Download - Doing Clever Things with the Semantic Web
Doing Clever Things With the Semantic Web
Mathieu d’AquinKnowledge Media Institute, the Open University
The Semantic Web
Using the Web to publish, share and exploit information/knowledge
From machines to machines
Using graph-based data modeling, knowledge representation (ontologies) and reasoning
Linked Data
As set of principles and technologies for a Web of Data– Putting the “raw” data
online in a standard representation (RDF)
– Make the data Web addressable (URIs)
– Link to other Data
http://lucero-project.info/lb/what-is-linked-data/
http://linkeddata.org
Metadata<rdf:RDF><channel rdf:about=“http://watson.kmi.open.ac.uk/blog”><title>Elementaries - The Watson Blog</title><link>http://watson.kmi.open.ac.uk:8080/blog/</link><description>"Oh dear! Where the Semantic Web is going to go now?" -- imaginary user 23</description><language>en</language><copyright>Watson team</copyright><lastBuildDate>Thu, 01 Mar 2007 13:49:52 GMT</lastBuildDate><generator>Pebble (http://pebble.sourceforge.net)</generator><docs>http://backend.userland.com/rss</docs>…
<rdf:RDF> <foaf:Image rdf:about='http://static.flickr.com/132/400582453_e1e1f8602c.jpg'> <dc:title>Zen wisteria</dc:title> <dc:description></dc:description> <foaf:page rdf:resource='http://www.flickr.com/photos/xcv/400582453/'/> <foaf:topic rdf:resource='http://www.flickr.com/photos/tags/vittelgarden/'/> <foaf:topic rdf:resource='http://www.flickr.com/photos/tags/wisteria/'/> <dc:creator> <foaf:Person><foaf:name>Mathieu d'Aquin</foaf:name> …
<rdf:RDF> <owl:Ontology rdf:about=""> <owl:imports rdf:resource="http://usefulinc.com/ns/doap#"/> </owl:Ontology> <j.1:Organization rdf:ID="KMi"> <rdfs:comment rdf:datatype="http://www.w3.org/2001/XMLSchema#string" >The Knoledge Media Institute of the Open University, Milton Keynes UK</rdfs:comment> </j.1:Organization> <j.1:Document rdf:ID="KMiWebSite"> …
FOAF
DCRSS TAP
WORDNET
NCI Galen
Music
…… …
…
…
UoD
The Semantic Web
Semantic Web Applications
Doing Clever Things With the Semantic Web
Intelligent AgentSmart FeaturesClever Things…
Knowledge/Problem Solving
Methods
What I Want to Talk About
Using the Semantic Web as A Knowledge Base– KMi Watson and Finding Ontologies
• Doing More with Links– Exploring the Web of Data
• Back to the Future– AI + Linked Data = Semantic Web?
Using the Semantic Web
Dynamically retrieving, exploiting and combining relevant semantic resources from the Semantic Web
Need for a Gateway to the Semantic Web
KMi Watson
Architecture
Interface
http://watson.kmi.open.ac.ukWatson: More than a Semantic Web Search Engine, Semantic Web Journal
Watson as a Service
Providing Web accessible APIs to a collection of online ontologies and semantic data sources
Example Application: Ontology Construction
Chose an entity to search
Get entities from online ontologies
Integrate statements Into the edited ontology
Reusing Knowledge from the Semanrtic Web with the Watson Plugin, Demo at ISWC 2008
Concept Relation Discovery
ka2.rdf
Researcher AcademicStaff
Sem
anti
c W
eb
Researcher
AcademicStaff
ISWC SWRCHam SeaFood
Sem
anti
c W
eb
HamSeaFood
Meat
Meat
SeaFood
Agrovoc NALT
pizza-to-go
wine.owl
NALT
Exploring the Semantic Web as Background Knowledge for Ontology Matching, Journal of Data Semantics
PowerAqua: Question Answering
From Semantic Web Research to Linked Data Applications
Watson as a platform to research applications and techniques on top of semantic web resources
But how can the Semantic Web be exploited and used in real-world application?
Starting from what we know best…
Applying Linked Data
The Open University is the largest University in the UK, where all the courses are realized at a distance
Creating the first University Linked Data platform: data.open.ac.ukDemonstrate the value of the technology and push the research through real-world scenarios
data.open.ac.uk
Applications
Resource Discovery
Mobile and Personal Semantics
ResearchExploration
Social
Example application: Finding relevant resources
Zablith et al, LinkedLearning 2011
Data as Web resources, accessible everywhere
See also: Zablith et al., COLD 2011
http://people.kmi.open.ac.uk/mathieu/about/discobro-discovering-linked-datat-resources-while-browsing/http://discovery.ac.uk/developers/competition/
Supporting Researchers: The Reading Experience Database
http://www.open.ac.uk/Arts/reading/
40,000 accounts of somebody reading something at some time in some place
Used by researchers in literature and history to explore research hypotheses
Experience
Person
Document
EventLocation
City Countrydate: Date
subClassOf
subClassOf
locatedIn
readerInvolved
textInvolved givesBackgroundTo
title: Stringdescription: Stringpublished: Date
creator/editor
providesExcerptFor
occupation
religion
originCountry
gender
LinkedEvent Ontology
CITO Citation Ontology
Dublin Core
FOAF
DBPedia
Back to the Future
The Semantic Web is both a vision and a reality
Making the Web more than a network of documents: the biggest, most distributed knowledge base ever
What could AI do with such a knowledge base?
Linked Data MiningFinding unexpected patterns in the use of the distributed data graph
Linked Data Mining: ExampleUsing Formal Concept Analysis + Reasoning to build a hierarchy of questions a linked dataset can answer
Use statistical metrics to identify the ones that are most likely to be interesting
http://lucero-project.info/lb/2011/06/what-to-ask-linked-data/
Extracting Relevant Questions to an RDF Dataset Using Formal Concept Amalysis at KCAP 2011
ReasoningTo analyze and understand raw data in relation with online resources
Example: Online personal information management
HTTP Ontology
Web Site Information
Location Information
Online Activities Ontology
Parameters and Website
info.
Personal Information
Trust Model
Enriched with linked data
www.youtube.com
Google Services
Entertainment Websites
Video Hosting
subsediaryOf
Company
type
Video sharing
subject/category
parent
www.google-analytics.com
Web Analytics
developer
subject/category
www.google.com
owner
Internet Search Engine
Search Engine
Web Search Engine
DBpedia freebase
Basic processing/analysis
Requests by User Agents
Requests by time of day
Interests
Trust
http://uciad.info
http://uciad.info
Understanding knowledge representation and data modeling
The Semantic Web also represents a very large, collaborative base of formally represented knowledge
This can also be mined, to discover things about knowledge representation and data modeling
Ontologies on the Semantic Web
Number of entities
Domain covered
Underlying description logic
Relationships between ontologies
DOOR: Towards a Formalization of Ontology Relations at KEOD 2009
Detecting versions of ontologiesWhen published on the Web, the information about the evolution of ontologies is lost
Using URI patterns to find candidate versions of ontologieshttp://loki.cae.drexel.edu/wbs/ontology/2003/10/iso-metadata http://loki.cae.drexel.edu/wbs/ontology/2004/01/iso-metadata
Applying machine learning algorithms (SVM, Naïve Bayes and Decision tree to recognize chains of versions of ontologies
Obtained 90% Precision (SVM)
Collected thousands of ontology version sequences to be analysed
For example, distribution of similarity in version and non-version ontologies (right)
Allocca at ESWC 2011
Agreement/Disagreement between ontologies
Ontologies are knowledge artifacts, they express opinions and beliefs and contradict each othersAssessing (dis)agreement in ontologies is very useful to understand how to combine knowledge from different sources
Assessing Statements related to SeaFood
Nb1: #ontologies in which the statement appears.Nb2: #ontologies containing entities matching the subject and object of the statement.
a: global agreement, d: global disagreement, cs: consensus, ct: controversy
21 different ontologies with a SeaFood concept
Agreement
Disagreement
Formally Measuring Agreement and Disagreement in Ontologies at K-CAP 2009
http://uciad.info
SeaFood disjointWith Meat
SeaFood subClassOf Meat
Vegan subClass Vegetarian
Using consensus to assess an ontology(a new NeOn toolkit plugin
AKT PortalThe brighter the blue the higher the positive consensus (higher agreement)
The brighter the red the lower the negative consensus (higher disagreement)
Dark = controversy: no clear cut between disagreement and agreement
Example: The statements attached to the class Employee are controversial: some ontologies agree, others disagree (often due to alternative representations of roles)
Visualizing Consensus with Online Ontologies to Support Quality in Ontology Development at ONTOQUAL@EKAW 2010
So my point is…The Semantic Web is a fantastic open field for AI
It is going to become omnipresent, hidden, personal
Exploring, Exploiting and Excavating the Semantic Web for
Research in technology (creating it and studying it)
Research in other areas
Everyday tasks
Still, after 10 years of research, represent new directions for many fields of the AI community, with their own issues, challenges and applications