Web Semantic Web enables Adaptive Hypermedia Linking in the Wild - PowerPoint PPT Presentation

Loading...

PPT – Web Semantic Web enables Adaptive Hypermedia Linking in the Wild PowerPoint presentation | free to download - id: b1c44-MThhY



Loading


The Adobe Flash plugin is needed to view this content

Get the plugin now

View by Category
About This Presentation
Title:

Web Semantic Web enables Adaptive Hypermedia Linking in the Wild

Description:

Journal of Digital Information, Volume 5 Issue 1. Article No. 247, 2004-05-27. 18 ... http://www.flickr.com/photos/torbein. Linking in the Wild geddit?? 46 ... – PowerPoint PPT presentation

Number of Views:130
Avg rating:3.0/5.0
Slides: 68
Provided by: seanbec
Learn more at: http://www.cs.man.ac.uk
Category:

less

Write a Comment
User Comments (0)
Transcript and Presenter's Notes

Title: Web Semantic Web enables Adaptive Hypermedia Linking in the Wild


1
Web Semantic Web enables Adaptive Hypermedia?
Linking in the Wild
  • Carole Goble
  • Information Management Group
  • The University of Manchester, UK
  • carole_at_cs.man.ac.uk

2
Thanks!
  • Sean Bechhofer
  • Yeliz Yesilada
  • Bernard Horan

David De Roure
3
Me, me, me, me, me
  • Semantic Web applications
  • Life Sciences
  • E-Science
  • The Semantic Grid
  • Old Skool Conceptual Hypermedia

4
This is the story of a Semantic Web Hypermedia
application
  • Good stuff
  • And what went wrong

5
A successful hypermedia of linked resources
6
Hypermedia lives on the web!
7
Web as a hypermedia
  • Simple is Good
  • Simple is Limited
  • NODES and links
  • Hard coding
  • Format restrictions
  • Ownership
  • Link typing. Not
  • Authoring and maintenance experience
  • Adaptation

8
Adaptive hypermedia on the Web
9
You, you, you, you, you
10
Where to start?Old Skool Open Hypermedia
Architecture
Link object
Anchor object
Anchor object
Link Service
Node Object
Node Object
Link Base
11
DLRScirca 1998
Web Server
Keywordlinkbases
XLinklinkbases
User profiles
Hyper link Agent
Publisherslinkbases
IR Agent
Proxy
Invertedindex
Peopledatabase
Person Agent
Ontology Agent
Citation Agent
Bibliographydatabase
Web Browser
Carr, Hughs, Hall, De Roure et al, University of
Southampton
12
Generic Links linking by words
13
(No Transcript)
14
OHS went out of fashion
  • The Web worked because it was simple
  • Web Open Hypermedia was an extra layer of
    software engineering
  • Work arounds

15
Meanwhilein adaptive hypermedia and recommender
system land.
16
(No Transcript)
17
The Next Big Thing Adaptive Web-Based
Systems Paul De Bra, Lora Aroyo and Vadim
Chepegin Journal of Digital Information, Volume
5 Issue 1Article No. 247, 2004-05-27
18
Adaptive hypermedia on the Web
19
Semantic Web on the Web
20
So what do YOU think the Semantic Web is about?
  • Semantic Web
  • Semantic Web technologies
  • Semantic technologies

21
The Semantic Web
The 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.
T. Berners-Lee, J. Hendler, O. Lassila, The
Semantic Web, Scientific American, May 2001
  • Resources should be more accessible to automated
    processes
  • Semantic markup
  • Metadata annotations that describe
    content/function

22
Scientific American, May 2001
Chuck D sez
Dont Believe the Hype!
  • Realising the complete vision is hard
  • Make a start by adding semantic annotation to web
    resources

23
Web page markup rendering information Hyper-link
s to related content
Semantic content is accessible to humans but not
(easily) to computers
24
Annotation assert facts using terms (metadata in
RDF) Represent terms and their relationships
(ontology in RDFS/OWL)
Event
Organisation
Conference
University
Topic
Funder
Date
Professional Body
Company
Location
25
Integration use a uniform common model in
RDF Connecting through shared terms and shared
instances Preserving context and provenance
26
Inference Logic-based classification and
validity checking using OWL Rules using SWRL
(Semantic Web Rule Language) RDF queries Just
making connections because so much stuff is
connected!
James Hendler Science and the Semantic Web
Science 299 520-521, 2003
27
Building a Semantic Web
  • Annotation
  • Associating metadata with resources
  • Integration
  • Integrating information sources
  • Inference
  • Reasoning over the information we have.
  • Could be light-weight (taxonomy)
  • Could be heavy-weight (logic-style)
  • Interoperation and Sharing are key goals

28
Languages Lego
SWRL
Expressive models
Inference
Model fusion
OWL
Controlled vocabularies
RDF(S)
Integration
Data fusion
Integration
RDF
Extensible metadata schemas that you dont have
to nail down
Annotation
XML
29
Semantic Web Myth Busting
  • Semantic Web people try to enforce meaning from
    the top
  • The Semantic Web people will require everybody
    to subscribe to a single predefined "meaning" for
    the terms we use
  • The Semantic Web will require users to
    understand the complicated details of formalised
    knowledge representation
  • The Semantic Web people will require us to
    manually markup all the existing web-pages

Acknowledgement Frank van Harmelen, panel WWW2006
30
Semantic Web Technologies for Adaptive
Hypermedia Applications
  • Metadata models knowledge models for Adaptive
    Hypermedia Systems
  • Build ontologies using reasoning
  • Build rule systems using SWRL
  • Create metadata using RDF

31
My experiences of USWTA
  • Promising
  • Validating models
  • Mixed sophistication.
  • OWL to build ontologies RDFS to deploy.
  • Data fusion and graph matching
  • Flexible and extensible self describing schemas
  • Open world more stuff to say

32
My experiences of USWTA
  • Hendler principle
  • A little semantics goes a long way! Easy to be
    seduced by technology and be too clever.
  • Technical and feature mismatches
  • Its clever but not my problem. Here is my
    problem, where is the feature? Graph matching,
    views.
  • Practicality mismatches
  • Its hard and expensive! Boring practical stuff
    needed. Like security. Trust. Concrete data
    types. Theory isnt enough.
  • Tool Maturity Scalability, reliability,
    performance
  • RDF Triple store death. Graph matching slow down.
  • Unbalanced lifecycle I have started but I cant
    finish
  • Technology visibility
  • ltangle brackets are not for people, and neither
    is RDF/gt

33
My experiences of USWTA
  • Generally good! But
  • Hendler principle A little semantics goes a long
    way!
  • Technical and feature mismatches Its clever but
    not my problem. Here is my problem, where is the
    feature?
  • Practicality mismatches Boring practical stuff
    needed. Trust. security. Trust. Concrete data
    types.
  • Tool Maturity Scalability, reliability,
    performance
  • Unbalanced lifecycle I have started but I cant
    finish
  • Technology visibility ltangle brackets r not 4
    people/gt

34
My experiences of USWTA
  • Hendler principle
  • Technical and feature mismatches
  • Practicality mismatches.
  • Tool Maturity
  • Unbalanced lifecycle
  • Visualisation to people

35
Semantic Web for Searchyawn.
Find images of Steve Furber
Sean Bechhofer
Alan Rector
Rev. Alan M. Gates, Associate Rector of the
Church of the Holy Spirit, Lake Forest, Illinois
36
Semantic Web for HypermediaWhat about using the
Semantic Web for linking??Hypermedia is a
Semantic Web ApplicationUse the Semantic WEB
for, like, you know, the WEB. Navigation and
stuff. For people.
37
The Semantic Web is a Web
Web page
Any Web Resource
URIgt
lta href
HTML
lta hrefhttp//gt
URI
URI
URI
RDF is like the web!
RDF
Courtesy, Jim Hendler
38
SW as Web
  • The Semantic Web is a Web a collection of
    linked nodes.
  • People will use it as well as agents, and for
    them, navigation of links is a key mechanism for
    exploring the space.
  • Question
  • Semantic Web approaches and technology offer the
    potential for dynamic linking of documents
  • How best to we apply these technologies in order
    to achieve our goals?

39
Web of Trust
Rules Web
?p -gt ?a pa
?p -gt ?a pa
?p -gt ?a pa
?p -gt ?a pa
The semantics are in the links
?p -gt ?a pa
Ontology Web
Navigate the Semantic Web
Metadata Annotation Web
adapt
Navigate the Web
Web
Deep web
40
Semantic Media Wikis
Courtesy Max Völkel
Link types Yay! KB within Wiki A Semantic
Hypermedia Application on the Web
41
Linking between pages A Semantic Hypermedia App.
on the Semantic Web
42
Add links based on annotation
43
Navigate based on annotation
44
Linking in the Wild
45
Linking in the Wild geddit??
http//www.flickr.com/photos/torbein
46
Conceptual Open Hypermedia ServicE
  • Users
  • Sun IT
  • Technology Office

Conceptual Hypermedia
Semantic Services
COHSE
Open Hypermedia Services
47
Link generator
  • Metadata provides a mechanism
  • for resource discovery provision of source
    anchors.
  • at browse time and authoring time.
  • Annotation allows both linking into and out of a
    resource.
  • Software agents that generates links

48
COHSE System
  • A software agent that generates and presents
    links on behalf of both an author and a reader.
  • Used at browsing time, provides just-in-time
    hypertext.
  • Used at authoring time, supports the authoring
    task.
  • Link Creation, not just Resource Discovery
  • This is not a radical new departure, but is the
    bringing together of existing, reasonably well
    understood technology.
  • It wont solve all the problems!

49
Generic Links Thesaurus
  • A thesaurus can bridge gaps between terms.

Thesaurus
Linkbase
Link Service
Document
Linked Document
50
Generic Links Ontology
  • An ontology can bridge gaps between concepts.

51
Annotation Links Ontology
  • Annotation allows explicit decoration of
    documents with concepts.

Annotations
Ontology
Linkbase
Link Service
Document
Linked Document
52
COHSE Architecture
Semantics and world knowledge
Presentation and delivery of results
Ontology Service
Parser
DLRS
Link Generator
Page Construction
Editorial Knowledge
Resource Service
Resource Discovery Metadata Management
53
  • Hypermedia over Web
  • using Semantic Web technologies
  • Generate your own Semantic Web
  • Hypermedia over the Semantic Web
  • If there is one.

54
Links Added to AH
55
Links associated with Concept
56
Following links to external site
57
Links added to external site
58
  • Adapt and evolve linking and navigation
    infrastructure
  • Change ontology -gt change links
  • Change annotations -gt change links
  • Change link base -gt change links
  • Alternative views / targets for different users
    or user groups
  • Multiple and personalised ontologies
    annotations
  • Monitor annotations concepts followed
  • Change the user, change the semantics

59
Suggest ontologies Based on user and page content
Ontology Service
Parser
DLRS
Link Generator
Page Construction
Editorial Knowledge
Resource Service
User Model
suggest resources adapt decisions
60
More stumbles?
  • The Web worked because it was simple
  • Web Semantic Web is an extra layer of software
    engineering
  • And Content!
  • The Ontology
  • The Annotation!
  • Why might it work this time??

61
The (Semantic Web) future is our (Hypermedia)
past
  • Was link management
  • Now RDF management
  • Was dangling links
  • Now dead annotation and
  • XPointer fraility
  • Was link authoring, now ontology authoring
  • Fashions just cycle!

62
Engineering Complexity
  • Keep the footprint small
  • Browser or proxy or server side link generation
  • Who cares! Have them all!
  • It will break. Get over it.
  • Vanilla technologies. Like COHSE portlet

63
Hypermedia authoring becomes Ontology authoring
  • Its a hurdle
  • Temptation to be too clever
  • Little bit of semantics goes a long way
  • Its for people navigating. More thinking, less
    reasoning.
  • Different business

64
Hypermedia authoring becomes Annotation Authoring
  • Get your link base!!!
  • Annotation generation
  • Using words terms
  • Annotation on demand
  • Generating annotations using search engines
  • Be open and be loose
  • Get the world to tag

65
Trend spotting
  • Web for People
  • Contacts social networks
  • Developing regions Web for all
  • Pervasive Web where you are
  • Link spam defence trusted Web
  • Security a safe Web
  • People Powered Web
  • Content social tagging
  • Search use user behaviour

66
Web 2.0 Rise of the Scruffies
Artificial Intelligence
Consensual
Lots
Reasoning
Ontologies
High-pain, High-gain
Semantics
Folksonomy
Some
Social tagging
Social networks
Thinking
Deferred pain
Collective Intelligence
Web
Lots
Some
67
But ontology authoring and annotation authoring
isnt hypermedia authoring
  • Oh dear

68
Seduced by knowledge
  • An ontology - appealing way of building links
    between resources for open, flexible linking
  • The ontology model drives the navigation between
    resources
  • Navigation based on super/subclass hierarchy
    BUT
  • .A Model of knowledge is NOT a Model of
    Navigation
  • Bound too tightly
  • Bending knowledge models to fit the paradigm

69
Conclusion like, Doh!
  • Separate but connect the navigational model used
    to link resources from the knowledge used to
    describe them.
  • Use disparate knowledge sources with differing
    levels of representation
  • Rich, heavy-weight ontologies
  • Taxonomies and light-weight ontologies
  • Folksonomies and unstructured models
  • Keywords
  • Text mining
  • One size does not fit all.

70
Navigation and Knowledge
?
?
  • What are appropriate models for intersite
    linking?
  • How best do we separate the ontology/knowledge
    from the navigation?
  • How might we map from our diverse knowledge
    structures (and their associated annotations)
    into navigational models?

?
?
?
?
?
?
71
Summary
  • The Semantic Web Web gives new opportunities
    for Adaptive Hypermedia
  • Both the Technologies and the Semantic Web
    itself.
  • Adaptive Hypermedia applications
  • Adapting navigation opportunities over the web
  • Navigating over the semantic web
  • But Linking in the World Wild Web is hard.
  • Temptations lurk .One size does not fit all
About PowerShow.com