Transcript Document
The Semantic Web Prof. Enrico Motta Knowledge Media Institute The Open University The World-Wide-Web • Huge information space and computing infrastructure • All the world knowledge and a huge number of services at your fingertips • Easy to publish information/services • Reasonable easy to locate resources (for humans - it is hard for machines) The WWW Revolution • Combination of ease of publishing, hugeness and removal of physical barriers implies revolutionary, disruptive, society-changing technology • Examples – Napster – Baghdad blogger beats BBC, CNN, etc… – Old lady in Yorkshire sets up very successful business from isolated cottage… – 17-year old kids become millionaires…etc, etc So what next for the web?… • Current web primarily for human consumption – Hard to locate resources for machines – Hard to locate info across web pages (for both humans and machines) – published information not easily processable by machines – Limited possibilities for largescale interoperability and the provision of ‘smart’ functionalities In a nutshell… • There is so much information out there, but current technology can do very little with it • And of course in the knowledge economy, access to the right information at the right time is a key competitive advantage <akt:Person rdf:about="akt:EnricoMotta"> <rdfs:label>Enrico Motta</rdfs:label> <akt:hasAffiliation rdf:resource="akt:TheOpenUniversity"/> <akt:hasJobTitle>kmi director</akt:hasJobTitle> <akt:worksInUnit rdf:resource="akt:KnowledgeMediaInstitute"/> <akt:hasGivenName>enrico</akt:hasGivenName> <akt:hasFamilyName>motta</akt:hasFamilyName> <akt:hasPrettyName>Enrico Motta</akt:hasPrettyName> <akt:hasPostalAddress rdf:resource="akt:KmiPostalAddress"/> <akt:hasEmailAddress>[email protected]</akt:hasEmailAddress> <akt:hasHomePage rdf:resource="http://kmi.open.ac.uk/people/motta/"/> </akt:Person> KMi Semantic Web <kmi741:Kmi-Academic-StaffMember rdf:about="kmi740:PeterScott"> <rdfs:label>PeterScott</rdfs:label> <akt741:has-research-interest rdf:resource="akt741:LearningResearch-Area"/> <akt741:has-research-interest rdf:resource="akt741:Telepresence -Research-Area"/> <akt741:has-research-interest rdf:resource="akt741:Hypermedia"/ > <akt741:has-postal-address rdf:resource="kmi740:Kmi-PostalAddress"/> <akt741:works-in-unit rdf:resource="kmi740:Cnm"/> <akt741:works-in-unit rdf:resource="akt741:KnowledgeMedia-Institute-At-The-OpenUniversity"/> <akt741:works-for rdf:resource="akt741:The-OpenUniversity"/> <akt741:has-job-title>Head of CNM / Senior Lecturer</akt741:has-jobtitle> <akt741:has-full-name>Peter Scott</akt741:has-full-name> <akt742:has-pretty-name>Peter Scott</akt742:has-pretty-name> <akt741:has-webaddress>news.kmi.open.ac.uk/pete rblog</akt741:has-web-address> </kmi741:Kmi-Academic-StaffMember> Conceptual Interoperablity Peter Scott Home Page Academic hasAffiliation hasAddress hasWebAddress... “What is Peter Scott’s Home Page?” "http://news.kmi.open.ac.uk/peterblog" Conceptual Interoperablity Academic hasAffiliation hasAddress hasWebAddress... Academic hasAffiliation hasAddress hasWebAddress... Conceptual Interoperablity Academic hasAffiliation hasAddress hasWebAddress... KMi Agent Academic hasAffiliation hasAddress hasWebAddress... Amsterdam Agent “Tell me more about Enrico Motta ”pretty cool guy…" The Semantic Web • Huge, distributed markup • Statements about web resources and entities in the world • Formal, declarative, machine processable information • Markup based on shared ontologies, to enable interoperability between agents (both human and software) So What? • Semantic Markup supports semantic search, based on meaning rather than keyword retrieval • Shared ontologies allow ‘conceptual interoperability’ • Conceptual interoperability enables the discovery, aggregation and use of information coming from multiple sites • Key is the use of formal, explicit, declarative, shared representations Trends and Possibilities…. • Semantic web is not a fad…here to stay – Some visionary papers may seem a bit over the top but reality is that • Semantic web is growing more quickly than the web • Huge amounts of distributed semantic markup will become available within 3-5 years • WWW has opened up huge new markets and removed geographical and marketing barriers • Semantic web will be about automating processes and achieving novel smart functionalities by leveraging machineprocessable contents Example: Travel Industry • Now: Expedia’s strategy is based on providing holiday packages customizations at no extra cost • Future: Automatic travel agents interoperating in real time with huge numbers of heterogenous providers to provide customtailored travel packages Example: Financial Sector • Now: Portfolio creation, management and customization slow and expensive • Future: Smart artificial brokers will outperform human brokers by being able to make use of large amounts of real-time data about financial products, environments and clients, to provide optimized, tailor-made financial portfolio for clients Example: Music Industry • Now: Digital music, ‘smart’ playlists, podcasting… • Future: All music ever created available online and massively marked-up. Huge possibilities for intelligently customized radio stations, for automated clearance of copyrights, etc, etc.. Conclusions • The semantic web is about producing massive amounts of distributed semantic annotations making explicit and formal what is implicit and informal • The availability of large scale semantic markup opens up the way for – Intelligent aggregation of information – Smart search engines able to reason about information coming from different sources, thus generating new knowledge – Dynamic agent interoperability on the web – Intelligent personalization of information