup.
That’s just schema.org references …

In principle, all the knowledge is there:
• GovTrack
• GeoNames
• DBPedia
• US Census
But even with LoD we cannot answer this query.

– GovTrack
GeoNames
US Census
requires intimate knowledge of the LoD data sets

(smart federation needed)
• Missing background (schema) knowledge.
(enhancements of the LoD cloud)
• Crucial info still hidden in texts.
(ontology learning from texts)
• Added reasoning capabilities (e.g., spatial).
(new ontology language features)

Linked Data: Variety
“Nancy Pelosi voted in favor of the Health Care Bill.”
Bills:h3962
H.R. 3962: Affordable
Health Care for America
Act
Votes:2009-887/+
people/P000197
Nancy Pelosi
On Passage: H R
3962 Affordable
Health Care for
America Act
Vote:
2009-887
vote:hasAction
vote:vote
dc:title
vote:hasOption
rdfs:label
Aye
dc:title
vote:votedBy
name

Linked Data federated querying
Query Upper level
ontology
LOD
IMDB
Dataset
LOD
Wikipedia
Dataset
(DBPedia)
Answer
Joshi, Jain, Hitzler et al. ODBASE 2012

Bootstrapping-based alignment
Jain, Hitzler et al, ISWC2010

Linked Data federated querying
Query Upper level
ontology
LOD
IMDB
Dataset
LOD
Wikipedia
Dataset
(DBPedia)
Answer
Joshi, Jain, Hitzler et al. ODBASE 2012

ALOQUS Illustration
“Identify films, the nations where they were shot and the population
of these countries”
SELECT ?film ?nation ?pop
WHERE {
?film protonu:ofCountry ?nation.
?film rdf:type protonu:Movie.
?film rdfs:label ?film_name.
?nation protont:populationCount ?pop.
}

Querying approach
Works very well, but only in some very limited cases.
Cannot deal with graph representations of even
very minimal complexity.

person/101396
“Smith, John”
name
R2R:
foaf:Person
type
Person_752
name
foaf:Person
type
“John Smith”
familyName
“Smith”
givenName
“John”
BCO-DMO:

Copernicus lunar crater located on earth – courtesy of Krzysztof
Janowicz http://stko.geog.ucsb.edu/location_linked_data
(missing reference coordinate system)

The linked data counter-hype
• “Ontologies don’t work, let’s just link data”
• “Okay, with a little bit of ontologies on top.”
• But then we don’t even know how to effectively query over
multiple linked datasets (without using a lot of manpower to
manually integrate them).
• It seems rather obvious that we need to get ontologies into the
picture, but how to do it while avoiding the drawbacks of strong
ontological commitments?

So What Now?

Ways forward?
How to establish a flexible conceptual architecture using data and
ontological modeling?

• Bottom-up homogenization of data representation.
• Avoidance of strong ontological commitments.
• Avoidance of standardization of specific modeling details.
• Well thought-out patterns can be very strong and versatile, thus
serve many needs.
We are currently establishing many geo-patterns in a series of
hands-on workshops, the GeoVoCamps, see http://vocamp.org/

Pattern1
Pattern1
Pattern2
Pattern2
Pattern2
Pattern3
Pattern3

Semantic
Trajectories

Semantics in OWL

Helpfulness of patterns
person/101396
“Smith, John”
name
R2R:
foaf:Person
type
Person_752
name
foaf:Person
type
“John Smith”
familyName
“Smith”
givenName
“John”
BCO-DMO:Even minimalistic reuse is
helpful:

Patterns
• Help to focus when modeling (one key notion at a time).
• Good ontology modeling implicitly employs the patterns idea
anyway. It’s just that you expose the patterns.
• An ontology composed of patterns exposes its internal
conceptual structure (as a composition of formal vocabulary
pieces).
• Well-designed patterns are widely reusable and adaptable.
• You don’t have to buy a whole ontology when you adopt a few
patterns from it.
• You can easily modify a pattern without giving up on a lot of
similarity to the original pattern (which can be leveraged for data
integration).
• You can separate the patterns from specific (application-driven)
modifications.
• You can separate the patterns from specific axiomatically
defined “views”.

Patterns Example
NSF EarthCube project “OceanLink”:
• Integration of existing ocean science data repositories.
• For faceted browsing and semantic search.
• To be done in a flexible, extendable, modular way.
• With minimal effort for additional data providers to integrate
their content.
National Science Foundation award 1354778 "EAGER: Collaborative Research: EarthCube Building Blocks,
Leveraging Semantics and Linked Data for Geoscience Data Sharing and Discovery."

OceanLink and EarthCube
EarthCube:
for the Geosciences
Earth system.”

OceanLink setup
OceanLink Patterns
R2R BCO-DMO WHOI
Library
AGU NSF
UI Views
User Interface
mappings

OceanLink patterns
Some central patterns:
• Cruise
• Trajectory
• Person
• Organization
• Roles of Agents
• Repository Object
• Data Set
• Document
We’re not starting from zero of course.

Cruise trajectory

Ways forward
• Establish a flexible conceptual architecture using data and
ontological modeling.
• A principled use of patterns, including
– the development of a theory of patterns and
– the provision of a critical amount of central patterns
may provide a primary path forward.

Thanks!

References
Semantic Web 1 (1-2), 39-44, 2010.
2010.
2009.

References
2013, pp. 41-44.

References
• Prateek Jain, Pascal Hitzler, Amit P. Sheth, Kunal Verma, Peter Z. Yeh, Ontology
Alignment for Linked Open Data. In P. Patel-Schneider, Y. Pan, P. Hitzler, P. Mika,
L. Zhang, J. Pan, I. Horrocks, B. Glimm (eds.), The Semantic Web - ISWC 2010. 9th
International Semantic Web Conference, ISWC 2010, Shanghai, China, November
7-11, 2010, Revised Selected Papers, Part I. Lecture Notes in Computer Science
• Amit Krishna Joshi, Prateek Jain, Pascal Hitzler, Peter Z. Yeh, Kunal Verma, Amit
P. Sheth, Mariana Damova, Alignment-based Querying of Linked Open Data. In:
Meersman, R.; Panetto, H.; Dillon, T.; Rinderle-Ma, S.; Dadam, P.; Zhou, X.;
Pearson, S.; Ferscha, A.; Bergamaschi, S.; Cruz, I.F. (eds.), On the Move to
Meaningful Internet Systems: OTM 2012, Confederated International Conferences:
CoopIS, DOA-SVI, and ODBASE 2012, Rome, Italy, September 10-14, 2012,
Proceedings, Part II. Lecture Notes in Computer Science Vol. 7566, Springer,
Heidelberg, 2012, pp. 807-824.
• Yingjie Hu, Krzysztof Janowicz, David Carral, Simon Scheider, Werner Kuhn, Gary
Berg-Cross, Pascal Hitzler, Mike Dean, Dave Kolas, A Geo-Ontology Design
Pattern for Semantic Trajectories. In: Thora Tenbrink, John G. Stell, Antony Galton,
Zena Wood (Eds.): Spatial Information Theory - 11th International Conference,
COSIT 2013, Scarborough, UK, September 2-6, 2013. Proceedings. Lecture Notes
in Computer Science Vol. 8116, Springer, 2013, pp. 438-456.

References
6643, Springer, 2011, pp. 80-92.

References
• Sebastian Rudolph, Markus Krötzsch, Pascal Hitzler, Cheap
Boolean Role Constructors for Description Logics. In: Steffen
Hölldobler and Carsten Lutz and Heinrich Wansing (eds.),
Proceedings of 11th European Conference on Logics in Artificial
Intelligence (JELIA), volume 5293 of LNAI, pp. 362-374. Springer,
September 2008.
• Adila Alfa Krisnadhi, Frederick Maier, Pascal Hitzler, OWL and
Rules. In: A. Polleres, C. d'Amato, M. Arenas, S. Handschuh, P.
Kroner, S. Ossowski, P.F. Patel-Schneider (eds.), Reasoning
Web. Semantic Technologies for the Web of Data. 7th
International Summer School 2011, Galway, Ireland, August 23-
27, 2011, Tutorial Lectures. Lecture Notes in Computer Science
Vol. 6848, Springer, Heidelberg, 2011, pp. 382-415.

EarthCube's OceanLink - Project Overview and Presentation Updates (March 2014)

Recomendados

Recomendados

Más contenido relacionado

La actualidad más candente

La actualidad más candente (20)

Destacado

Destacado (6)

Similar a EarthCube's OceanLink - Project Overview and Presentation Updates (March 2014)

Similar a EarthCube's OceanLink - Project Overview and Presentation Updates (March 2014) (20)

Más de EarthCube

Más de EarthCube (20)

Último

Último (20)

EarthCube's OceanLink - Project Overview and Presentation Updates (March 2014)