Web Services: Tues 15:30-17:00

elbowsspurgalledInternet and Web Development

Oct 21, 2013 (3 years and 7 months ago)

80 views

Web Services: Tues 15:30
-
17:00

08 November 2005

15:17



“Automated Business
-
to
-
Business Integration of a Logistics Supply Chain
using Semantic Web Services Technology”

Chris Preist
, Javier Esplugas Cuadrado, Steve Battle, Stuart Williams, Stephan
Grimm



Pasted from <
http://iswc2005.semanticweb.org/Ptimetable.html
>





Semantic web enabled web services



Case studies

o

1: HP messaging



3 service providers: road, sea, road



Existing supply chain



Probl
em w shipment service so need to rapidly find another



Macro architecture

o

Logistics coordinator

o

Discovery service

o

Service providers



Find replacement service & insert into supply chain



Service providers

o

Register class of service



Service requestor arch

o

Discov
ery & pre
-
contractual

o

Protocol mgt



Abstract status



Concrete status

o

Domain kb (rdf base)



Incl ontology



Instance info

o

Domain logic



Coord of services



Intelligent decision making



Simple workflow

o

Message transport



Discovery

o

Service & goal descs



Use logistics on
tology in OWL



Capabilities in OWL

o

Look for intersection betw goals & provision



Protocol

o

Abstract



State machine



Business level

o

Concrete



How each indiv action is implemented



Lift

edifact message into abstract info



Then
Lower
ed into rosettanet message



Impleme
ntation

o

Java, Tomcat, Axis

o

RACER

o

JENA



Benefits

o

Reduces false positives

o

Rapid integration of new service suppliers



Can be applied elsewhere

o

Need to describe domain ontologies



Qs

o

Need to focus on protocol integration



“Ontological Approach to Generating Per
sonalized User Interfaces for Web
Services”

Deepali Khushraj
, Ora Lassila



Pasted from <
http://iswc2005.semanticweb.org/Ptimetable.html
>





Key ideas

o

Automate generation of dynamic UIs for W
Ss

o

Then creation of
personalized

UIs



Req semantic info re user



SWS

o

OWL
-
S & WSMO



Used OWL
-
S



Generate UI

o

Ontology representing UI model



Concepts for



Field ordering



Labels



Widget types



Ordering & selection of options for fields w pre
-
determined vals



Grouping



Scheme for creating RDF/OWL instances

o

Rendering alg uses OWL
-
S desc & UI model to dynamically render the
UI



Semantic caching

o

Cache data annotated w semantic model

o

Data from



Personal profile



CoaC (from FoaF)



PIM info



Context



History of I/Os



Corporate data



S
ocial context

o

Alg adds & evicts objects from cache



Possibly including related objects

o

Retrieval



Implicit types det using reasoner



Results prioritized based on data source type, usage freq,
semantic distance



Filtering based on context & hsitory



Process mode
l used to prioritize

o

Rendering alg customizations



Eliminate user input if answer known



Change UI widgets



Provide def vals



Reorder/narrow down lists



See paper for diag of arch

o

Transformation proxy to add semantic info to objects without it (e.g.
calendar)



F
uture

o

Migrate to XForms

o

Policy
-
based extensions

o

Tool for creating UI models

o

Optimization & enhancement of semantic cache mgt alg



Qs

o

Privacy: put user semantic cache on mobile



Then merge on the fly

o

Use of semantic distance to determine best match

o

Need notio
n of 'task' incorporated



Part of context?

o

Cache compartmentalized and drawn together at run time



“An Application of Semantic Web Technologies to Situation Awareness”

Christopher Matheus
, Mieczyslaw Kokar, Kenneth Baclawski, Jerzy Letkowskios



Pasted fr
om <
http://iswc2005.semanticweb.org/Ptimetable.html
>





Problem domain

o

Situation Awareness (SAW)



Sensory info



Fuse into higher order rels



Context dependent & goal directed



Req domain kn

o

Reqs



D
omain kn



Spec conditions that define higher
-
order rels



Reasoning about time
-
dep sensor info



Research focus

o

Defn SAW using
Speckware



Converted from OWL

o

Prototype



Methodology

o

Subject matter experts



Develop ontologies



Dev rules to define complex rels



Grounded

in observable data annotated by ontologies

o

Jess

inference engine



Domain kn

o

OWL provides good basis



But lack of explicit implication: not sufficient

o

Used RuleML then SWRL



Has formalism



Additional representational capabilities



More complex rels



Easy to conv
ert SWRL to Jess



** Presentation stopped after power cut to this side of Galway!!