semantic web and knowledge management

69
Semantic Web and Knowledge Management Ching-Long Yeh 葉葉葉 Department of Computer Science and Engineering Tatung University Taipei, Taiwan [email protected] (msn) http://www.cse.ttu.edu.tw/chingyeh

Upload: sydnee-bennett

Post on 04-Jan-2016

23 views

Category:

Documents


0 download

DESCRIPTION

Semantic Web and Knowledge Management. Ching-Long Yeh 葉慶隆 Department of Computer Science and Engineering Tatung University Taipei, Taiwan [email protected] (msn) http://www.cse.ttu.edu.tw/chingyeh. Content. Introduction WWW: HTML, HTTP, browsers XML and its Protocol - PowerPoint PPT Presentation

TRANSCRIPT

Page 1: Semantic Web and Knowledge Management

Semantic Web and Knowledge Management

Ching-Long Yeh 葉慶隆Department of Computer Science and Engineering

Tatung UniversityTaipei, Taiwan

[email protected] (msn)http://www.cse.ttu.edu.tw/chingyeh

Page 2: Semantic Web and Knowledge Management

Semantic Web and Knowledge Management 2

ContentContent

• Introduction – WWW: HTML, HTTP, browsers – XML and its Protocol– Business Automation: RosettaNet, ebXML– Semantic Web: WWW + metadata layer – Semantic Grid

• Semantic Web – Overview– Reasoning in Prolog– Languages: RDF, RDFS, OWL, OWL-S, SWRL, SPARQL– Ontologies:

• RSS, FOAF, iCalendar, vCard, DC(Q), musicBrainz– Semantic Web System Architecture

• Knowledge-Engineering Approach to Knowledge Management – KE methodology: CommonKADS

• Our Current Research – Lesson Learned in Project Management Based on Semantic Web – From Text to RDF

• Summary

Page 3: Semantic Web and Knowledge Management

Semantic Web and Knowledge Management 3

Web Technology OverviewWeb Technology Overview

• WWW– Infrastructure

• HTML, HTTP, URI, browsers– Services

• Search engine and directory navigation

• WWW + XML– Web Service (UDDI, WSDL, SOAP)

• SOA (Registry, provider, requester)– ebXML

• SOA for business automation– discovery, implementation, run-time phases

• Business process + message service– Semantic Web

• Meaning processing automation• WWW + metadata layer (OWL+RDF)• Services automation (WWW+OWL-S/RDF)• Semantic Grid

Page 4: Semantic Web and Knowledge Management

Semantic WebSemantic Web

Page 5: Semantic Web and Knowledge Management

Semantic Web and Knowledge Management 5

Semantic WebSemantic Web

• The Semantic Web is a vision:

the idea of having data on the web defined and linked in a way that it can be used by machines not just for display purposes, but for automation, integration and reuse of data across various applications

Page 6: Semantic Web and Knowledge Management

Semantic Web and Knowledge Management 6

Semantic WebSemantic Web

• The Semantic Web = a Web with a meaning.

"If HTML and the Web made all the online documents look like one huge book, RDF, schema, and inference languages will make all the data in the world look like one huge database“

Tim Berners-Lee, Weaving the Web, 1999

Page 7: Semantic Web and Knowledge Management

Semantic Web and Knowledge Management 7

Introduction from W3C SW ActivityIntroduction from W3C SW Activity

• The Semantic Web is a web of data.

• The Semantic Web is about two things. – Common formats for interchange of data,

• On the original Web we only had interchange of documents.

– Language for recording how the data relates to real world objects• That allows a person, or a machine, to start off in one database, and then move

through an unending set of databases which are connected not by wires but by being about the same thing.

Page 8: Semantic Web and Knowledge Management

Semantic Web and Knowledge Management 8

The Semantic Web ArchitectureThe Semantic Web Architecture

(http://www.w3.org/2001/Talks/0228-tbl/slide5-0.html)

Tim Berners-Lee:“Axioms, Architecture and Aspirations”W3C all-working group plenary Meeting28 February 2001

URI Unicode

XML Namespaces

XML Schema

Sig./Ency.

RDF M&S

RDF Schema

Ontology (OWL)

Rules (SWRL)

Logic (FOL)

Proof

Trust

I. Horrocks, et al. Semantic web architecture: Stack or two towers? In F. Fages and S. Soliman, (eds.), Principles and Practice of Semantic Web Reasoning (PPSWR 2005), number 3703 in LNCS, pages 37-41. SV, 2005. http://www.cs.man.ac.uk/~horrocks/Publications/download/2005/HPPH05.pdf

Page 9: Semantic Web and Knowledge Management

Semantic Web and Knowledge Management 9

Reasoning in Prolog Reasoning in Prolog (1)(1)

• Facts and rules about members of a family

parent(tom,bob).parent(pam,bob).parent(tom,bob).parent(tom,liz).parent(bob,ann).parent(bob,pat).parent(pat,jim).female(pam).male(tom).male(bob).female(liz).female(pat).female(ann).male(jim).

offspring(Y,X):- parent(X,Y).mother(X,Y):- parent(X,Y),female(X).grandparent(X,Z):- parent(X,Y),parent(Y,Z).sister(X,Y):- parent(Z,X),parent(Z,Y),female(X), X\==Y.predecessor(X,Z):-

parent(X,Z).predecessor(X,Z):-

parent(X,Y), predecessor(Y,Z).

Page 10: Semantic Web and Knowledge Management

Semantic Web and Knowledge Management 10

Reasoning in Prolog Reasoning in Prolog (2)(2)

• The following unlisted facts can be derived by using the rules.

offspring(bob,pam).offspring(bob,tom).offspring(liz,tom).offspring(ann,bob).offspring(pat,bob).offspring(jim,pat).

mother(pam,bob).mother(pat,jim).

grandparent(tom,ann).grandparent(tom,pat).grandparent(pam,ann).grandparent(pam,pat).grandparent(tom,ann).grandparent(tom,pat).grandparent(bob,jim).

sister(liz,bob).sister(ann,pat).sister(pat,ann).

predecessor(pam,bob).predecessor(tom,bob).predecessor(tom,liz).predecessor(bob,ann).predecessor(bob,pat).predecessor(pat,jim).predecessor(pam,ann).predecessor(pam,pat).predecessor(pam,jim).predecessor(tom,ann).predecessor(tom,pat).predecessor(tom,jim).predecessor(bob,jim).

Page 11: Semantic Web and Knowledge Management

RDF and Schema LanguagesRDF and Schema Languages

Page 12: Semantic Web and Knowledge Management

Semantic Web and Knowledge Management 12

RDF M&SRDF M&S• RDF (Resource Description Framework)

– Beyond Machine readable to Machine understandable

• RDF consists of two parts– RDF Model (a set of triples)

– RDF Syntax (different XML serialization syntaxes)

• RDF Schema for definition of Vocabularies (simple Ontologies) for RDF (and in RDF)

Page 13: Semantic Web and Knowledge Management

Semantic Web and Knowledge Management 13

RDF Data ModelRDF Data Model• Resources

– A resource is a thing you talk about (can reference)

– Resources have URI’s

– RDF definitions are themselves Resources (linkage, see requirement 1)

• Properties – slots, define relationships to other resources or atomic values

• Statements– “Resource has Property with Value”

– (Values can be resources or atomic XML data)

• Similar to Frame Systems

Page 14: Semantic Web and Knowledge Management

Semantic Web and Knowledge Management 14

A Simple ExampleA Simple Example• Statement

– “Ora Lassila is the creator of the resource http://www.w3.org/Home/Lassila”

• Structure– Resource (subject) http://www.w3.org/Home/Lassila– Property (predicate) http://www.schema.org/#Creator– Value (object) "Ora Lassila”

• Directed graph

http://www.w3.org/Home/Lassilas:Creator Ora Lassila

Page 15: Semantic Web and Knowledge Management

Semantic Web and Knowledge Management 15

EmailName

s:Creator

http://www.w3.org/Home/Lassila

Another ExampleAnother Example

• To add properties to Creator, point through an intermediate Resource.

Person://fi/654645635

Ora Lassila [email protected]

Page 16: Semantic Web and Knowledge Management

Semantic Web and Knowledge Management 16

Example:Example: Bag Bag

• The students incourse 6.001 are Amy, Tim,John, Mary,and Sue

Rdf:Bag

/Students/Amy

/Students/Tim

/Students/John

/Students/Mary

/Students/Sue

bagid1

/courses/6.001

students

rdf:type

rdf:_1

rdf:_2

rdf:_3

rdf:_4

rdf:_5

Page 17: Semantic Web and Knowledge Management

Semantic Web and Knowledge Management 17

rdf:_1

rdf:_2

rdf:_3

rdf:typesource

ftp.eu.net

ftp.cs.purdue.edu

ftp.x.org

Example:Example: Alternative Alternative

• The source code for X11 may be found at ftp.x.org, ftp.cs.purdue.edu, or ftp.eu.net

altid

rdf:Althttp://x.org/package/X11

Page 18: Semantic Web and Knowledge Management

Semantic Web and Knowledge Management 18

Representing Prolog Facts in RDFRepresenting Prolog Facts in RDF

parent(pam,bob).parent(tom,bob).parent(tom,liz).parent(bob,ann).parent(bob,pat).parent(pat,jim).female(pam).male(tom).male(bob).female(liz).female(pat).female(ann).male(jim).

Page 19: Semantic Web and Knowledge Management

Semantic Web and Knowledge Management 19

OWLOWLW3C Web Ontology LanguageW3C Web Ontology Language

• OWL provides three increasingly expressive sublanguages: OWL Lite, OWL DL, and OWL Full.

Page 20: Semantic Web and Knowledge Management

Semantic Web and Knowledge Management 20

OWLOWLW3C Web Ontology LanguageW3C Web Ontology Language

OWL Lite language constructs

RDF Schema Features: Class rdf:Property rdfs:subClassOf rdfs:subPropertyOf rdfs:domain rdfs:range Individual

(In)Equality: equivalentClass equivalentProperty sameAs differentFrom allDifferent

Property Characteristics: inverseOf TransitiveProperty SymmetricProperty FunctionalProperty InverseFunctionalProperty

Property Type Restrictions: allValuesFrom someValuesFrom

Restricted Cardinality: minCardinality (only 0 or 1) maxCardinality (only 0 or 1) cardinality (only 0 or 1)

Header Information: ontology imports

Page 21: Semantic Web and Knowledge Management

Semantic Web and Knowledge Management 21

Ontology SpectrumOntology Spectrum

What is an Ontology?What is an Ontology?

Catalog/ID

GeneralLogical

constraints

Terms/glossary

Thesauri“narrower

term”relation

Formalis-a

Frames(properties)

Informalis-a

Formalinstance

Value Restrs.

Disjointness, Inverse, part-

of…

Page 22: Semantic Web and Knowledge Management

Semantic Web and Knowledge Management 22

Creating Your Own OntologyCreating Your Own OntologyA Simple Knowledge-Engineering MethodologyA Simple Knowledge-Engineering Methodology

Step 1: Determine the domain and scope of the ontology

– Why, what, who, competency questions

Step 2: Consider reusing existing ontologies

Step 3: Enumerate important terms in the ontology

Step 4: Define the classes and the class hierarchy

Step 5: Define the properties of classes—slots

Step 6: Define the facets of the slots

Step 7: Create instances

Page 23: Semantic Web and Knowledge Management

Semantic Web and Knowledge Management 23

Obtaining RDF schema from ontology libraryObtaining RDF schema from ontology library

• SchemaWeb: http://www.schemaweb.info/default.aspx • Swoogle: http://swoogle.umbc.edu/ • DAML ontology library: http://www.daml.org/ontologies/

Page 24: Semantic Web and Knowledge Management

Semantic Web and Knowledge Management 24

Examples of RDF schemaExamples of RDF schema

• RSS 1.0: http://www.schemaweb.info/schema/SchemaDetails.aspx?id=12

• MusicBrainz: http://www.schemaweb.info/schema/SchemaDetails.aspx?id=168

• Resume: http://www.schemaweb.info/schema/SchemaDetails.aspx?id=89 • FOAF: http://www.schemaweb.info/schema/SchemaDetails.aspx?id=29

Page 25: Semantic Web and Knowledge Management

Semantic Web and Knowledge Management 25

RDFCalendar

FOAF

Page 26: Semantic Web and Knowledge Management

Semantic Web and Knowledge Management 26

OWL-S: Ontology for Semantic Web OWL-S: Ontology for Semantic Web ServicesServices

• Some motivating tasks– Automatic Web service discovery

– Automatic Web service invocation

– Automatic Web service composition and interoperation

– Automatic Web service execution monitoring

Page 27: Semantic Web and Knowledge Management

Semantic Web and Knowledge Management 27

High-level View of the Service OntologyHigh-level View of the Service Ontology

ServiceProfile

ServiceModel

ServiceGrounding

ServiceResourceprovides

presents

describedBy

supports

What the service does

How it works

How toAccess it

Page 28: Semantic Web and Knowledge Management

Semantic Web and Knowledge Management 28

Top Level of the Process OntologyTop Level of the Process Ontology

Process Profile

AtomicProcess Composite

ProcessSimpleProcess

ControlConstruct

SequenceRepeatUntil

ProcessComponent=Process U ControConstruct

ProcessComponent=Process U ControConstruct

ProcessComponent=Process U ControConstruct

hasProcesshasProfile

computedInputcomputedOutputcomputedEffect

computedPreconditioninvocab

compsedBy

expandcollapserealizes

realizedBy

hasGrounding

InputPrecondition

Outputeffect

Condition

Page 29: Semantic Web and Knowledge Management

Semantic Web and Knowledge Management 29

Grounding a Service to a Concrete Realization Grounding a Service to a Concrete Realization

Process Model

Atomic Process

Operation

DL-Based Types

Inputs/Outputs

Message

Binding to SOAP, HTTP, etc.

WSDL

OWL-S

Page 30: Semantic Web and Knowledge Management

Semantic Web and Knowledge Management 30

SWRL: Semantic Web Rule LanguageSWRL: Semantic Web Rule LanguageExamplesExamples

hasParent(?x1,?x2) hasBrother(?x2,?x3) hasUncle(?x1,?x3)∧ ⇒

Implies(Antecedent(hasParent(I-variable(x1) I-variable(x2)) hasBrother(I-variable(x2) I-variable(x3))) Consequent(hasUncle(I-variable(x1) I-variable(x3))))

Page 31: Semantic Web and Knowledge Management

Semantic Web and Knowledge Management 31

SPARQL: RDF Query LanguageSPARQL: RDF Query LanguageExamplesExamples

SELECT ?x WHERE { ?x <http://www.w3.org/2001/vcard-rdf/3.0#FN> "John Smith" }

SELECT ?x, ?fname WHERE {?x <http://www.w3.org/2001/vcard-rdf/3.0#FN> ?fname}

SELECT ?givenName WHERE { ?y <http://www.w3.org/2001/vcard-rdf/3.0#Family> "Smith" . ?y http://www.w3.org/2001/vcard-rdf/3.0#Given ?givenName . }

PREFIX vcard: <http://www.w3.org/2001/vcard-rdf/3.0#> SELECT ?givenName WHERE { ?y vcard:Family "Smith" . ?y vcard:Given ?givenName . }

PREFIX vcard: <http://www.w3.org/2001/vcard-rdf/3.0#> SELECT ?g WHERE { ?y vcard:Given ?g . FILTER regex(?g, "r", "i") }

PREFIX info <http://somewhere/peopleInfo#> SELECT ?resource WHERE { ?resource info:age ?age . FILTER (?age >= 24) }

Page 32: Semantic Web and Knowledge Management

Semantic Web System ArchitecturesSemantic Web System Architectures

Page 33: Semantic Web and Knowledge Management

Semantic Web and Knowledge Management 33

TypicalTypical System ArchitectureSystem Architecture

Page 34: Semantic Web and Knowledge Management

Semantic Web and Knowledge Management 34

Layered ArchitectureLayered Architecture

Page 35: Semantic Web and Knowledge Management

Semantic Web and Knowledge Management 35

System ArchitectureSystem Architecture

Page 36: Semantic Web and Knowledge Management

Semantic Web and Knowledge Management 36

SesameSesameA generic Architecture for Storing and Querying RDF and RDF SchemaA generic Architecture for Storing and Querying RDF and RDF Schema

Page 37: Semantic Web and Knowledge Management

Semantic Web and Knowledge Management 37

SesameSesame

Page 38: Semantic Web and Knowledge Management

Semantic Web and Knowledge Management 38

Annotea Basic Architecture Annotea Basic Architecture

Page 39: Semantic Web and Knowledge Management

Knowledge Management Based on Semantic WebKnowledge Management Based on Semantic Web

Page 40: Semantic Web and Knowledge Management

Semantic Web and Knowledge Management 40

What is knowledge management?What is knowledge management?

• Knowledge is seen as a resource

• This means for knowledge management taking care that the resource is– delivered at the right time

– available at the right place

– present in the right shape

– satisfying the quality requirements

– obtained at the lowest possible costs

• to be used in business processes

Selected from the course slides of CommonKADS

Page 41: Semantic Web and Knowledge Management

Semantic Web and Knowledge Management 41

Knowledge assets

Apply yourbest knowledge

Construct newknowledge

Value chain

Continuous improvement of knowledge Continuous improvement of knowledge assetsassets

Selected from the course slides of CommonKADS

Page 42: Semantic Web and Knowledge Management

Semantic Web and Knowledge Management 42

Knowledge management & knowledge Knowledge management & knowledge engineeringengineering

• Organization analysis feeds into knowledge management (and vice versa)

• Knowledge modeling provides techniques for knowledge identification and development

• Knowledge engineering focuses on common / reusable elements in knowledge work

Selected from the course slides of CommonKADS

Page 43: Semantic Web and Knowledge Management

Semantic Web and Knowledge Management 43

Knowledge engineeringKnowledge engineering

• process of– eliciting,

– structuring,

– formalizing,

– operationalizing

• information and knowledge involved in a knowledge-intensive problem domain,

• in order to construct a program that can perform a difficult task adequately

Selected from the course slides of CommonKADS

Page 44: Semantic Web and Knowledge Management

Semantic Web and Knowledge Management 44

Problems in knowledge engineeringProblems in knowledge engineering

• complex information and knowledge is difficult to observe

• experts and other sources differ

• multiple representations:– textbooks

– graphical representations

– heuristics

– skills

Selected from the course slides of CommonKADS

Page 45: Semantic Web and Knowledge Management

Semantic Web and Knowledge Management 45

A Short History of A Short History of Knowledge SystemsKnowledge Systems

1965 19851975 1995

general-purpose search engines

(GPS)

first-generation rule-based systems

(MYCIN, XCON)

emergence of structured methods

(early KADS)

mature methodologies

(CommonKADS)

=> from art to discipline =>

Selected from the course slides of CommonKADS

Page 46: Semantic Web and Knowledge Management

Semantic Web and Knowledge Management 46

CommonKADS Model SetCommonKADS Model Set

OrganizationModel

TaskModel

AgentModel

KnowledgeModel

CommunicationModel

DesignModel

Context

Concept

Artefact

Selected from the course slides of CommonKADS

Page 47: Semantic Web and Knowledge Management

Semantic Web and Knowledge Management 47

Why context modeling?Why context modeling?

• Often difficult to identify profitable use of (knowledge) technology

• Laboratory is different from the ''real'' world

• Acceptability to users very important

• Fielding into ongoing process not self evident

• Often not clear what additional measures to take

Selected from the course slides of CommonKADS

Page 48: Semantic Web and Knowledge Management

Semantic Web and Knowledge Management 48

Goals for context modelingGoals for context modeling

• Identify problems and opportunities

• Decide about solutions and their feasibility

• Improve tasks and task-related knowledge

• Plan for needed organizational changes

Selected from the course slides of CommonKADS

Page 49: Semantic Web and Knowledge Management

Semantic Web and Knowledge Management 49

Role of knowledge systemsRole of knowledge systems

• "automation" is not the right way to look at KSs

• tasks are usually too complex

• much better view: KS as process-improvement tool

• typical role of KS: active intelligent assistant

Selected from the course slides of CommonKADS

Page 50: Semantic Web and Knowledge Management

Semantic Web and Knowledge Management 50

• Step 1: Carry out a scoping and feasibility study– Tool: Organization Model (OM)

• Step 2: Carry out impact and improvement study– Tool: Task and Agent Models (TM, AM)

• zooming in/refinement of organization model

• Each study consists of an analysis part and a “constructive” decision-making part

Context modelling processContext modelling process

Selected from the course slides of CommonKADS

Page 51: Semantic Web and Knowledge Management

Project Management Based on Semantic WebProject Management Based on Semantic Web

Page 52: Semantic Web and Knowledge Management

Semantic Web and Knowledge Management 52

Project DescriptionProject Description

Lesson Learned in PM

CommonKADS

Organization modelTask modelAgent modelKnowledge modelCommunication modelDesign models

Implementation using SW technology

System

Page 53: Semantic Web and Knowledge Management

Semantic Web and Knowledge Management 53

System Architecture

Page 54: Semantic Web and Knowledge Management

Semantic Web and Knowledge Management 54

客戶客戶

第三方組織第三方組織

下包廠商下包廠商

Team ATeam A

Team CTeam C

Team DTeam D

主管主管 業務業務

後勤後勤

Team BTeam B

Communication LayerCommunication Layer

Communication ServicesCommunication Services

SMTP

HTTPHTTP

SOAPSOAP

HTTPS

FTPFTP

RSS

Page 55: Semantic Web and Knowledge Management

Semantic Web and Knowledge Management 55

客戶客戶

第三方組織第三方組織

下包廠商下包廠商

Team ATeam A

Team CTeam C

Team DTeam D

主管主管 業務業務

後勤後勤

Team BTeam B

Communication LayerCommunication Layer

Communication ServicesCommunication Services

Security ServicesSecurity Services

Security LayerSecurity Layer

身分確認管理

存取權限管理存取權限管理

加解密

政策管理與管制政策管理與管制

Page 56: Semantic Web and Knowledge Management

Semantic Web and Knowledge Management 56

客戶客戶

第三方組織第三方組織

下包廠商下包廠商

Team ATeam A

Team CTeam C

Team DTeam D

主管主管 業務業務

後勤後勤

Team BTeam B

Communication LayerCommunication Layer

Communication ServicesCommunication Services

Security ServicesSecurity Services

Security LayerSecurity Layer

DataData IntegrationIntegration ServicesServices

Data Integration LayerData Integration Layer

資料轉換版本控制版本控制

二維關聯

資料分類與聚集資料儲存資料儲存

Page 57: Semantic Web and Knowledge Management

Semantic Web and Knowledge Management 57

客戶客戶

第三方組織第三方組織

下包廠商下包廠商

Team ATeam A

Team CTeam C

Team DTeam D

主管主管 業務業務

後勤後勤

Team BTeam B

Communication LayerCommunication Layer

Communication ServicesCommunication Services

Security ServicesSecurity Services

Security LayerSecurity Layer

DataData IntegrationIntegration ServicesServices

Data Integration LayerData Integration Layer

Info.(Meta Data) Info.(Meta Data) IntegrationIntegration ServicesServices

Info.(Meta Data) Info.(Meta Data) IntegrationIntegration LayerLayer

註解資訊編製

後設資料編製後設資料編製

生命週期生命週期與程序管理與程序管理

專案狀態與背景資訊管理

Page 58: Semantic Web and Knowledge Management

Semantic Web and Knowledge Management 58

客戶客戶

第三方組織第三方組織

下包廠商下包廠商

Team ATeam A

Team CTeam C

Team DTeam D

主管主管 業務業務

後勤後勤

Team BTeam B

Communication LayerCommunication Layer

Communication ServicesCommunication Services

Security ServicesSecurity Services

Security LayerSecurity Layer

DataData IntegrationIntegration ServicesServices

Data Integration LayerData Integration Layer

Info.(Meta Data) Info.(Meta Data) IntegrationIntegration ServicesServices

Info.(Meta Data) Info.(Meta Data) IntegrationIntegration LayerLayer

Knowledge(Info.) Knowledge(Info.) IntegrationIntegration ServicesServices

Knowledge (Info.) Knowledge (Info.) IntegrationIntegration LayerLayer語意層級導覽

專案事件管理專案事件管理專案進度呈報服務專案進度呈報服務

行事曆管理

線上審查線上審查

即時資訊即時資訊管理管理

語意層級搜尋

Page 59: Semantic Web and Knowledge Management

Semantic Web and Knowledge Management 59

客戶客戶

第三方組織第三方組織

下包廠商下包廠商

Team ATeam A

Team CTeam C

Team DTeam D

主管主管 業務業務

後勤後勤

Team BTeam B

Communication LayerCommunication Layer

Communication ServicesCommunication Services

Security ServicesSecurity Services

Security LayerSecurity Layer

DataData IntegrationIntegration ServicesServices

Data Integration LayerData Integration Layer

Info.(Meta Data) Info.(Meta Data) IntegrationIntegration ServicesServices

Info.(Meta Data) Info.(Meta Data) IntegrationIntegration LayerLayer

Knowledge(Info.) Knowledge(Info.) IntegrationIntegration ServicesServices

Knowledge (Info.) Knowledge (Info.) IntegrationIntegration LayerLayer

System Management System Management ServicesServices

System Management System Management LayerLayer

被授權範圍內管理被授權範圍內管理

整體系統管理與管制

Page 60: Semantic Web and Knowledge Management

From Text to RDFFrom Text to RDF

Page 61: Semantic Web and Knowledge Management

Semantic Web and Knowledge Management 61

GoalGoal

• We aim at the automatic creation of metadata from documents for the Semantic Web using a low-cost natural language processing technology, i.e., information extraction.

Page 62: Semantic Web and Knowledge Management

Semantic Web and Knowledge Management 62

System architecture for System architecture for managing the metadata layer of managing the metadata layer of Semantic Web Semantic Web

Contentprovider

Contentprovider

Contentprovider

Contentprovider

Annotator EditorWrapper

InstancesOntology

Inference Engine

Conceptual Search

Social Interactions

Tailored Service

Semantic Navigation

useruser

Back end

Know

ledge-based system

Front end

Information Extractor

Page 63: Semantic Web and Knowledge Management

Semantic Web and Knowledge Management 63

Integrating information Integrating information extraction function with extraction function with Semantic Web system Semantic Web system

Search engine (google.com)

Keywords

Relevant documents

IE engine

Frame Instance Store

RDF Store

RDF store engine

Frame engine

Service front ends

User

Export

Extracted content

Back End extraction component

Service Front End

Ontology-based store

Page 64: Semantic Web and Knowledge Management

Semantic Web and Knowledge Management 64

Components in the information extraction system Components in the information extraction system

Chinese tokenizer

GazetterSentencesplitter

POStagger

Domain eventmatcher

Input documents

Extracted content

Page 65: Semantic Web and Knowledge Management

Semantic Web and Knowledge Management 65

IR vs. IEIR vs. IE

IR

IE

Page 66: Semantic Web and Knowledge Management

Semantic Web and Knowledge Management 66

Extracted Domain Events in RDFExtracted Domain Events in RDF

Page 67: Semantic Web and Knowledge Management

Semantic Web and Knowledge Management 67

Query the Extracted ContentsQuery the Extracted Contents

Page 68: Semantic Web and Knowledge Management

Semantic Web and Knowledge Management 68

A Test of the IE SystemA Test of the IE System

• We take one hundred financial news as our test data and use dozens of JAPE rules to extract the specific domain events.

• After the processing of domain events matching, we then calculate the precision rate and recall rate of our system.

• We first manually extract the target events within these financial news and we obtain 120 events of interest.

• After the domain event matching, it returns 25 results, among which 22 are correct. So the precision rate is 88%, and the recall rate is 18%.

Page 69: Semantic Web and Knowledge Management

Thank you.Thank you.