niso virtual conference: the semantic web coming of age: technologies and implementations
DESCRIPTION
Feb 19, 2014: NISO Virtual Conference: The Semantic Web Coming of Age: Technologies and Implementations Deck includes presentations from: Ramanathan V. Guha, Google Fellow; Founder of Schema.org; Pierre-Paul Lemyre, Director of Business Development, Lexum; Bob Du Charme, Director of Digital Media Solutions, TopQuadrantTRANSCRIPT
NISO Virtual Conference: The Semantic Web Coming of Age: Technologies and
Implementations
February 19, 2014
Speakers: Ramanathan V. Guha, Ralph Swick,
Kevin Ford, Henry Story, Pierre-Paul Lemyre, Bob DuCharme
http://www.niso.org/news/events/2014/virtual/semantic/
NISO Virtual Conference: The Semantic Web Coming of Age: Technologies and Implementations
Agenda11:00 a.m. – 11:10 a.m. – Introduction Todd Carpenter, Executive Director, NISO
11:10 a.m. - 12:00 p.m. Keynote AddressRamanathan V. Guha, Google Fellow; Founder of Schema.org
12:00 p.m. - 12:45 p.m.: The W3C Semantic Web InitiativeRalph Swick, Domain Lead of the Information and Knowledge Domain, W3C
12:45 p.m. - 1:30 p.m. Lunch Break
1:30 p.m. - 2:15 p.m. Semantic Web Applications in Libraries: The Road to BIBFRAMEKevin Ford, Network Development & MARC Standards Office, Library of Congress
2:15 p.m. - 3:00 p.m.: The Social Data Graph: The Friend of a Friend (FOAF) ProjectHenry Story, Chief Technical Officer & Co-founder at Stample
3:00 p.m. - 3:15 p.m. Afternoon Break
3:15 p.m. - 3:45 p.m. Sharing Information on the Semantic Web: The Need for a Global License RepositoryPierre-Paul Lemyre, Director of Business Development, Lexum
3:45 p.m. - 4:15 p.m. Semantic Web Applications in Publishing Bob Du Charme, Director of Digital Media Solutions, TopQuadrant
4:15 p.m. - 5:00 p.m. Conference Roundtable: Services that Build on Others Semantic Web Data: Semantic Search Beyond RDFModerated by: Todd Carpenter, Executive Director, NISO
Towards a web of Data
R.V.GuhaGoogle
Outline of talk
• How did we end up here … a personal perspective– The tortuous path through standards and products
• Schema.org– Why schema.org, principles, how does it work– Status : schemas, adoption, partners, applications
• Reports from Google, Bing, Yahoo! & Yandex
• Looking forward: Schemas in the pipeline, Research problems
One day, 16 years ago, …
• Ralph Swick, Ora Lasilla, Tim Bray, Eric Miller and myself
• Trying to make sense of a flurry of activity– XML, MCF, CDF, Sitemaps, …
• There were a number of problems – PICS, Meta data, sitemaps, …
• But one unifying idea
Context: The Web for humans
Structured Data
Web server
HTML
Goal: Web for Machines & Humans
Web server
Structured Data
Apps
What does that mean?
Chuck Norris
Ryan, Oklahama
March 10th 1940
birthdate
birthplace
Actor
type
How do we get there?
• How does the author give us the graph– Data Model– Syntax– Vocabulary– Identifiers for objects
• Why should the author give us the graph?
Going depth first
• Many heated battles– Lot of proposals, standards, companies, …
• Data model– Trees vs DLGs vs Vertical specific vs who needs one?
• Syntax– XML vs RDF vs json vs …
• Model theory anyone– We need one vs who cares vs what’s that?
Timeline of ‘standards’
• ‘96: Meta Content Framework (MCF) (Apple)• ’97: MCF using XML (Netscape) RDF, CDF• ’99 -- : RDF, RDFS • ’01 -- : DAML, OWL, OWL EL, OWL QL, OWL RL• ’03: Microformats• And many many many more … SPARQL, Turtle, N3, GRDDL,
R2RML, FOAF, SIOC, SKOS, …
• Lots of bells & whistles: model theory, inferencing, type systems, …
But something was missing …
• Fewer than 10k sites were using these standards
• Something was clearly missing and it wasn’t more language features
• We had forgotten the ‘Why’ part of the problem
• The RSS story
’07 - :Rise of the consumers
• Yahoo! Search Monkey, Google Rich Snippets, Facebook Open Graph
• Offer webmasters a simple value proposition
• Search engines to webmasters:– You give us data … we make your results nicer
• Usage begins to take off– 1000x increase in markup’ed up pages in 3 years
Yahoo Search Monkey
• Give websites control over snippet presentation• Moderate adoption
– Targeted at high end developers – Too many choices
Yandex Enhanced Snippets & Services
• Snippets for Web Search
• Data for Vertical Search Services
• Various syntax and vocabularies
• Yandex specific vocabs for unexamined use cases
Google Rich Snippets: Reviews
Google Rich Snippets: People
Google Rich Snippets: Events
Google Rich Snippets: Recipes
Google Rich Snippets: Recipe View
Google Rich Snippets
• Multi-syntax• Adhoc vocabulary for each vertical• Very clear carrot • Lots of experimentation on UI• Moderately successful• Scaling issues with vocabulary
Situation in 2010
• Too many choices/decisions for webmasters– Divergence in vocabularies
• Too much fragmentation • N versions of person, address, …
• A lot of bad/wrong markup– ~25% for microformats, ~40% with RDFA– Some spam, mostly unintended mistakes
• Absolute adoption numbers still rather low– Less than 100k sites
In the meantime …
• The Web has grown– >25 million independent active sites – Trillions of web pages– ~5 billion pages change every day– Spam, malware, …
• In other words: scaling problems along every dimension
Schema.org
• Work started in August 2010– Google, Yahoo!, Microsoft & then Yandex
• Goals:– One vocabulary understood by all the search engines– Make it very easy for the webmaster
• It is A vocabulary. Not The vocabulary.– Webmasters can use it together other vocabs– We might not understand the other vocabs. Others might
Schema.org: report card
• Over 15% of all sites/pages now have schema.org• Over 5 million sites, over 25 billion entity references
• In other words– Same order of magnitude as the web– Not just ‘promising new stuff’ anymore
1.06.1
1
1.08.1
1
1.10.1
1
1.12.1
1
1.02.1
2
1.04.1
2
1.06.1
2
1.08.1
2
1.10.1
2
1.12.1
2
1.02.1
3
1.04.1
3
1.06.1
3
1.08.1
3
1.10.1
3
% urls
% urls
Schema.org: Major sites
• News: Nytimes, guardian.com, bbc.co.uk,• Movies: imdb, rottentomatoes, movies.com• Jobs / careers: careerjet.com, monster.com, indeed.com• People: linkedin.com, • Products: ebay.com, alibaba.com, sears.com, cafepress.com,
sulit.com, fotolia.com• Videos: youtube, dailymotion, frequency.com, vinebox.com• Medical: cvs.com, drugs.com• Local: yelp.com, allmenus.com, urbanspoon.com• Events: wherevent.com, meetup.com, zillow.com, eventful• Music: last.fm, myspace.com, soundcloud.com
Schema.org: categories
• Most used categories by occurrence– Person, Offer, Product, PostalAddress, VideoObject,
ImageObject, BlogPosting, WebPage, Article, AggregateRating, LocalBusiness, Place, Organization, MusicRecording, JobPosting, Recipe, Book, Movie, Blog, Photograph, ImageGallery
• Most used categories by domains– ImageObject, WebPage, PostalAddress, BlogPosting,
Product, Person, Offer, Article, LocalBusiness, Organization, Blog, AggregateRating, Review, VideoObject, Place, Event, Rating, AudioObject, MusicRecording, Store
Schema.org: properties
• Top properties by occurrence– name, url, image, description, offers, author, price,
thumbnailUrl, datePublished, addressLocality, address, itemOffered, duration, streetAddress, isFamilyFriendly, priceCurrency, playerType, paid, regionsAllowed, postalCode, hiringOrganization, jobLocation,
• Top properties by domain– Name, description, url, image, contentURL, address,
author, telephone, price, postalCode, offers, ratingValue, priceCurrency, datePublished, addressRegion, availability, email, bestRating, creator, review, location, startDate
Schema.org principles: Simplicity
• Simple things should be simple– For webmasters, not necessarily for consumers of markup– Webmasters shouldn’t have to deal with N namespaces
• Complex things should be possible– Advanced webmasters should be able to mix and match
vocabularies
• Syntax– Microdata, usability studies– RDFa, json-ld, …
Schema.org principles: Simplicity
• Can’t expect webmasters to understand Knowledge Representation, Semantic Web Query Languages, etc.
• It has to fit in with existing workflows
• Avoid KR system driven artifacts– domainIncludes/rangeIncludes– No classes like ‘Agent’ – Categories and attributes should be concrete
Schema.org principles: Simplicity
• Copy and edit as the default mode for authors– It is not a linear spec, but a tree of examples
• Vocabularies– Authors only need to have local view– But schema.org tries to have a single global coherent
vocabulary
Schema.org principles: Incremental
• Started simple – ~ 100 categories at launch
• Applies to every area– Add complexity after adoption– now ~1200 vocab items– Go back and fill in the blanks
• Move fast, accept mistakes, iterate fast
Schema.org Principles: URIs • ~1000s of terms like Actor, birthdate
– ~10s for most sites– Common across sites
• ~10ks of terms like USA– External enumerations
• ~10m-100m terms like Chuck Norris and Ryan, Oklahama– Cannot expect agreement on these– Consumers can reconcile entity
references
Chuck Norris
Ryan, Oklahama
March 10th 1940
birthplace
Actor
type
citizenOf
USA
birthdate
Schema.org Principles: Collaborations
• Most discussions on public W3C lists
• Work closely with interest communities
• Work with others to incorporate their vocabularies– We give them attribution on schema.org– Webmasters should not have to worry about where each
piece of the vocabulary came from– Webmasters can mix and match vocabs
Schema.org Principles: Collaborations
• IPTC /NYTimes / Getty with rNews• Martin Hepp with Good Relations• US Veterans, Whitehouse, Indeed.com with Job Posting• Creative Commons with LRMI• NIH National Library of Medicine for Medical vocab.• Bibextend, Highwire Press for Bibliographic vocabulary• Benetech for Accessibility• BBC, European Broadcasting Union for TV & Radio schema• Stackexchange, SKOS group for message board• Lots and lots and lots of individuals
Schema.org Principles: Partners
• Partner with Authoring platforms– Drupal, Wordpress, Blogger, YouTube
• Drupal 8– Schema.org markup for many types
• News articles, comments, users, events, …
– More schema.org types can be created by site author– Markup in HTML5 & RDFa Lite– Coming out early 2014
Schema.org & W3C
• Web Schemas – vocabulary discussion Part of W3C Semantic Web Interest Group
• W3C Community Group(s) e.g. http://w3.org/community/schemabibex/
• Wiki, Mercurial, email lists, … Informal collaboration rather than classic standardization Ideas for extending schema.org welcomed Also room this week Thu/Fri, see blog.schema.org
W3C Data Activity• Subsumes & expands on Sem Web & eGov
• Success for Sem Web in several communities who publish self-describing, re-usable structured data
• schema.org makes it easier for developers to use common vocabulary
• W3C pleased with the success of schema.org and continues to encourage data formats that support mixing of vocabs
Schema.Org Usage @ MS
Today• Used primarily to support Bing’s Rich Captions• Also recommended for Windows 8 App Contracts
Tomorrow• Extended usage in Rich Captions and Bing Search• Development support via new Bing Platform Dev
Center• Innovative experiences in other Microsoft products
Bing Rich Captions
• Ensure your Schema.Org annotations represent the primary content on the page -> ensures correct caption data is shown
• Ensure your Schema.Org annotations are appropriate and relevant to the content of the page -> improves your chances of being shown
• Rich Captions do not support RDFa 1.1 or JSON-LD -> stick with RDFa 1.0 or Microdata for now
Yahoo! Search
• Rich results– Blanco et al. Enhanced Results for Web Search, SIGIR 2011– Mika & Potter. Metadata Statistics for a Large Web Corpus, LDOW 2012
• Related entities– Blanco et al.
Entity recommendations in Web Search– Wed 15:15 Search track
Yahoo! Media
• Schema.org across the Yahoo! Network– Article, Photo and Video markup– Q&A and Sports under development
• Personalization– Based on past reading behavior, Facebook profile, implicit feedback– Concepts from a news taxonomy and entity-graph– Nicolas Torzec. The Y! Knowledge Base: Making Knowledge Reusable at
Yahoo! SemTech 2013
Applications
• Applications drive adoption
• First generation of applications– Rich presentation of search results
• Many new applications are coming up– On search page and beyond
Newer Applications: Knowledge Graph
Newer Applications: Knowledge Graph
Non web search Applications
• Searching for Veteran friendly jobs
Non search applications: Google Now
Pinterest: Schema.org for Rich Pins
Non search Applications
• Open Table website confirmation email Android Reminder
Future application
• Clinical trials• 4000+ clinical trials at any time in the US alone• Almost all the data ‘thrown away’• All that gets published is a textual ‘abstract’
• Over half the trials are redundant• Earlier trials have the data• Assumptions, etc. cannot be re-examined• Longitudinal studies extremely hard, but super
Vocabularies in the pipeline
• Actions (Potential Actions), Events• Accessibility• Commerce: Orders, Reservations, …• Communication: Fleshing out TV, Radio, Email, Q&A, …
• Media: Scholarly works, Comics, Serials• Sports• and many many more …
Big initiatives underway
• Representing time– Lot of triples with associated time interval– Hard to force into the triple model– Looking at named graphs
• Tabular / CSV data– Census data, Scientific data, etc.– Need mechanisms for external specification of the meaning of these tables
Research Problems
• Propositional Attitudes– Fictional characters– Possible actions
• Mapping Entities across sites– Billions of entities– But hundreds of billions of entity references– Very large scale cross site entity mapping
Schema.org: concluding
• Provide webmasters – A good reason for adding structured data markup– Allow more advanced authors to add much richer markup
from many different vocabularies– An easy way to do it
• Many different applications emerging
• Many interesting research problems left
• Light at the end of the tunnel
Questions?
MCF
• Introduced the Directed Labeled Graph (DLG) model
• Used to describe site structure• Provided simple visualization plugin• ~3m downloads, few thousand sites (not too bad for ‘96)• Documentation was about a page• Lead to MCF using XML (4 pages including diagrams)
97-99: RDF / RDFS
• MCF (4 pages) got renamed to RDF• Data model becomes a religion• Lot of new things … reification, sequences, chains, model theory, …
• Final version done only by 2002• RDF Primer is 100 pages long!• But … used by very few websites
‘99: RSS 0.91
• Netscape needs customizable home page ala my.yahoo
• No $s to do deals• Create format and open it up to everyone
– 2 year target was 5k feeds– 14 years later 100m+ feeds
2002: OWL
• Tim BL, et. al. ‘Semantic Web’ paper• Answer to lack of adoption: we need more features
• Lots of Darpa and EU funding• Very powerful solutions … looking for problems• ~10 years later … < 100 sites use it
‘04-07: Microformats
• Reaction to the complexity of RDF & OWL• Retrofitted vCard, etc. into HTML hCard• Adoption by Google & Yahoo! made it wildly successful
• Evolutionary dead-end: no new vocabulary in over 5 yrs.
Lessons
• The RSS story• Make sure you have your carrot!
– Carrots work much better than sticks!• Find the right initial level of generality• Start simple and iterate fast• Optimize for flexibility
Sharing Information on the Semantic Web:The Need for a Global License Repository
Distributed Production of Information• Originates in research (not
commercial venture)
• Access is its driving force (not property)
• Has defined the Internet as we know it– Open standards & Open Source Software
(OSS)
– Web 2.0
– And now the semantic web
Distributed Production of Information
Why does it work?
OSS• Motives to produce OSS
– Ethical / policy reasons
– Non-monetary incentives
– Increase the speed of market adoption
– Co-create and appropriate value• By building on previous works• By integrating external contributions
OSS• Each contributor is adding to the pool
of knowledge available to all
• This knowledge is more valuable than what any contributor can achieve individually
• Not new phenomenon (science, music, education, ...)
OSS - Legal Framework• Collaborative software development
existed before– Under informal agreements
– Under bilateral contractual schemes
• OSS licenses created a favourable legal environment– By favouring reciprocity (BSD)
– By securing openness (GPL)
OSS• What the success of OSS makes us
see clearly
– In a networked world, centralized corporate production can sometime be surpassed by distributed production
– Mass adoption requires a proper legal framework
OSS - Legal Framework
Is this conclusion applicable only to software?
Web 2.0• Extensive use of Web services
facilitating mass collaboration– Rich Internet applications
– Web forums
– Blogs
– Wikis
– Folksonomies (Social tagging)
Web 2.0• Web-as-participation-platform
– Architecture of participation
– Users become producers
– Collective intelligence
Web 2.0 - Legal Framework• Adaptation of OSS licenses
– GNU Free Documentation License
– OpenContent License
– Creative Commons (CC)
Web 2.0 - Legal Framework• Extended the favourable
environment to all types of freely accessible information– By specifying the applicable reuse
conditions (no permission required)
– By clarifying the spectrum of potential rights
– By standardizing the licensing process and making automated retrieval possible (CC)
Web 2.0 - Legal Framework
The question of online right management should be solved by
now?
Semantic web• Collaborative initiatives developed
independently from each other– Vertical information flow (Information
silos)
• Accessibility & reusability of information call for reciprocity between them and other sources of information– Horizontal information flow (seamless
interoperability)
Semantic Web• Using technology for data sharing
and reuse– Data modeling (XML, RDF)
– Syndication technologies (RSS)
– Ontologies (OWL)
– Heuristic (text-recognition)
Semantic Web• Web understandable by computers
– Data get a meaning
– Dynamic discovery, composition and execution
– Layers of services
Semantic Web• Current state of implementation
– Information sources are pre-qualified (limited dynamic discovery)
• Public domain sources• Use of APIs under bilateral agreements
– Reproduction of information is often limited (links are the norm, mashup still the exception)
– Management of personal information is a nightmare
Semantic web -Legal Framework
Here again successful mass adoption is not just about technology
The Legal Issue:Fragmentation of Rights• Rights are fragmented
– By reuse conditions
– By jurisdictions
– By formats / domains
• Scalable to the smallest element (website > webpage > data)
The Legal Issue:Fragmentation of Rights• Not a new problem
– Requests to limit the proliferation of OSS licenses (FSF)
– Initiatives to standardize licensing (CC)
• Not fundamental as long as humans are in charge of the reuse of information
The Legal Issue:Fragmentation of Rights• Seamless interoperability through
semantic technologies require computers to automatically– Retrieve applicable licenses
– Resolve their respective terms
– Select information with adequate conditions for the anticipated reuse
The Legal Issue:Fragmentation of Rights• Standardization under CC is helping
– Embedded license information ease their retrieval
– Computer readable version ease their resolution
– Standardized terminology and low number ease the selection
The Legal Issue:Fragmentation of Rights• But it is a partial solution
– Most content is not licensed under CC (and often cannot)
– Copyright holders have the right to attach alternative conditions to their content
– CC is not generally accepting alternative licenses
• Example of difficulties– Website Terms of Use vs Google “Usage
rights” feature
The Legal Issue:Fragmentation of Rights
A higher level resolution mechanism is required
The Solution:A Global License Repository?• A database of varying copyright
licenses and their conditions
• A standardized approach to licenses resolution and selection (expand the CC model?)
• A Web service that can be queried by users and computers
The Solution:A Global License Repository?• Issues that need to be addressed
– Unlimited number of reuse conditions
– Format and domain specific restrictions
– Internationalization
– Versioning of licenses
– Compatibility between licenses (relicensing)
The Solution:A Global License Repository?• Possible solutions
– Organizing conditions and restrictions into groups or categories?
– Managing licenses at the lowest possible level and associating related ones?
– Limiting the designation of compatibility to the most common licenses?
The Solution:A Global License Repository?• A successful implementation requires
– Promotion and large-scale adoption of a standard tagging model for information
– Involvement of copyright holders in feeding and updating the database
– Transparency and quality control procedures generating trust in the system
The Solution:A Global License Repository?• A successful implementation requires
– Scalability insuring efficient interactions at every level of development
– Provision of outputs under standardized formats
– Provision of simple APIs facilitating interactions with the repository
The Solution:A Global License Repository?• Architecture
– Use of open standards and OSS to display transparency
– Use of collaborative technologies to distribute updates
– Use of aggregative technologies to promote exploitation and reuse
Conclusion• Facilitating the sharing of information
is the core function of the Internet– Semantic web is a no-brainer in principle
• But mass adoption requires a legal framework– Providing a lawful mean to co-create and
appropriate value
– Securing the existing rights of copyright holders
Conclusion
TechnologyLegal Framework
OSS Copyleft
Web 2.0Creative
Commons
Semantic Web ???
© Copyright 2014 TopQuadrant Inc. 100
Semantic Web Applications in Publishing
Bob DuCharmeFebruary 19, 2014
© Copyright 2014 TopQuadrant Inc. 101
Outline
http://snee.com/rdf/niso2014/
RDF
Taxonomies, semantics, and content
Metadata management
Content creation
© Copyright 2014 TopQuadrant Inc. Slide 102
RDF
● Resource Description Format● W3C Standard● Syntaxes exist, but ultimately a data model● Very easy to aggregate● Tools exist to treat relational data and
spreadsheets as RDF
© Copyright 2014 TopQuadrant Inc. Slide 103
An RDF “statement”: the triple
● (Subject, predicate, object)● “This resource, for this property, has this
value.”● “John Smith has a hire date of 2012-10-11.”● “Chair 523 located in in room 47.”● “index.html has the title ‘My Home Page'.”
© Copyright 2014 TopQuadrant Inc. Slide 104
Using URIs
A triple?
Subject: index.html
Predicate: title
Object: “My Home Page”
© Copyright 2014 TopQuadrant Inc. Slide 105
Using URIs*
A triple?
Subject: index.html
Predicate: title
Object: “My Home Page”A triple!
Subject: <http://www.myco.com/members/index.html>Predicate: <http://purl.org/dc/elements/1.1/title>Object: “My Home Page”
*Uniform Resource Identifier, not Uniform Resource Locator—an identifier, not an address.
© Copyright 2014 TopQuadrant Inc. Slide 106
URIs as triple objects
<urn:isbn:9780062515872><http://purl.org/dc/elements/1.1/creator>“Tim Berners-Lee” .or…
<urn:isbn:9780062515872><http://purl.org/dc/elements/1.1/creator><http://topbraid.org/ids/TimBernersLee> .or…
<urn:isbn:9780062515872><http://purl.org/dc/elements/1.1/creator><http://www.w3.org/People/Berners-Lee/card#i> .
© Copyright 2014 TopQuadrant Inc. Slide 107
URIs as triple objects<urn:isbn:9780062515872><http://purl.org/dc/elements/1.1/creator><http://www.w3.org/People/Berners-Lee/card#i> .
<http://www.w3.org/People/Berners-Lee/card#i><http://www.w3.org/2000/01/rdf-schema#label>“Tim Berners-Lee” .
107
© Copyright 2014 TopQuadrant Inc. Slide 108
URIs as triple objects<urn:isbn:9780062515872><http://purl.org/dc/elements/1.1/creator><http://www.w3.org/People/Berners-Lee/card#i> .
<http://www.w3.org/People/Berners-Lee/card#i><http://www.w3.org/2000/01/rdf-schema#label>“Tim Berners-Lee” .
108
< http://www.w3.org/People/Berners-Lee/card#i><http://xmlns.com/foaf/0.1/mbox> “[email protected]” .
< http://www.w3.org/People/Berners-Lee/card#i><http://dbpedia.org/property/almaMater><http://dbpedia.org/resource/The_Queen's_College,_Oxford> .
<http://dbpedia.org/resource/The_Queen's_College,_Oxford><http://dbpedia.org/property/established> 1341 .
© Copyright 2014 TopQuadrant Inc. Slide 109
http://www.w3.org/People/Berners-Lee/card#i
urn:isbn:9780062515872
http://dbpedia.org/resource/The_Queen's_College,_Oxford
“Tim Berners-Lee”
“[email protected]”1341
http://purl.org/dc/elements/1.1/creator
http://xmlns.com/foaf/0.1/mbox
http://www.w3.org/2000/01/rdf-schema#label
http://dbpedia.org/property/almaMater
http://dbpedia.org/property/established
A “graph”
© Copyright 2014 TopQuadrant Inc. Slide 110
http://www.w3.org/People/Berners-Lee/card#i
urn:isbn:9780062515872
http://dbpedia.org/resource/The_Queen's_College,_Oxford
“Tim Berners-Lee”
“[email protected]”1341
http://purl.org/dc/elements/1.1/creator
http://xmlns.com/foaf/0.1/mbox
http://www.w3.org/2000/01/rdf-schema#label
http://dbpedia.org/property/almaMater
http://dbpedia.org/property/established
SPARQL: look for patterns within the graph
© Copyright 2014 TopQuadrant Inc. Slide 111
Defining structure● Optional
● Nice option in “schema vs. schemaless” debate
● RDFS (RDF Schema Language)● Enumerate classes, properties, and relationships between them
# Prefixes stand in for base URIs, like with XMLfoaf:Person rdf:type owl:Class . viaf:19831453 rdf:type foaf:Person .
● OWL (Web Ontology Language)● Further describe classes and properties, enable fancier inferencing● For example: Jack has a spouse value of Jill; "spouse" is an
owl:SymmetricProperty; software can then infer that Jill has a spouse value of Jack. (Semantics!)
– Both are W3C standards, both expressed with triples (and therefore easy to aggregate)
© Copyright 2014 TopQuadrant Inc. 112
Simple Knowledge Organization System
Controlled vocabulary
Taxonomy
Thesaurus
Ontology
© Copyright 2014 TopQuadrant Inc. 113
Taxonomies
Mammal
Dog
Bulldog Collie
Horse Cat
Above: subset-of relationship. Alternatives: part-of, instance-of.
metadata!
© Copyright 2014 TopQuadrant Inc. 114
Thesaurus
Mammal
Building
Dog
Bulldog Collie
Horse Cat
House
Residential Commercial
Doghouse
(use for: mutt, cur)
Related term
© Copyright 2014 TopQuadrant Inc. 115
Simple Knowledge Organization System
Controlled vocabulary
Taxonomy
Thesaurus
Ontology
© Copyright 2014 TopQuadrant Inc. 116
Simple Knowledge Organization System
Controlled vocabulary
Taxonomy
Thesaurus
OntologySKOS: the W3C’s OWL ontology forcreating thesauri, taxonomies, and controlled vocabularies.
© Copyright 2014 TopQuadrant Inc. 117
http://myCompany.com/animals/c43209101
preferred label (English): "dog"
preferred label (Spanish): "perro"
preferred label (French): "chien"
alternative label (English): "mutt"
alternative label (Spanish): "chucho"
history note: "Edited by Jack on 5/4/11"
related term: http://myCompany.com/shelters/c3048293
SKOS: standardized properties
© Copyright 2014 TopQuadrant Inc. 118
http://myCompany.com/animals/c43209101
preferred label (English): "dog"
preferred label (Spanish): "perro"
preferred label (French): "chien"
alternative label (English): "mutt"
alternative label (Spanish): "chucho"
history note: "Edited by Jack on 5/4/11"
related term: http://myCompany.com/shelters/c3048293
product: http://myCompany.com/vaccinations/c2197503
foo code: “5L-MN1-003”
SKOS: custom properties
© Copyright 2014 TopQuadrant Inc. 119
Who is using SKOS?
AGROVOC
New York Times: People, Organizations, Locations, Subject Descriptors
Library of Congress subject headers
AGFA drug admin. forms
NASA: many categories
© Copyright 2014 TopQuadrant Inc. 120
TopQuadrant’s TopBraid EVN
© Copyright 2014 TopQuadrant Inc. 121
Metadata management
Content metadata is more than just keywords
3 Vs of Big Data: Volume, Velocity and Variety
Gartner October 2013 poll on which is most difficult: Variety: 16%
Volume: 35%
Variety: 49%
© Copyright 2014 TopQuadrant Inc. 122
Metadata in images from 3 suppliersSupplier 1 Supplier 2 Supplier 3
file name file name file name
file size file size file size
last modified last modified last modified
shutter speed shutter speed
GPS latitude GPS latitude
GPS longitude GPS longitude
creator name
X resolution X resolution Horz. resolution
Y resolution Y resolution Vertical resolution
copyright notice re-use rights
keywords subject
aperture setting F stop
(etc.)
© Copyright 2014 TopQuadrant Inc. 123
Accommodating overlapping metadata
Store it all; each is just a triple
Identify relationships for easier searching “aperture setting” = “F stop”
“copyright notice” a subproperty of “re-use rights”
“Y resolution” a subproperty of “vertical resolution”
Image data a simple case. Consider audio, video, management of digital rights…
© Copyright 2011 TopQuadrant Inc. Slide 124
Metadata and…
Prosopography: the study of careers, especially of individuals linked by family, economic, social, or political relationships.
© Copyright 2014 TopQuadrant Inc. 125
Content creation
From an article on Massive Open Online Courses in the February 8, 2014 Economist:
"...says Tylor Cowen, a co-founder of Marginal Revolution University, it is possible that textbook publishers are better equipped than universities to develop MOOCs profitably."
© Copyright 2014 TopQuadrant Inc. Slide 126
Available Data: The Linked Open Data Cloud
© Copyright 2014 TopQuadrant Inc. Slide 127
Wikipedia page on Andrew Johnson
© Copyright 2014 TopQuadrant Inc. Slide 128
DBpedia page for Andrew Johnson
(further down the page…)
NISO Virtual ConferenceThe Semantic Web Coming of Age: Technologies and Implementations
NISO Virtual Conference • February 19, 2014
Questions?All questions will be posted with presenter answers on the NISO website following the webinar:
http://www.niso.org/news/events/2014/virtual/semantic/
Thank you for joining us today. Please take a moment to fill out the brief online survey.
We look forward to hearing from you!
THANK YOU