OSM Lecture & Exercise

wrendeceitInternet και Εφαρμογές Web

21 Οκτ 2013 (πριν από 3 χρόνια και 5 μήνες)

76 εμφανίσεις

OSM Lecture (14:45
-
16:15)

Takahira

Yamaguchi



OSM Exercise (16:30
-
18:00)

Susumu
Tamagawa


TBL 1
st

Proposal


Information

Management
: A Proposal

(1989)




Links have the following types:


depends on


is part of


made


refers to


uses


is an example of



Users should do data analysis by WWW:


Automatic analysis such as comparison


Automatic generation of mailing lists


Deriving real organization structure

TBL 2nd Proposal (2001)

Tim
Berners

Lee, James
Hendler
,
Ora

Lassila
: The Semantic Web,
http://www.sciam.com/article.cfm?articleID=00048144
-
10D2
-
1C70
-
84A9809EC588EF21&sc=I10032


Semantic Web


Make Web real Hypertext


Data(instance) should have types (class)


Links(property) should have types


Links = relationship between resources


Data Analysis


SW can make inference

“The
Semantic Web

will globalize
KR
,


just as the
WWW

globalized
hypertext

http://www.semanticweb.org/about.html#bigpicture

Big Picture for Semantic Web (2001)

Intelligence


Machine Intelligence



Knowledge Representation


(1980
-

:
Expert Systems


1990
-
: Knowledge Modeling


-
>
Ontologies
)




Web Intelligence



(2000
-
: Semantic Web)


Expert System Overview

KB

KA

IE

WS

DI

EF


(Domain Expert)

User

Major Components

Knowledge
Base

Inference Engine

Working Space

Other Components



Knowledge
Acquisition


Explanation
Facility


Dialog Interface

Expert Systems from KM

How Organizations Manage What They
Know,

by Thomas H. Davenport & Laurence
Prusak

(1998)


Technical systems in narrow domain knowledge



not been supplanted as knowledge providers



difficult to maintain or add knowledge (rules) (Authorizer’s
Assistant ES for credit authorization

→ knowledge is stable and so still used in AMEX

XCON →knowledge changes and so DEC stopped using it)



Why so hard to maintain rules?


other experts besides developer do not understand


what rule primitives mean.

So they should be specified.


ontologies

(in the filed of knowledge engineering)



Ontoligies

in KE and Semantic Web

KE


90
-
: an explicit specification of
conceptualization by Tom Gruber
93


Ontolingua


General Ontology

CYC,
WordNet
, EDR



Ontology Development
Methodology





Semantic Web


95
-
97
:
XML
as arbitrary structures


97
-
98:
RDF


98
-
99:
RDFS



2001
:
DAML+OIL 2
nd

proposal by TBL


2004.2.10:
OWL



2009.10

OWL2


2010.7
RIF

Semantic Web Stack


2001


-
>

2006

2001

2006

2010

Web

Pages

Google

Rich Snippets

YAHOO!

Search

Monke
y

Dapper

Piggy Bank

Scraping

GRDDL

RDFa

HTML5

microdata

Annotation

microformats

Life Science

Government

Media

Linked Open Data (LOD)

Open

Data

Government

Science

Music

Pictures


USA

UK

Science

Commons

MusicBrainz

Flickr

Folksonomy

Wikipedia

Collective Intelligence

See Movie on Semantic Web
Introduction (4 minutes)

RDF

(Resource Description Framework)

RDF Basic (Data) Model
(1)

Resource

information resource identifiable by URI

Property
: relationship between resources


or resource’s attribute


Statement = (Subject, Predicate, Object)

(
Tuple
)



(Resource, Property,(relationship), Resource)



(Resource, Property(attribute) , Value)


URI

URI

URI

RDF Basic Model

2




RDF Data Model
-
> Arced Graph with
Lables


Subject, Object: Node

Predicate: Arc (Property)


Resource: Ellipse (Subject, Object )

Literal: Rectangle

(Object )

http://www.yamaguti.

comp.ae.keio.ac.jp/


Takahira

Yamaguchi

dc
: creator

Resource

Property

Literal

Statement

RDF Structured Model (1)

Putting together two or more statements

Some resource is object in statement 1 and

subject in statement 2.



Ueda

Tama

breed

Ueda

breed

Tama

white

male

sex

color

RDF Structured Model (2)

Blank Node

http://www.bb2.com
’s author is
Ryou

Imai.

His mail address is
webmaster@imai.com
.

http://
www.bb2.com


Ryou

Imai

webmaster@imai.com

author

Name

mail address

Blank Node

To put together resources

Outside web sites cannot see it.

RDF Syntax (1)

http
://yamaguchi.com

Takahira

Yamaguchi

dc:creator

<
rdf:RDF


xmlns:rdf
="http://www.w3.org/1999/02/22
-
rdf
-
syntax
-
ns#"


xmlns:dc
="http://purl.org/dc/elements/1.1/">


<
rdf:Description

rdf:about
="http://kanzaki.com">


<
dc:creator
>
Takahira

Yamaguchi</
dc:creator
>


</
rdf:Description
>

</
rdf:RDF
>

@prefix dc: <http://purl.org/dc/elements/1.1/> .

@prefix
rdf
:

<http://www.w3.org/1999/02/22
-
rdf
-
syntax
-
ns#> .

<
http://kanzaki.com>


dc:creator


Takahira

Yamaguchi"
.

<http://kanzaki.com>
<http://purl.org/dc/elements/1.1/creator>


Takahira

Yamaguchi
"
.

RDF Model

RDF Syntax XML

RDF Syntax Notation3

RDF Syntax N
-
Triple

RDF
XML

S祮ta砠xㄩ1䉡獩挠䵯B敬

rdf
: RDF Model & Syntax name space

Description element

statement

About attribute

Subject resource’s URI

Description element

property(
dc:creator
)

value

Takahira

Yamaguchi


http
://yamaguchi.com

Takahira

Yamaguchi

dc:creator

<
rdf:Description

rdf:about
=

http
://yamaguchi.com
">


<
dc:creator
>
Takahira

Yamaguchi</
dc:creator
>

</
rdf:Description
>

<
rdf:Description



rdf:about
=

http
://yamaguchi.com


dc:creator
=

Takahira

Yamaguchi”
/>

Shorter Version

RDF
XML

S祮ta砠x㈩2却牵捴畲敤e䵯M敬

<
rdf:RDF

...
>


<
rdf:Description

rdf:about
=“
urn: ISBN
-
13:978
-
4274203480
">


<
dc:creator
>
Takahira

Yamaguchi</
dc:creator
>


<
dc:publisher
>


<
rdf:Description

rdf:about
=" http://
www.ohmsha.co.jp
">


<
ex:webmaster
>Someone’s name</
ex:webmaster
>


</
rdf:Description
>


</
dc:publisher
>


</
rdf:Description
>

</
rdf:RDF
>

urn
: ISBN
-
13:

978
-
4274203480

http://www.ohmsha.co.jp
/


Takahira

Yamaguchi

Someone

s name

dc:creator

dc:publisher

ex:webmaster

RDFS

RDF Scheme

RDFS

What is Ueda? What class?

What is Tama? What class?

Ueda

breed

Tama

white

male

sex

color

What does ‘breed’ mean?

What does ‘sex’ mean?

What does ‘color’ mean?

If software does not make sense of the above names (symbols),

there is no way to enable semantic
-
based intelligent processing.

-
>
Ontologies

(RDFS+OWL)

RDFS vocabulary


RDFS 4 Basic Classes


rdfs:Resource
:



rdfs:Class




rdf:Property




rdfs:Literal





RDFS 7 Basic Properties


rdf:type

class
-
instance


rdfs:subClassOf

superclass
-
subclass


rdfs:subPropertyOf

super property
-
subproperty


rdfs
: domain
(what subject should be)


rdfs
: range
(what object should be)

rdfs:label

rdfs:comment


rdf
:
-
>
http://www.w3.org/1999/02/22
-
rdf
-
syntax
-
ns#

prefix

rdfs
:
-
>
http://www.w3.org/2000/01/rdf
-
schema#

prefix

How to make class

<
rdf:Description

rdf:ID
=“animal">


<
rdf:type

rdf:resource
="
http://www.w3.org/2000/01/rdf
-
schema #Class
"/>


or <
rdf:type

rdf:resoucre
=“&
rdfs;Class
”/>


<
rdfs:subClassOf

rdf:resource
=“#living thing"/>

</
rdf:Description
>


<
rdfs:Class

rdf:ID
=“dog">


<
rdfs:subClassOf

rdf:resource
=“#animal"/>

<
/
rdfs:Class
>

living thing

animal

dog

is
-
a

is
-
a

How to make “person” class

<rdf:Description rdf:ID=

Person

>


<rdf:type rdf:resource=

&rdfs;Class

/>


<rdfs:subClassof rdf:resource=

rdfs;Resource

/>


<rdfs:Class ID=

Person

>

<rdfs:subClassof rdf:resource=

rdfs;Resource

/>

</rdfs:Class>


How to make property


rdfs:subPropertyOf

super property
-
subproperty


rdfs
: domain
(what subject should be)


rdfs
: range
(what object should be)

<
rdf:Property

rdf:ID
=“breed">


<
rdfs:subPropertyOf

rdf:resource
=“
...
"/>


<
rdfs:domain

rdf:resource
=“#person"/>


<
rdfs:range

rdf:resource
=“#cat"/>

</
rdf:Property
>

How to make instance

rdf:type

class
-
instance


<
rdf:RDF


xmlns:rdf
="http://www.w3.org/1999/02/22
-
rdf
-
syntax
-
ns#"


xmlns:rdfs
="http://www.w3.org/2000/01/rdf
-
schema#" >



<
rdf:Description

rdf:about
=“#Tama">


<
rdf:type

rdf:resource
="#cat"/>


</
rdf:Description
>



<
rdf:Description

rdf:about
=“#Ueda">


<
rdf:type

rdf:resource
=“#person"/>


<breed

rdf:resource
=“#tama"/>


</
rdf:Description
>

</
rdf:RDF
>

Ueda

breed

rdf:type

Tama

person

cat

rdf:type

RDF/RDFS

William

Charles

subject


Resource


predicate


Property


object
(
Literal
)

hasFather

statement

Model

Syntax

<rdf:Description about=“William”>


<hasFather>Charles<hasFather/>


<rdf:type resource=“Person”/>

</rdf:Description>

Resource

Property

Literal

Person

rdf:type

object
(
Resource
)

<rdf:Class rdf:about=“Person”>


<rdfs:subClassOf resource=“&rdfs;Resource”/>
</rdf:Class>

<rdf:Property rdf:about=“hasFather”>

</rdf:Property>

Syntax

Model

Person

hasFather

rdfs:Class

rdf:Property

rdf:type

rdf:type

rdfs:Resource

rdfs:subClassOf

(Class)

(Property)



































RDF



R D F S