Hera: Development of Semantic Web Information Systems

1
Hera Development of Semantic Web Information
Systems

• Geert-Jan Houben
• Peter Barna
• Flavius Frasincar
• Richard Vdovjak
• hera_at_wwwis.win.tue.nl

2
Overview

• WIS design
• Hera methodology, RDF(S)
• Conceptual model and integration
• Application model and adaptation
• User interaction

3
Motivation

• From Web pages to Web information system (WIS)
• Technologies from Semantic Web RDF(S)
• Hera uses RDF(S) for effective support of WIS
  design

4
WIS Design

• Generation of hypermedia presentations
  navigation structure
• Presentation objects, e.g. pages
• Navigation connections, e.g. hyperlinks
• Integration from different sources transparent
  repository
• Management of semi-structured data
• Personalization user adaptation

5
Adaptation

• Presentations must be adaptable to different
  users/user platforms
• Devices (PC, PDA, WAP Phone, WebTV etc.)
• Device capabilities (display size, memory size,
  network speed, etc.)
• User preferences (desired layout, navigation
  patterns, etc.)
• User browsing history

6
HERA Architecture
7
Hera WIS Design Methodology

• RMM, OOHDM, WebML, etc.
• Sequence of steps in designing a web application
• Model-driven approach
• data/navigation/presentation
• Data transformations
• towards HTML, WML, SMIL, etc.
• Use of RDF(S) to specify different models and
  XSLT to transform (meta)data
• subclass/subproperty
• extensibility, e.g. CC/PP vocabulary

8
(No Transcript)
9
(No Transcript)
10
Conceptual Model (CM)

• Provides a uniform semantic view over different
  data sources that are integrated within a given
  Web application
• Consists of hierarchies of concepts relevant
  within the given domain, their properties, and
  relations
• Encoded in RDF(S)

11
Conceptual Model Example
12
Source Clusters
Photo Rental

• Sources are
• Autonomous
• (Virtually) grouped to clusters based on the
  content they provide
• RDF(S), RQL capable

Photo Stock Agency
13
Integration Model

• IM decouples the CM and Sources
• Articulations
• actual links between the CM and the source
  ontologies
• (a part of it) serves as a query on the source
  side
• Decorations
• offer a way to rank sources within the same
  cluster
• capture explicitly designers knowledge about
  sources
• open possibilities for queries with constraints
• e.g. Im interested in the answer within 1s,
  otherwise forget it

14
Integration Model Ontology

• Path expression
• Articulation
• Decoration
• Processing instruction

Application independent
15
Application Model (AM)

• Captures navigational view over CM, describing
  hypermedia aspects
• Slices are meaningful presentation units
• Associated to concepts from CM
• Containing properties/attributes and possibly
  other slices
• Slices are linked together with slice
  relationships
• Aggregation relationships index, tour, indexed
  guided tour etc.
• Reference relationships link with an anchor
  specified
• Encoded in RDF(S)

16
Application Model Example

17
Adaptation

• WIS are accessed through multitude of devices and
  by different users
• Device capabilities
• User preferences
• Browsing history
• Adaptation based on conditioning the appearance
  of slices in AM

18
Adaptation/User Model

• Captures two kinds of adaptation
• Adaptability takes into account the context in
  which the user will use the presentation (e.g.
  the browsing platform)
• Adaptivity means that the presentation changes
  itself according to the state of the users
  mind while being browsed
• Consists of
• Device/User Profile captures static visual and
  platform preferences encoded in CC/PP
• User Model represents the dynamic users state,
  e.g. did the user visit (learn) this slice
  (concept)
• Application and Update Rules describe the
  behavior of the presentation (e.g. conditional
  slices in AM) and keep the user model up-to-date
  (AHAM rules)

19
(No Transcript)
20
Profile Example

• Device/User Profile (CC/PP encoding)
• Screen size 100x80
• Preferred language English

ltrdfDescription rdfaboutProfilegt ltccppcompon
entgt ltprfHardwarePlatformgt ltprfImageCapablegtNo
lt/prfImageCapablegt ltprfScreenSizegt100x80lt/prfS
creenSizegt lt/prfHardwarePlatformgt lt/ccppcomp
onentgt
ltccppcomponentgt ltupUserPreferencesgt ltupLangua
gegtEnglishlt/upLanguagegt lt/upUserPreferencesgt
lt/ccppcomponentgt lt/rdfDescriptiongt
21
(No Transcript)
22
Adaptation Model Syntax

• Adaptability Condition
• Adaptivity Condition

ltrdfsClass rdfIDSlice.painting.picture
sliceconditionprfImageCapableYesgt
ltrdfsubClassOf rdfresourceSlice/gt lt/rdfsCla
ssgt
ltrdfsClass rdfIDSlice.painter.main
sliceconditionumBiography falsegt
ltrdfsubClassOf rdfresourceSlice/gt lt/rdfsCla
ssgt
ltrdfsClass rdfIDSlice.painting.main
sliceconditionumPainter gt 10gt
ltrdfsubClassOf rdfresourceSlice/gt lt/rdfsCla
ssgt
23
User Interaction

• E-commerce applications often require WIS with
  broader functionality than just navigation
  through static web sites (known patterns
  shopping carts, on-line payments, searches, etc.)
• The functionality includes also richer means of
  interaction with users via interaction elements
  buttons, text entry forms, checkboxes, etc.
• Navigation objects (slices) with data content may
  depend on the interaction and/or system business
  logic, so it cannot be static

24
User Interaction Specification

• Extension of AM specification
• Structural interaction elements as active
  attributes capturing user actions (buttons, text
  fields, etc.)
• Behavioural operations assigned to interaction
  elements (dynamic navigation, data manipulation,
  adaptation, call of external Web Services, etc.)
• Consequence on WIS architecture need for an
  engine providing the operations (as AHA! for
  adaptivity)

25
Shopping Cart example of CM data manipulations
26
Conclusion Future Work

• Explicit semantics in models, expressing
  concepts, their hierarchies and relationships
• Adaptation in all design steps including the CM
  and IM, and full support of adaptivity
• Experiment with higher ontology languages (e.g.
  OWL) as the basis for the different models
• Further development of authoring tools helping
  the designer to build models in all design steps
• Query and transformation language, e.g. RAL

27
(No Transcript)