Semantic Web, NON-technically speaking
- 1. Semantic Web in Context Segment at TLT Center “Web2.0 Day” Eric Hoffer Second Integral at Seton Hall July 16, 2008
- 2. Agenda About me Your focus on Web2.0 Versions of the Web, relatively speaking The Semantic Web or Web3.0 What does it have to do with you
- 3. So, about me marketing, strategy, people, processes, puzzles, tools, usability, products, productivity… research, analytics, financial markets, work, the world, people… living and doing their jobs discovery – information – connecting - its about helping real people living and doing their work unlock “SME-tise” and enable the many
- 4. me (one more thing)… I’m interested in lots of different things... Visual Usability Running Value Networks Finance me
- 5. me (one more thing)… I’m interested in lots of different things... particularly where they interact…
- 6. me (one more thing)… I’m interested in lots of different things... particularly where they interact… and in their outer reaches…
- 7. me (one more thing)… I’m interested in lots of different things... particularly where they interact… and in their outer reaches… in the interstices - where they overlap. Biology and Anthropology System Dynamics and Economics Usability and Value Networks As you might experience with:
- 8. Interstices Why am I telling you this? To me, this is what semantic technologies and the Semantic Web are about. Enabling the crossing of lines, bridging gaps, unlocking expertise, getting things done… for people.
- 9. Technical aspects… There’s all sorts of technical details, but let’s just get some context
- 10. Web Progression Web (original) – Pages and Links Web2.0 Web3.0 (apologies in advance for using a version-type naming scheme – which doesn’t really exist).
- 11. Web (original, Web1.0) - about About companies showing their stuff And about people finding the stuff
- 12. Web1.0 (links) (Image borrowed from Josh Bancroft’s tinyscreenfuls.com) This is the view as far as the machine is concerned
- 13. Web Progression Web1.0 – Pages and Links Web2.0 Web3.0
- 14. Web Progression Web1.0 – Pages and Links (static, proprietary, no user contribution, and technical) Web2.0 Web3.0
- 15. Web Progression Web1.0 – Pages and Links (static, proprietary, no user contribution, and technical) Web2.0 – Participatory Web3.0
- 16. Web2.0 - about About people taking part in the web, sharing ideas, opinions, creations… and others taking advantage and enjoying using the web as a platform to deliver or piece together capabilities
- 17. Web2.0 – elements rating/reviewing sharing collaborating socializing blogging pull tagging mashing
- 18. Web1.0 2.0 (how?) Under the hood, XML brought “structure” to web pages, by separating people-bound information from computer-bound formatting instructions, companies and services could now provide wrappers - into which, people just fill in their own content. Image source: http://www.flickr.com/photos/fleur-design
- 19. Web2.0 – examples rating/reviewing – amazon, ebay… sharing – flickr, youtube… collaborating – wikis, google docs… socializing – myspace, facebook, linkedin… blogging – typepad, blogger, wordpress… pull – RSS, ATOM tagging – delicious, zigtag… mashing – google maps…
- 20. Web2.0 (reviews, rating, tagging)
- 21. Web Progression Web1.0 – Pages and Links (static, proprietary, no user contribution, and technical) Web2.0 – Participatory Web3.0
- 22. Web Progression Web1.0 – Pages and Links (static, proprietary, no user contribution, and technical) Web2.0 – Participatory (dynamic, contributive, connecting, interactive, and easy) Web3.0
- 23. Great stuff, Web2.0. So what’s missing? XML separated complexity of presentation layer from content, BUT didn’t structure the content (text and data). So, computers can’t tell one word from another, nor differentiate between different meanings of a word. (e.g. “orange”…). Therefore, computers still couldn’t help us to get better answers, (just nicer looking ones that might be close to what we ask about).
- 24. Enter, the S emantic W eb, or semantic web, Web3.0…
- 25. "Semantic Web is an extension of the current Web in which information is given well-defined meaning, better enabling computers and people to work in cooperation." (Tim Berners-Lee, James Hendler, Ora Lassila, The Semantic Web , Scientific American, May 2001)
- 26. Web3.0/Semantic Web – about If Web2.0 is for “people” (being part of it and connecting between each other and things), Web3.0 is for “machines” (and connections between data - for the benefit of people).
- 27. Web3.0/Semantic Web – about Adding granularity to what is on the page, to enable identification and classification of items (people, places, things, events) on page – so the computer can help find the right pages, to grab the right data, mix things together in like terms...
- 28. Web Progression Web1.0 – Pages and Links (static, non-interactive, non-dynamic, and technical) Web2.0 – Participatory (dynamic, contributive, connecting, interactive, and easy) Web3.0 –Semantic. Interlinking. Relative.
- 29. Web Progression Web1.0 – Pages and Links (static, non-interactive, non-dynamic, and technical) Web2.0 – Participatory (dynamic, contributive, connecting, interactive, and easy) Web3.0 –Semantic. Interlinking. Relative. (compatibility for data interoperability – to intelligently get to answers)
- 30. BUT HOW, and to what end? Remember all those Web 2.0 things? Well 3.0 is the same stuff, but enhanced with contextual information. Relatively easy in closed environment – mapping to likely choices – but out in the wild is another story. Consider simple tagging – In context can reveal more. And mashing - Think APIs – but on a really broad scale.
- 31. Mashing
- 32. Sharing + Tagging + Mapping See live example
- 33. Isn’t all that just Web2.0? What’s the Semantic Web part? contextualization and meaning draw logical conclusions (inferencing) digest natural language (NLP) comply with W3C Standards (more later) …
- 34. Web3.0 From the outside, it is – or rather, should be, just enhanced experience for the user Fewer and better search results Actual answers to questions Information from different resources being dynamically pulled together and provided to you in common terms.
- 36. Web2.0 3.0 – how? Under the hood of the transition this time… Source: http://www.flickr.com/photos/suzanneandsimon Adding granularity to the people-bound portion of what is on the page – so the computer can use it to help us do those things.
- 37. Web2.0 3.0 – how? Somewhat technically speaking, this means: Interpreting text/data Classifying it Using discovered classification to steer through relationships between terms Traversing relationships to iteratively form subsequent inquiry “ Semagically” getting just what you want
- 38. What about the Semantic Web jargon? URIs – (uniform resource identifiers) each and every “thing” has a standard name and a permanent place to be referenced RDF and OWL – language and syntax to describe relationships between things – (incl: ontologies) Reasoners – systems like Pellet use formal logic to infer or draw conclusions based relationships (father’s brother is uncle…) GRDDL – gleans and extracts rdf data from web pages SPARQL – language for asking questions (querying) and receiving answers of semantically compliant data.
- 39. Remember this slide? <rdf:Description rdf:about="http://www.w3.org/TR/rdf-syntax-grammar"> <ex:editor> <rdf:Description> <ex:homePage> <rdf:Description rdf:about="http://purl.org/net/dajobe/"> </rdf:Description> </ex:homePage> </rdf:Description> </ex:editor> </rdf:Description> <rdf:Description rdf:about="http://www.w3.org/TR/rdf-syntax-grammar"> <ex:editor> <rdf:Description> <ex:fullName>Dave Beckett</ex:fullName> </rdf:Description> </ex:editor> </rdf:Description> <rdf:Description rdf:about="http://www.w3.org/TR/rdf-syntax-grammar"> <dc:title>RDF/XML Syntax Specification (Revised)</dc:title> </rdf:Description>
- 40. Remember this slide? <rdf:Description rdf:about="http://www.w3.org/TR/rdf-syntax-grammar"> <ex:editor> <rdf:Description> <ex:homePage> <rdf:Description rdf:about="http://purl.org/net/dajobe/"> </rdf:Description> </ex:homePage> </rdf:Description> </ex:editor> </rdf:Description> <rdf:Description rdf:about="http://www.w3.org/TR/rdf-syntax-grammar"> <ex:editor> <rdf:Description> <ex:fullName>Dave Beckett</ex:fullName> </rdf:Description> </ex:editor> </rdf:Description> <rdf:Description rdf:about="http://www.w3.org/TR/rdf-syntax-grammar"> <dc:title>RDF/XML Syntax Specification (Revised)</dc:title> </rdf:Description> This is the granular form. But SW folks now get that regular people brought in via web2.0 won’t write this. So don’t worry about it.
- 41. For those who live in spreadsheets One way to think of this granularity is with a spreadsheet - Row of items implies relationship between items, (columns are characteristics) OK to repeat data in multiple rows, but better to reference one “real” item (repeats would be references too) Item changes. Everything pointing to it reflects change … like collapsing redundant data into a single named entity, and explicitly defining the relationships between the columns. See animation
- 42. Web2>3 Let’s go back to tagging (tag clouds) See Del.icio.us Combine Visualize Weight Emphasize Focus Emerge Expose Discover
- 43. Web2>3 (clustering) See ZigTag site
- 44. NLP and Entity Extraction (using Calais) Try this live by going to http://sws.clearforest.com/calaisviewer/ In this example we discussed Calais using natural language processing to “read” text submissions, re-displaying with items marked and classified. We then considered how this can enable asking the web questions on a grand scale, and tabulation of results of NLP from millions of pages.
- 45. … applied to blogging (e.g. Zemanta or Tagaroo) See live example In this example we looked at how the Zemanta plug-in “reads” a blog post while you are writing it, and provides images and links for you to include in the post – based on its “sense” that these are related to what you are writing.
- 46. … embedding semantics (e.g. using SIOC)
- 47. If you want to get into the weeds a bit Behind the scenes at the “semantic web” companies you’ve probably heard of – like: Metaweb’s Freebase Radar Networks’ Twine Soon-to-be Microsoft’s Powerset… it is all about named entities, classes and type data - and these services, while not making it so apparent, are trying to build this (whereas Google is, at least outwardly, more about statistical ranking – likelihood of meaning). See Trueknowledge at http://beta.trueknowledge.com/ and try finding an entity (person, place, event, thing…) and try out “Add Knowledge” to see the process of unambiguous entry.
- 48. Linking Data (global scale – possibilities emerge) Source: DERI http://richard.cyganiak.de/2007/10/lod/
- 49. Progression from 50,000 feet
- 51. Web Science Source: WSRI - http:// webscience.org /
- 52. Web Science Source: WSRI - http:// webscience.org /
- 53. Web Science Source: WSRI - http:// webscience.org /
- 54. Other links and examples http://webscience.org/ and http://www.w3.org/2001/sw/ http:// sws.clearforest.com/calaisviewer / http://wiki.dbpedia.org and http://deri.org/ and http://sioc-project.org/ http://www.talis.com/platform/podcasts http:// danbri.org / http:// mashmaker.intel.com /web/ - learning masher http:// beta.trueknowledge.com http://www.readwriteweb.com/ (e.g archives/trueknowledge_demo.php) http:// healthmap.org /en and http:// tagmaps.research.yahoo.com /
- 55. Contact Info: Eric Hoffer Second Integral, LLC http://www.secondintegral.com email: ehoffer@secondintegral.com