A Non-Technical, Example-Driven Introduction to Linked Data

LINKED DATA ONTOLOGIES APPLICATIONS (AT STKO)
HOW LINKED DATA AND SEMANTIC WEB
TECHNOLOGIES FOSTER THE
PUBLICATION, RETRIEVAL, REUSE, AND
INTEGRATION OF DATA
A NON-TECHNICAL, EXAMPLE-DRIVEN INTRODUCTION
Krzysztof Janowicz, Grant McKenzie, and Yingjie Hu
STKO Lab
University of California, Santa Barbara, USA
UCSB Library, April 2014
AN EXAMPLE-DRIVEN INTRODUCTION TO LINKED DATA JANOWICZ, MCKENZIE, AND HU

WHAT IS LINKED DATA?
LINKING DATA AS NEXT-GENERATION INFRASTRUCTURE
Data Silos
Web services
Databases
Web pages
hinder ad-hoc combination
enforce data models
limit re-usability

FROM DOCUMENTS TO DATA
FROM LINKED DOCUMENTS TO LINKED DATA
Use Uniform Resource Identiﬁers (URI) to identify entities, link them to other
entities, encode information about these entities using the
machine-understandable RDF, and make them available on the Web.

FROM DOCUMENTS TO DATA
BERNERS-LEE’S LINKED DATA PRINCIPLES AND STARS
Four Rules for Linked Data
Use URIs as names for things
Use HTTP URIs so that people can look up those names.
When someone looks up a URI, provide useful information, using the standards
(RDF*, SPARQL)
Include links to other URIs. so that they can discover more things.
Is your Linked Open Data 5 Star?
Available on the web (whatever format) but with an open licence, to be Open Data
Available as machine-readable structured data (e.g. excel instead of image
scan of a table)
as (2) plus non-proprietary format (e.g. CSV instead of excel)
All the above plus, Use open standards from W3C (RDF and SPARQL) to
identify things, so that people can point at your stuff
All the above, plus: Link your data to other people’s data to provide
context
See http://www.w3.org/DesignIssues/LinkedData.html

EXPLORING LINKED DATA
EXPLORING LINKED DATA RELATED TO SANTA BARBARA
Follow-your-nose: Explore information related to Santa Barbara using
Linked Data (DBpedia).

FINDING RELATIONS BETWEEN PLACES
Finding Relations between places (via people, events, objects,...).

REASONATOR:
BROWSE WIKIDATA
’Wikidata is a
free knowledge
base that can be
read and edited
by humans and
machines alike.’

SEARCHING THE WEB OF DOCUMENTS
This is still how most (Web) search works today. 20 million results, no hit.

SEARCHING THE WEB OF (LINKED) DATA
Populated places have a population, are located, occupy a certain area,...

GOOGLE’S KNOWLEDGE GRAPH

GOOGLE’S KNOWLEDGE GRAPH
Google’s Web search is changing towards query answering.

QUERYING AND INTEGRATION OF LINKED DATA
THE GLOBAL GRAPH OF LINKED DATA
The examples before involved only one data set, but there is much more...

THE GLOBAL GRAPH OF LINKED DATA

INTEGRATION AND QUERY FEDERATION
Integration by searching equivalent classes or/and same features
in data sets. This requires ontology matching and alignment.

ONTOLOGIES IN COMPUTER/INFORMATION SCIENCE
ONTOLOGIES IN COMPUTER/INFORMATION SCIENCE
An ontology is an explicit speciﬁcation of a conceptualization used
to achieve a shared understanding of a particular domain of interest. (adopted from Gruber (1993))

ENTITIES, CONCEPTS AND CATEGORIES
An entity is an individual (real world) object.

A concept/class is a (mental) template used for grouping entities.

A category is the set of entities grouped using a particular concept.

SUBSUMPTION AND SIMILARITY
SUBSUMPTION
Concepts can be organized within hierarchical structures.
[this subClassOf relation is transitive]

SUBSUMPTION AND SIMILARITY
SIMILARITY
Some concepts (and entities) are more similar than others.

ONTOLOGY LANGUAGES
UNDER THE HOOD OF A MAP LEGEND ONTOLOGY
NC = {LegendItem, Symbol, Label, FeatureType} (1)
NR = {consistsOf, isLabelFor, isLabelOf, SymbolizedBy} (2)
¬∃N. (3)
LegendItem ∃consistsOf.Symbol ∃consistsOf.LegendItem (4)
Label ∃SymbolizedBy.Symbol ∀SymbolizedBy.Symbol (5)
≤ 1isLabelFor. (6)
≤ 1isLabelOf. (7)
≤ 1SymbolizedBy. (8)
Label ∃isLabelFor.FeatureType (9)
Label Symbol ⊥ (also for Symbol, Label, FeatureType, LegendItem) (10)
SymbolizedBy−
◦ isLabelFor depictedBy−
(11)
¬∃consistsOf−
Legend (12)
. . . (13)

SEMANTIC INTEROPERABILITY
WHY NOT JUST STANDARDIZE MEANING? (CITY OR TOWN?)

WHY NOT JUST STANDARDIZE MEANING? (CITY OR TOWN?)
California:
City ≡ Town
Utah:
Town ≡ < (population, 1000)
Pennsylvania:
Town ≡ {Bloomsburg}

SEMANTIC INTEROPERABILITY – MEANINGFUL LINKS
Unfortunately, our data sources use
exactly the same terminology (e.g.,
connection) to talk about totally different
and contradicting facts (e.g., separation)
While we can still syntactically integrate
and reuse information, the results may be
misleading or even meaningless
We need heterogeneity preserving semantic interoperability methods

LINKED DATA AND VOCABULARIES
ONTOLOGIES TO MAKE YOUR DATA MORE USABLE
Five Stars of Linked Data Vocabulary Use
Linked Data without any vocabulary.
There is dereferencable human-readable information about the used
vocabulary.
The information is available as machine-readable explicit
axiomatization of the vocabulary.
The vocabulary is linked to other vocabularies
Metadata about the vocabulary is available (in a dereferencable
and machine-readable form).
The vocabulary is linked to by other vocabularies.
See http://semantic-web-journal.net/content/
five-stars-linked-data-vocabulary-use

LINKED DATA AND MAPS
A TINY ONTOLOGY FOR ESRI’S ARCGIS ONLINE
This fragment of the ontology developed for ArcGIS Online deﬁnes
the relations (e.g., isOwnerOf) between items (e.g., map services),
users, and user groups.

WHICH BASEMAP IS MOST POPULAR (BASED ON VIEWS)?
SELECT DISTINCT ?baseMap ?numViews
WHERE { ?baseMap arcgis:isBaseMapOf ?item .
?baseMap arcgis:numViews ?numViews }
ORDER BY DESC(?numViews) LIMIT 10
Listing 1: SPARQL query for most popular maps based on views.

WHICH BASEMAP IS MOST POPULAR (BASED ON USAGE)?
SELECT ?baseMap (count(distinct ?item) as ?usedTimes)
WHERE { ?baseMap arcgis:isBaseMapOf ?item }
GROUP BY ?baseMap
ORDER BY DESC(?usedTimes) LIMIT 10
Listing 2: SPARQL query for most popular maps based on usage.

LINKED DATA PORTAL FOR ARCGIS ONLINE (DEMO)
Semantics-enabled and Linked Data-driven similarity search interface for ArcGIS Online.
http://stko-exp.geog.ucsb.edu/linkedarcgis/

LINKED DATA AND GAZETTEERS
USCB’S ALEXANDRIA DIGITAL LIBRARY GAZETTEER (2006)
Interface of the original Alexandria Digital Library Gazetteer.

ONTOLOGY-POWERED, LINKED DATA-DRIVEN ADL GAZETTEER (DEMO)
5 Million places merged from multiple authoritative data sources. Contains
multiple alternative (e.g., historic) names, provenance information, 1200
geographic feature classes, polygon data, GeoSPARQL endpoint, etc.
[Still a lot of work to be done].

WHY NOT JUST USE GEONAMES?
A SPARQL query for people living near the Gulf of Guinea will return about 7
billion! See http://stko.geog.ucsb.edu/location_linked_data for more examples.

LINKED DATA AND SCIENTOMETRICS
DEKDIV: EXPLORING BIBLIOGRAPHIC LINKED DATA (DEMO)
System: http://stko-exp.geog.ucsb.edu/lak/; paper: http://bit.ly/1dW2NER

LINKED DATA AND SCIENTOMETRICS
SEMANTIC WEB JOURNAL: LINKED SCIENTOMETRICS
System: http://semantic-web-journal.com/SWJPortal/;
paper: http://bit.ly/1ilwbRU

A Non-Technical, Example-Driven Introduction to Linked Data

More Related Content

What's hot (16)

Similar to A Non-Technical, Example-Driven Introduction to Linked Data (20)

More from kjanowicz (20)

Recently uploaded (20)

A Non-Technical, Example-Driven Introduction to Linked Data