Topic Maps: What Works and
What Doesn’t?
31 October 2007
A304 - 2:45-3:30 PM PDT
Presented by Jay Ven Eman, Ph.D., CEO
Access Innovations, Inc. / Data Harmony
505.998.0800 / www.accessinn.com / www.dataharmony.com
[email protected]
New Technologies


Meta data
W3C



OWL
SKOS
Topic Maps
Copyright  2007 Access Innovations, Inc.
Meta data


What is it in this context?
How does it work in a semantic
environment?
Copyright  2007 Access Innovations, Inc.
“Is MLB a sport, entertainment, or business?”
Copyright  2007 Access Innovations, Inc.
Semantic Web?
October 31, 2007
“Is MLB a sport, entertainment, or business?”
By
About
Smith
Professional baseball
Entertainment
Business
Summary In
Story
brief ...
Arial
1.98
Copyright  2007 Access Innovations, Inc.
1.98?


Price?
Price of what?





$, , Ÿ, £?
Wholesale? Retail? Sale?
How?
?

Newspaper?
Stadium seat?
Article?
Copyright  2007 Access Innovations, Inc.
“Meaning” starts with a knowledge
organization system (KOS)






SKOS
Uncontrolled list
Name authority file
Synonym set/ring
Controlled vocabulary
Taxonomy
Thesaurus
Topic Map
Ontology
LOTS OF OVERLAP!
Copyright  2007 Access Innovations, Inc.
Not complex - $
Highly complex - $$$$
Meta Data - the “Meaning Markers”




Data about data
Information about information
Included
Added
Copyright  2007 Access Innovations, Inc.
Data about ‘stuff’ - like what?






Author name
Date of creation
Language used in the creation
Title of the creation
Subject of the creation
Keywords...
Copyright  2007 Access Innovations, Inc.
Narrowing the focus

Keywords (AKA subject headings, index
terms, identifiers, etc.) are one type of
meta data.
Copyright  2007 Access Innovations, Inc.
For example...

A bibliographic database record usually
includes information such as author, title,
language, date of creation, and subject
area.

So does a traditional library card catalog
Copyright  2007 Access Innovations, Inc.
But did you think about…



The legend on a street map?
The yellow pages in a telephone book?
The aisle signs in a supermarket?
Copyright  2007 Access Innovations, Inc.
Meaning of meta data


Meta data is information
that ‘points’ to a
explanation or a
resolution
Meta data makes
statements about an
information resource or
object
Copyright  2007 Access Innovations, Inc.
Sidebar - meta data or metadata?

‘Metadata’ is “a word coined by Jack E.
Myers to represent current and future
lines of products implementing the
concepts of his MetaModel, and also to
designate his company, The Metadata
Company, that would develop and market
those products.”
Copyright  2007 Access Innovations, Inc.
Metadata





A term not used prior to 1969
Used first in 1973
Registered U.S. Trademark (in 1986),
owned by Jack Myers
Metadata granted incontestable status
in 1991
Designed to be a term with no particular
meaning
Copyright  2007 Access Innovations, Inc.
Included
Meta Data
Added
<DOC Date=10/31/07>
<TI> “Is MLB a sport, entertainment, or business?”</TI>
<Byline>
Smith
</Byline>
<ST>
Professional baseball </ST>
<ST>
Entertainment
<ST>
Business
<AB>
In brief ...
<Text>
There was a time ...</Text>
</ST>
</ST>
</AB>
</DOC>
Object
Meta data as indexing language
List of words
Synonyms
Taxonomy
Thesaurus
INCREASING COMPLEXITY / RICHNESS
Ambiguity control
Synonym control
Ambiguity control Ambiguity cont’l
Synonym control Synonym cont’l
Hierarchical rel’s Hierarchical rel’s
Associative rel’s
Copyright  2007 Access Innovations, Inc.
Taxonomy / thesaurus





Main Term (MT)
Top Term (TT)
Broader Terms (BT)
Narrower Terms (NT)
Related Terms (RT)




Aka subject term, heading, node,
category, descriptor, class
TAXONOMY
See also (SA)
Scope Note (SN)
History (H)
NonPreferred Term (NP)

Used for (UF), See (S)
Copyright  2007 Access Innovations, Inc.
THESAURUS
Term record
Various views
New Frontiers from
the World Wide Web
Consortium:
OWL & SKOS
Copyright  2007 Access Innovations, Inc.
The old frontier?
Term record
Various views
Taxonomy, Thesaurus, &
Ontology

Taxonomies and thesauri are not ontologies

They are entities

Ontology – science of describing kinds of
entities

“an explicit and formal specification of a
conceptualization”
Copyright  2007 Access Innovations, Inc.
Ontology

From philosophy – the science of
describing

Kinds of entities in the world

How they are related
Copyright  2007 Access Innovations, Inc.
OWL

Web Ontology Language

W3C Recommendation 10 February 2004

http://www.w3.org/TR/2004/Rec-owl-guide-20040210/

http://www.w3.org/TR/2004/Rec-owl-ref-20040210/

http://www.w3.org/TR/2004/Rec-webont-req-20040210/
Copyright  2007 Access Innovations, Inc.
Copyright  2007 Access Innovations, Inc.
Taxonomic classification
Kingdom:
Animalia
Phylum:
Chordata
Class:
Aves
Order:
Strigiformes
Families:
Strigidae
Tytonidae
Copyright  2007 Access Innovations, Inc.
Copyright  2007 Access Innovations, Inc.
Spotted Owl
Web Ontology language - OWL
 OWL

output
Provides semantic meaning to these kinds of
entities

Web resource

Accessible to automated processes
Copyright  2007 Access Innovations, Inc.
OWL

“…is intended to provide a language that
can be used to describe



the classes and
relations between them
that are inherent in Web documents and
applications.”
Copyright  2007 Access Innovations, Inc.
OWL
Formalize a domain by defining

Classes

Properties of those classes
Define individuals

Assert properties about them
Reason about these

Classes and

Individuals
Copyright  2007 Access Innovations, Inc.
OWL Ontology

May include
1.
2.
3.
Classes
Properties
Instances
SKOS
Topic

Capture semantics

Multiple, distributed, related ontology schema

Normative OWL exchange syntax

RDF/XML
Resource Description Framework/Extensible
Markup Language
Copyright  2007 Access Innovations, Inc.
Structure of
controlled vocabularies
List of words
Synonyms
Taxonomy
Thesaurus
INCREASING COMPLEXITY / RICHNESS
Ambiguity control
Synonym control
Ambiguity control Ambiguity cont’l
Synonym control Synonym cont’l
Hierarchical rel’s Hierarchical rel’s
Associative rel’s
Copyright  2007 Access Innovations, Inc.
Hierarchical View
Term
Copyright  2007 Access Innovations, Inc.
Taxonomy term record







<TermInfo>
<T>Agrotechnology</T>
<BT>Biotechnology</BT>
<NT>Animal management technologies</NT>
<NT>Controlled environment agriculture</NT>
<NT>Genetically modified crops</NT>
</TermInfo>
Source: www.DataHarmony.com
Copyright  2007 Access Innovations, Inc.
Thesaurus term record














<TermInfo>
<T>Agrotechnology</T>
<BT>Biotechnology</BT>
<NT>Animal management technologies</NT>
<NT>Controlled environment agriculture</NT>
<NT>Genetically modified crops</NT>
<RT>Agricultural science</RT>
<RT>Food technology</RT>
<UF>Plant engineering</UF>
<Scope></Scope>
<Editorial_Note></Editorial_Note>
<Facet></Facet>
<History></History>
</TermInfo>
Source: www.DataHarmony.com
Copyright  2007 Access Innovations, Inc.
OWL term record
<PreferredTerm rdf:ID="T131">
<rdfs:label xml:lang="en">Agrotechnology</rdfs:label>
<BroaderTerm rdf:resource="#T603" newsindexer:alpha="Biotechnology"/>
<NarrowerTerm rdf:resource="#T252" newsindexer:alpha="Animal
management technologies"/>
<NarrowerTerm rdf:resource="#T1221" newsindexer:alpha="Controlled
environment agriculture"/>
<NarrowerTerm rdf:resource="#T2166" newsindexer:alpha="Genetically
modified crops"/>
<Related_Term rdf:resource="#T127" newsindexer:alpha="Agricultural
science"/>
<Related_Term rdf:resource="#T2020" newsindexer:alpha="Food
technology"/>
<Non-Preferred_Term rdf:resource="#T3898" newsindexer:alpha="Plant
engineering"/>
</PreferredTerm>
Source: www.DataHarmony.com
Copyright  2007 Access Innovations, Inc.
SKOS


Simple Knowledge Organization System
SKOS Core Guide



W3C Working Draft 2 November 2005
http://www.w3.org/TR/2005/WD-swbp-skos-core-guide20051102/
SKOS Core Vocabulary Specification


W3C Working Draft 2 November 2005
http://www.w3.org/TR/2005/WD-swbp-skos-core-spec-20051102/
Copyright  2007 Access Innovations, Inc.
SKOS

May include
1.
2.
3.



Classes (RDFS)
Properties (RDF)
Instances??
OWL
Express structure and content of concept schemes
Multiple, distributed, related SKOS schemes
Normative SKOS exchange syntax
RDF/XML

Resource Description Framework/Extensible Markup Language
Copyright  2007 Access Innovations, Inc.
SKOS
Specifically for “concept schemes”








Thesauri
Classification schemes
Subject headings lists
Taxonomies
Terminologies
Glossaries
And other types of controlled vocabularies
Copyright  2007 Access Innovations, Inc.
SKOS

Models concept schemes


A set of concepts
OPTIONALLY includes statements about
semantic relationships between concepts
Directionality implied - interpretations (‘skos:Concept’ and properties)
 Not people, organizations, places, etc.

Copyright  2007 Access Innovations, Inc.
Source:
Copyright  2007 Access Innovations, Inc.
DH SKOS Output
<skos:Concept rdf:about="#T1">
<skos:prefLabel>Agriculture</skos:prefLabel>
<skos:altLabel>Agribusiness</skos:altLabel>
<skos:altLabel>Agronomy</skos:altLabel>
<skos:altLabel>Farming</skos:altLabel>
<status>Accepted</status>
</skos:Concept>
Copyright  2007 Access Innovations, Inc.
DH SKOS Output
<skos:Concept rdf:about="#T2">
<skos:prefLabel>American music</skos:prefLabel>
<skos:broader rdf:resource="#T66" local:alpha="Music styles"/>
<skos:related rdf:resource="#T27" local:alpha="Country and western music"/>
<skos:related rdf:resource="#T51" local:alpha="Jazz music"/>
<skos:related rdf:resource="#T99" local:alpha="Rhythm and blues music"/>
<skos:related rdf:resource="#T101" local:alpha="Rock music"/>
<status>Accepted</status>
</skos:Concept>
Copyright  2007 Access Innovations, Inc.
DH SKOS Output
<skos:Concept rdf:about="#T3">
<skos:prefLabel>Architecture</skos:prefLabel>
<skos:broader rdf:resource="#T113" local:alpha="Visual and performing arts"/>
<skos:scopeNote>Refers to the art and practice of designing and building
structures</skos:scopeNote>
<status>Accepted</status>
</skos:Concept>
<skos:Concept rdf:about="#T4">
<skos:prefLabel>Band music</skos:prefLabel>
<skos:broader rdf:resource="#T49" local:alpha="Instrumental music"/>
<skos:related rdf:resource="#T5" local:alpha="Bands (Music)"/>
<status>Accepted</status>
</skos:Concept>
Copyright  2007 Access Innovations, Inc.
A Brief Discussion of
Topic Maps
Statements about what?
Sports
Baseball
Amateur baseball
Little league
Professional baseball
MLB
“Is MLB a sport, entertainment,
or business?”
Copyright  2007 Access Innovations, Inc.
Topic Maps






ISO standard - ISO 13250:2002
For merging back-of-the-book indexes
Collection of structured markup
Describing KOS
Associating KOS with information
resources (objects)
Separation of KOS from objects
Topic Maps

Three main concepts
1.
2.
3.

Names of things
Occurrences of the named things
Associations between names
Three additional constructs
1.
2.
3.
Identity
Facet
Scope
OWL
Topic with occurrence
Professional baseball
descriptor-for
Topic map layer
Information resources layer
“Is MLB a sport,
entertainment, or
business?”
http://www.newindexer.com/mlb.htm/
Topics, associations,
occurrences
doc-type
MLB
Sports
member-of
article
http://www.newindexer.com/mlb.htm/
use-for
descriptor-for
Professional baseball
author-of
related-to
Baseball
member-of
Professional athletes
Amateur baseball
member-of
Little league
member-of
http://www.swaa.org
Smith
Problems with Semantic Web






Complexity
Lack of tools
Lack of skills
Limited resources
Gaming the system
The syllogism trap





KOS biases
Lack of agreement
Lack of interest
Good enough
Topic Maps vs. OWL
Lack of agreement

“Symbionese Liberation Army credited with
offing an SUV”



About - ‘revolutionaries’ or ‘freedom fighters’
About - ‘revolutions’ or ‘freedom movements’
“Symbionese Liberation Army accused of
firebombing SUV”


About - ‘terrorists’ or ‘anarchists’
About - ‘terrorism’ or ‘anarchy’
The syllogism trap



Humans are mortal
Greeks are human
Therefore, Greeks are mortal



New Mexicans speak Spanish
The author lives in New Mexico
Therefore, ...
Source: Clay Shirky, “The Semantic Web, Syllogism, and Worldview”
www.shirky.com/writings/semantic_syllogism.html/ and
Dave McComb, presentation at DAMA-I, May 2005 www.wilshireconferences.com
The syllogism humor trap



I am a nobody
Nobody is perfect
Therefore, I am perfect
Bonus:
I don't approve of political jokes.
I've seen too many of them get elected.
Topic Maps vs. OWL




TMCL
Topic maps
XTM, HyTM, LTM
ISO






OWL
RDF Schema
RDF
RDF/XML, N3
SOAP, WSDL
W3C
Full-text search and applied
indexing languages



Full-text search engines - getting better??
Thesauri applied using machine
automated indexing - easier, faster,
cheaper
Taxonomic navigation



Faceted navigation
Table of contents drilldown - taxonomy views
Query disambiguation
Copyright  2007 Access Innovations, Inc.
Full-text search and applied
indexing languages





Long history
Many richly developed thesauri with legs
Tools that work
Large body of professionals
Almost as rich
Copyright  2007 Access Innovations, Inc.
Tools that work!
Almost as rich
Hierarchical View
Term Record
ANSI/NISO Z39.19-200x
Clearer disambiguation?
Temperature
Planets
IsA
TypeOf
IsA
BrandOf
Mercury
Roman god
IsA
Metallic element
Automobile
Clearer disambiguation?

Thesaurus statement





Mercury (planet)
mercury (metal)
Mercury (automobile)
Mercury (mythical being)
mercury (temperature)
Clearer disambiguation?

OWL statement
<PreferredTerm rdf:ID="T3195">
<rdfs:label xml:lang="en">Mercury (Planets)</rdfs:label>
<BroaderTerm rdf:resource="#T3896"
newsindexer:alpha="Planets"/>
</PreferredTerm>
Thesaurus to SKOS

Thesaurus label

Main Term (MT)
Top Term (TT)
Broader Terms (BT)
Narrower Terms (NT)
Narrower Term Instance
Related Terms (RT)
 See also (SA)
NonPreferred Term (NP)
 Used for (UF), See (S)
Scope Note (SN)
History (H)









SKOS Label

<skos:Concept rdf:about=”numeric">
<skos:hasTopConcept
rdf:resource=”numeric"
local:alpha=”TopTerm"/>
<skos:broader rdf:resource=”numeric"
local:alpha=”BroaderTerm"/>
<skos:Narrower rdf:resource=”numeric"
local:alpha=”NarrowerTerm"/>
<skos:related rdf:resource=”numeric"
local:alpha=”RelatedTerm"/>
<skos:altLabel>NonpreferredTerm</sko
s:altLabel>
<rdf:Property rdf:ID=”ScopeNote">
<rdf:Property rdf:ID=”History">







Thesaurus to Ontology (OWL)

Thesaurus Label

OWL Label

Main Term (MT)
Top Term (TT)
Broader Terms (BT)
Narrower Terms (NT)
Narrower Term Instance
Related Terms (RT)

See also (SA)
NonPreferred Term (NP)

Used for (UF), See (S)
Scope Note (SN)
History (H)

<PreferredTerm rdf:ID=”numeric">
<TopTerm rdf:ID=“numeric”>
<BroaderTerm rdf:resource=”numeric"
newsindexer:alpha=”BroaderTerm"/>
<NarrowerTerm rdf:resource=”numeric"
newsindexer:alpha=”NarrowerTerm"/>
<Related_Term rdf:resource=“numeric"
newsindexer:alpha=”RelatedTerm"/>
<Non-Preferred_Term
rdf:resource=”numeric"
newsindexer:alpha=”Non-preferredTerm"/>
<owl:DatatypeProperty
rdf:ID="Scope_Note">















<owl:DatatypeProperty rdf:ID=”History">
Objectives for search &
navigation

ASIS&T -- virtual library


ASRT -- internal information control


Organization chart
Naval Postgrad -- Homeland security degree


Subject matter
Curriculum outline
SLA -- Web content

Public Web navigation
Copyright  2007 Access Innovations, Inc.
Naval Postgraduate School’s Homeland Security Taxonomy
Naval Postgraduate School’s Homeland Security Taxonomy
SLA website and thesaurus
SLA search
Myth of topic maps





And OWL, SKOS
Not a myth
They do work
Limited adoption
Narrow, tightly defined niches
Copyright  2007 Access Innovations, Inc.
Thank you. Questions?
Topic Maps: What Works and
What Doesn’t?
31 October 2007
A304 - 2:45-3:30 PM PDT
Presented by Jay Ven Eman, Ph.D., CEO
Access Innovations, Inc. / Data Harmony
505.998.0800 / www.accessinn.com / www.dataharmony.com
[email protected]
Activity in the field

Ontologies


SKOS


http://www.w3.org/2001/sw/WebOnt/impls
http://www.w3.org/TR/swbp-skos-coreguide/#secref
Topic Maps

http://www.topicmaps.org/
Copyright  2007 Access Innovations, Inc.
Resources




www.accessinn.com
www.dataharmony.com
www.iso.org
www.ontopia.com



Lars Marius Garshol, “Metadata? Thesaurui?
Taxonomies? Topic Maps!”
Steve Pepper, “The TAO of Topic Maps”
www.topicmaps.org
Copyright  2007 Access Innovations, Inc.
Resources

Cory Doctorow, “Metacrap: Putting the Torch to Seven Straw-men
of the Meta-utopia,” http://www.well.com/~doctorow/metacrap.htm

Russell Glass, “Is Anyone Going to Tag all of this Stuff?,”
http://zoominfo.blogs.com/soughtafter/2005/03/semantic_web_is.html

Clay Shirky, “The Semantic Web, Syllogism, and Worldview,”
www.shirky.com/writings/semantic_sllogism.html

Pete Norvig, “Semantic Web Ontologies: What Works and What
Doesn’t,”
www.alwayson-network.com/comments.php?id=P7480_0_3_0_C
Copyright  2007 Access Innovations, Inc.
Descargar

90 Minutes to a Great Taxonomy Part 1