Semantic Web Services

woodruffpassionateInternet and Web Development

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

135 views

CSI5389

www.site.uottawa.ca/~afato092

1

Semantic Web Services

Sheila A. McIlraith, Tran Cao Son, and Honglei Zeng

Stanford University

2001

Presenter:
Ali Fatolahi

CSI 5389

Winter 2006

CSI5389

www.site.uottawa.ca/~afato092

2

The Problem


Web


Toward service provider


Computers


Moving inside devices


Web usage


Getting automated


Web Page Change


Needs an API Change

CSI5389

www.site.uottawa.ca/~afato092

3

The Solution


Web Services need to be


Computer
-
Interpretable


Use
-
Apparent


Agent
-
Ready


Semantic Web Service is a Web Service


Unambiguous and Machine
-
Understandable


Properties, Capabilities, Interfaces, and Effects

CSI5389

www.site.uottawa.ca/~afato092

4

Realization


AI
-
Inspired Content Markup Languages


Ontology Interface Layer (OIL
1
)


DARPA Agent Markup Language (DAML
2
)


DAML +OIL
3


DAML

L
4


Still in Progress

CSI5389

www.site.uottawa.ca/~afato092

5

Approach of This Paper


An Agent
-
Independent
Declarative API


Data and Metadata


Properties and Capabilities


Interface for Execution


Prerequisites and Consequences


Semantic Markup Exploiting Ontology


DAML Family


Distributed Knowledgebase (KB)

CSI5389

www.site.uottawa.ca/~afato092

6

Role of the Distributed KB


Enabling Agents to Perform Automatic


Discovery


Execution


Composition and Interoperation


A Means for Agents to


Populate their Local KBs


Reason about Web Services

CSI5389

www.site.uottawa.ca/~afato092

7

The Manual Web Service Usage


Use a search engine to find a service


Either read the web page associated with
that service or execute the service


Fill out the forms and click the button


Inform the user about possible compositions


Interact with the user step by step




CSI5389

www.site.uottawa.ca/~afato092

8

The Automatic Web Service Usage


User just defines her objectives


An agent does everything using


WS Descriptions on WS Sites


Ontology
-
Enhanced Search Engines


Main Tasks in Automatic Mood


Discovery


Execution


Composition and Interoperation

CSI5389

www.site.uottawa.ca/~afato092

9

Automatic Discovery


Automatically Locating Web Services


Given Particular Service and Its Properties


Semantic Markup at the Web Sites


To Specify the Service


How to Find?


Service Registry


(Ontology
-
Enhanced) Search Engine

CSI5389

www.site.uottawa.ca/~afato092

10

Automatic Execution


Automatically Executing a Web Service


By a Computer Program or Agent


Semantic Markup of Web Services


A Declarative, Computer
-
Interpretable API for
Executing Services


Tells the Agent What Input Is Necessary


What Information Will Be Returned


How to Execute and Interact With the Service

CSI5389

www.site.uottawa.ca/~afato092

11

Automatic Composition and
Interoperation


Automatic


Selection


Composition


Interoperation


Given a High
-
Level Description of the Objective


Semantic Markup of Web Services


A software can manipulate this markup


Specification of the Objectives, Constraints, …

CSI5389

www.site.uottawa.ca/~afato092

12

State of the art


Today they are not automatic


Lack of Content Markup and a Suitable Markup
Language


Academic Research


Agent matchmaking research (Lark
5
)


XML
-
based standards in industry


UDDI, WSDL, ebXML, …

CSI5389

www.site.uottawa.ca/~afato092

13

What distinguishes this paper?


Expressive Semantic Web Markup
Language


Well
-
Defined Semantics


A Semantic Layer


Sitting On Top of WSDL


A Richer Level of Description


More Sophisticated Interactions


Reasoning

CSI5389

www.site.uottawa.ca/~afato092

14

Markup Strategy


Markup Subjects


Web Services


User and Group Constraints and Preferences


Agent Procedures


Markup Tool


DAML Ontology


Sharing Common Concepts


Specification and Reuse


Concept Mapping


Composition of New Concepts

CSI5389

www.site.uottawa.ca/~afato092

15

Domain
-
Independent

Service

Complex Service

Primitive Service

Domain
-
Specific

Buy

Buy Ticket

Buy Movie Ticket

Buy Airline Ticket

Customer

Origin

Destination

Departure Date

N includes 1

CSI5389

www.site.uottawa.ca/~afato092

16

WS Discovery Markup


Associates properties with services


Relevant to Classification and Selection


Example: BuyUALTicket Service


Service
-
Independent Property Types


Company Name, The Service URL, Unique Service
Identifier, The Intended Use, …


Service
-
Specific Property Types


Valid Methods of Payment, Travel Bonus Plans
Accepted, …

CSI5389

www.site.uottawa.ca/~afato092

17

WS Execution Markup


Requires a dataflow model


Function Metaphor


Process or Conversation Model


It must enable the agent to


automatically construct and execute a Web
service request


interpret and potentially respond to the
service’s response.

CSI5389

www.site.uottawa.ca/~afato092

18

WS Composition Markup


AI
-
Based
action metaphor


Plan Domain Description Language (PDDL
6
)


Parameters, Preconditions and Effects


Defines Consequences of WS Execution


Buying a Book


Credit Card will be debited


Each WS is conceived as an action


PDDL translation to AI action already exists

CSI5389

www.site.uottawa.ca/~afato092

19

Constraints and Preferences
Markup


Bob Prefers to Drive if


Travel time is less than 3 hours


His company mandates him to


Travel by American Airlines


Markup Language: DAML
-
L


Agent Usage is more challenging than the
Markup itself.

CSI5389

www.site.uottawa.ca/~afato092

20

Agent Technology


Model
-
Based Programming


Comprises a
Model


The Agent’s KB


And a
Program


The generic procedure we wish to execute


Agent Programming Language Golog


Congolog


Situation Calculus


Open Agent Architecture (OAA
7
) sends requests

CSI5389

www.site.uottawa.ca/~afato092

21

How it works?


User requests a generic procedure


Agent populates its local KB


adds the user’s constraints to its KB


provides a logical encoding of the preconditions
and effects of the Web service actions


ConGolog instantiates user request into


a sequence of primitive actions


Agent finds appropriate service for each action




CSI5389

www.site.uottawa.ca/~afato092

22

Assessment


There is no sample of:


Situation Calculus


ConGolog


Semantic Markup





We just see the screen shots and general
terms


But it’s a promising work!

CSI5389

www.site.uottawa.ca/~afato092

23

What’s up today?


Based on


J. Nandigam et al. “Semantic Web Services”.2005. [8]


People still trying out


Languages


Environments


Frameworks


Ontologies

CSI5389

www.site.uottawa.ca/~afato092

24

Some Ideas


What about WSFL?


For Procedures


Agent
-
Mediated Semantic WS Technology


Both the Good and the Bad


Semantic Web without Agents?!


Do not interrupt user please!


Opportunity for Agent Technology


Threat for Semantic Web Service

CSI5389

www.site.uottawa.ca/~afato092

25

Thank You, Any Question?

CSI5389

www.site.uottawa.ca/~afato092

26

References

1.
F. van Harmelen and I. Horrocks, “FAQs on OIL: The Ontology Inference Layer,”
IEEE
Intelligent Systems
, vol. 15, no. 6, Nov./Dec. 2000, pp. 69

72.

2.
J. Hendler and D. McGuinness, “The DARPA Agent Markup Language,”
IEEE Intelligent
Systems
, vol. 15, no. 6,Nov./Dec. 2000, pp. 72

73.

3.
www.daml.org/2000/10/daml
-
oil

1.
www.daml.org/2001/03/daml+oil
-
index.html

4.
Somewhere in
www.stanford.edu
!

5.
K. Sycara et al., “Dynamic Service Matchmaking among Agents in Open Information
Environments,”
J. ACM SIGMOD Record
, vol. 28, no. 1, Mar. 1999, pp. 47

53.

6.
M. Ghallab et al.,
PDDL: The Planning Domain Definition Language, Version 1.2
, tech. report
CVC TR

98

003/DCS TR

1165, Yale Center for Computational Vision and Control,Yale
Univ.,New Haven, Conn., 1998.

7.
D.L. Martin, A.J. Cheyer, and D.B. Moran, “The Open Agent Architecture: A Framework for
Building Distributed Software Systems,”
Applied Artificial Intelligence
, vol. 13, nos. 1

2, Jan.

Mar. 1999, pp. 91

128.

8.
Jagadeesh Nandigam, Venkat N Gudivada and Mrunalini Kalavala. “SEMANTIC WEB
SERVICES” Proceedings of the CCSC: Midwestern Conference, JCSC 21, 1 (October 2005). pp.
50
-
63.