• Aucun résultat trouvé

A Survey of the Landscape and State-of-Art in Semantic Wiki

N/A
N/A
Protected

Academic year: 2022

Partager "A Survey of the Landscape and State-of-Art in Semantic Wiki"

Copied!
64
0
0

Texte intégral

(1)

Semantic Wiki Mini-Series 1st session:

A Survey of the Landscape and State-of-Art in Semantic Wiki

Co-chairs:

Sebastian Schaffert (Salzburg Research, Austria), Max Völkel (FZI Karlsruhe)

2008-10-23

Semantic Wikis: The Wiki Way to the Semantic Web?

1

(2)

Agenda

History

State of the Art

Trends Introduction

2

(3)

Introduction

Introduction

3

(4)

Introduction: Semantic Wikis

 wiki principles

 metaweb

 two perspectives on Semantic Wikis

 characteristics of Semantic Wikis

 example

4

Introduction

(5)

Wiki Principles

wikis allow anyone to edit

wikis are easy to use and do not require additional software

wiki content is easy to link

wikis support versioning of all changes

wikis support all media

5

Introduction

(6)

Nova Spivack: Metaweb

6

Semantic Wikis Introduction

(7)

Two Perspectives on Semantic Wikis

 Wikis for Metadata

 Metadata for Wikis

no clear separation, but tendencies!

7

Introduction

(8)

Wikis for Metadata

creating metadata on the Semantic Web is difficult!

– requires domain knowledge

– requires knowledge engineering skills – complicated, insufficient tools

Wikis for metadata:

– simplified technological access to the creation of metadata

– collaboration of domain experts and knowledge engineers

– dynamically evolving knowledge networks and knowledge models

8

Introduction

(9)

Metadata for Wikis

 Wikis

huge amounts of digital content (e.g. Wikipedia)

strong connection of content via hyperlinks

 problem: structure exists, but is only used for presentation and not accessible by computers

finding relevant content is increasingly difficult

integration and exchange between different systems is difficult

9

Introduction

(10)

Semantic Wikis

annotation of existing structures with machine readable metadata

links carry meaning, typing of links, typing of pages

context dependent adaptation and presentation

different domains have different ways of presenting content, personal preferences, etc.

improved, „intelligent“, search and navigation

queries to the structure, visualisation of structure, derived information

improved interoperability between systems

exchange of content, integration of different systems, agents, etc

10

Introduction

(11)

Semantic Wikis: Example

11

Introduction

(12)

History

History

12

(13)

1995: The First Wiki

Wiki

First developed by Ward Cunningham as an add-on to the Portland Pattern Repository on 1995.03.25

http://c2.com/cgi/wiki?WikiDesignPrinciples  our interpretation

Incremental - Pages can cite other pages, including pages that have not been written yet.  network of pages

Organic - The structure and text content of the site is open to editing and evolution.

 different from classical content management systems

Universal - The mechanisms of editing and organizing are the same as those of writing so that any writer is automatically an editor and organizer.  integrated creation and organization of content

Unified - Page names will be drawn from a flat space so that no additional context is required to interpret them.  humans can remember names

Precise - Pages will be titled with sufficient precision to avoid most name clashes, typically by forming noun phrases.  names are quasi-unique

Tolerant - Interpretable (even if undesirable) behavior is preferred to error messages.

 usability: novice users have less fear to start using it

Observable - Activity within the site can be watched and reviewed by any other visitor to the site.  exchange of meta-information

Convergent - Duplication can be discouraged or removed by finding and citing similar or related content. 

History

13

(14)

2004/2005: First Semantic Wikis

Platypus Wiki from Stefano Campanini, Paolo

Castagna, Roberto Tazzoli presented at ISWC2004

WikSAR from David Aumüller wins best Demo award at ESWC2005

History

14

(15)

2005: Wikipedia became popular

Comparing search volume on Google Trends on 2008-10-22

History

15

(16)

2006: Wikis became popular

Comparing search volume on Google Trends on 2008-10-22

History

16

(17)

2006: Semantic Wikis followed the trend

Web Search Volume, Worldwide, 2004 – 2008-10-22, /!\ Scales are different between diagrams!

Wiki

Ontology Semantic web

Semantic wiki

History

17

(18)

2006: Semantic Wiki as a research topic

 2005: [swikig] mailing list launched

 2006: First Workshop on Semantic Wikis: From Wiki to Semantics [SemWiki2006] at ESWC2006, Budva, Montenrego

 2006: Second Workshop on Semantic Wikis: Wiki- based Knowledge-Engineering [WibKe2006] at WikiSym 2006 in Odense, Denmark

 2008: Third Workshop on Semantic Wikis: The Wiki Way of Semantics [SemWiki2008] at ESWC2008,

Tenerife

http://semwiki.org

History

18

(19)

State of the Art

State of the Art

19

(20)

What is a Semantic Wiki? I/II

Semantic Wikis* try to combine the strengths of

Semantic Web

– machine processable, – data integration

– complex queries

Wiki

– easy to use and contribute, – strongly interconnected, – collaborative.

Emergence of Semantic Wikis from to sources:

A) Semantic technologies for wikis („ST4W“) – i.e. better navigation, better queries – Most semantic wiki engines are here

B) Wikis for semantic technologies („W4ST“) – i.e. Ontology engineering, ontology learning – E.g. Many papers on mining wikipedia

State of the Art

* http://Semwiki.org, Schaffert & Völkel, 2006 20

(21)

What is a Semantic Wiki? II/II

A Semantic Wiki is like the Semantic Web in a Petri dish

Many terms emerge – how to consilidate the vocabulary?

Many people work together – how to achieve consensus?

Queries over multiple resources

Import of semantic web data

Export to other semantic web tools

Versioning

Access rights

Trust

...

State of the Art

21

(22)

Semantic Wiki Engines

AceWiki – controlled english

Artificial Memory – personal knowledge management

BOWiki – biomedical domain

Confluence Plugins (Metadata, Scaffolding) - commercial

Hypertext Knowledge Workbench – personal knowledge management

IkeWiki

SWiM - offshoot of IkeWiki

KiWI – successor in scope of KiWi project

OntoWiki – free-form database

OpenRecord – free-form database

SweetWiki – semantic tagging

Semantic MediaWiki (MediaWiki extension) – Semantic Wikipedia

HaloExtension – extension of Semantic MediaWiki, browsing & refactoring

Semantic Forms – free-form database

... Many more Semantic MediaWiki extensions

SWOOKI – a peer-to-peer based SemWiki

State of the Art

http://semanticweb.org/wiki/Semantic_Wiki_State_Of_The_Art 22

(23)

SemWiki2006 Results

How is metadata created?

Incentives for creating formal data

Low in semantic web, higher in semantic wikis with direct benefit

Page vs. Concept

How is metadata used?

Trust - Can trustworthiness of article content be determined from the article metadata?

Navigation - alternative views on the data

Search …

Automated content generation including reasoning

Ontology engineering

Why/for what are Semantic Wikis used?

Like normal wikis, but more sophisticated, doing everything better

Integration

Integartion of structured text and RDF world still unsolved

No common wiki metadata ontology

State of the Art

23

(24)

SemWiki2008

Alexandre Passant and Philippe Laublet.

Towards an Interlinked Semantic Wiki Farm

Christoph Lange.

Mathematical Semantic Markup in a Wiki:

The Roles of Symbols and Notations

Max Völkel.

Hypertext Knowledge Workbench

Andrea Bonomi, Alessandro Mosca, Matteo Palmonari and Giuseppe Vizzari.

Integrating a Wiki in an Ontology Driven Web Site: Approach, Architecture and Application in the Archaeological Domain

Jochen Reutelshoefer, Joachim Baumeister and Frank Puppe.

Ad-Hoc Knowledge Engineering with Semantic Knowledge Wikis

Christoph Lange, Sean McLaughlin and Florian Rabe.

Flyspeck in a Semantic Wiki

Cezary Kaliszyk, Pierre Corbineau, Freek Wiedijk, James McKinna and Herman Geuvers.

A real Semantic Web for mathematics deserves a real semantics

Florian Schmedding, Christoph Hanke and Thomas Hornung.

RDF Authoring in Wikis

Axel Rauschmayer.

Next-Generation Wikis: What Users Expect; How RDF Helps

Malte Kiesel, Sven Schwarz, Ludger van Elst and Georg Buscher.

Using Attention and Context Information for Annotations in a Semantic Wiki

Karsten Dello, Lyndon Nixon and Robert Tolksdorf.

Extending the Makna Semantic Wiki to support workflows

Tobias Kuhn.

AceWiki: Collaborative Ontology Management in Controlled Natural Language

Sau Dan Lee, Patrick Yee, Thomas Lee, David Cheung and Wenjun Yuan.

Descriptive Schema: Semantics-based Query Answering

Markus Luczak-Rösch and Ralf Heese.

A Generic Corporate Ontology Lifecycle

Charbel Rahhal, Hala Skaf-Molli and Pascal Molli.

SWOOKI: A Peer-to-peer Semantic Wiki

Gero Scholz. Semantic MediaWiki with Property Clusters

Joshua Bacher, Robert Hoehndorf and Janet Kelso.

BOWiki: ontology-based semantic wiki with ABox reasoning

State of the Art

 More application oriented than 2006

24

(25)

Semantic Wikis: Trends

Trends

25

(26)

Semantic Wikis: Trends

 Application Areas

what kinds of application areas can be addressed by Semantic Wikis?

 Platform

what kinds of software will Semantic Wikis develop into?

 Technology

what kinds of technological development/improvements will Semantic Wikis see?

Trends

26

(27)

Application Areas

Trends

27

(28)

Application Areas

 Knowledge Management

 Semantic Wikipedia / Semantic Encyclopaedia

 eLearning

 Ontology Engineering

Trends

28

(29)

Knowledge Management

 for me: primary application area

 from “knowledge is power” to “sharing is power”

 supporting the user by semantic technologies

Trends

29

(30)

Knowledge Management: Examples

 connect software documentation (design documents, code documentation) about

components with relevant bug reports and present developer a summary view of his tasks

 allow project managers in consultancies to share project knowledge, e.g. “look for projects that are similar to mine” or “generate instances of all

relevant QM process definitions for my project setup”

Trends

30

(31)

Knowledge Management: Examples

 allow project managers to modify project workplan in different ways, e.g. as a table, as a Gantt

diagram, … with direct connection to ERP system

 allow head of department to get a summary view over all projects

Trends

31

(32)

Knowledge Management: Challenges

 different perspectives on same content

 integration with existing tools (and here the Semantic Web can help)

 requires heavy support for the user, e.g. extensive reasoning, calculation, …

 often very formal environments (contradiction with Wiki Philosophy)

Trends

32

(33)

Semantic Wikipedia

 making the “wisdom of the crowds” in Wikipedia (and similar applications) accessible

 not restricted to Wikipedia, not even to Wikis as technology (see “platform” later)

Trends

33

(34)

Semantic Wikipedia: Challenges

 requires high performance and scalability (i.e. little reasoning)

 community needs to be convinced to make use of semantic features (only if immediate benefit)

Trends

34

(35)

Learning

 ePortfolio systems: collection of learning artefacts, reflection on learning

 collaborative story telling

 personal development planning and alignment with actual achievements

Trends

35

(36)

Learning: Challenges

 requires functionalities current Wikis cannot provide, e.g. collaborative text writing

 require lots of metadata for planning

Trends

36

(37)

Ontology Engineering

 make ontology development simpler

 allow knowledge workers and ontology engineers to collaborate in one system

Trends

37

(38)

Ontology Engineering: Challenges

 allow different perspectives on same content (ontology engineer: ontology view, knowledge worker: domain specific view or wiki view)

 full support for ontologies and reasoning

Trends

38

(39)

Platform

Trends

39

(40)

Semantic Wiki Platform

 wiki as philosophy rather than technology: same principle holds for most other Web 2.0/Social Web applications

 breaking information and system boundaries:

integrating information and giving different perspectives on the same information

 Semantic Wikis as generic platform for developing many different kinds of Social Web applications

Trends

40

(41)

Wiki as Philosophy

wikis allow anyone to edit

wikis are easy to use and do not require additional software

wiki content is easy to link

wikis support versioning of all changes

wikis support all media

same holds for other social software applications!

Trends

41

(42)

Breaking Information and System Boundaries

 integration of different kinds of content in one system (wiki text, photos, code, …)

 different perspectives on the same content (wiki, blog, social network, tagit, …)

 users edit the system behaviour, not only the content (e.g. widgets - zembly, custom layouts, declarative rules)

Trends

42

(43)

Example: Wiki

Trends

http://showcase.kiwi-project.eu 43

(44)

Example: TagIT

Trends

http://showcase.kiwi-project.eu 44

(45)

Example: Blog

 no image (yet) but entries to wiki/tagit could also be displayed in blog style (ordered by creation

time)!

Trends

45

(46)

Example: Social Networking

 user information in the wiki could be used as basis for social networks (e.g. based on tags)

 information represented as foaf data (RDF)

 just another perspective on the same data!

Trends

46

(47)

Example: Community Equity

 Community Equity: valuation system for community content developed by Sun

content can be rated by users -> information equity

tags inherit information equity -> tag equity

users inherit information equity for their content ->

contribution equity

users inherit tag equity for the tags of their content ->

skills equity

Trends

47

(48)

Technology

Trends

48

(49)

Semantic Wikis as Testbed for the Semantic Web

Semantic Wikis connect the real world with the Semantic Web

Semantic Wikis are the “Semantic Web in Small”, because a Wiki is “Web in Small”

Semantic Wikis share many common properties with the Semantic Web

most technologies developed on the Semantic Web can be used and evaluated in Semantic Wikis

(my challenge: if it is not useful in Semantic Wikis, it is not useful at all!)

Trends

49

(50)

Challenge 1: Proof Benefit

 the Semantic Web and Semantic Wikis must show how they are beneficial to ordinary users

Trends

50

(51)

Challenge 2: User Interfaces

 all users like simple interfaces; tools like Protégé are way too complicated

 how to do as much semantics as possible with as little user exposure as possible

Trends

51

(52)

Challenge 3: Personalisation

 semantic data offers the possibility for personalising content presentation

 e.g. preferences, observed behaviour, context

Trends

52

(53)

Challenge 4: Tagging

 users like tagging (various reasons: simplicity, low cognitive barrier, …)

 how to „lift“ non-semantic tags to the Semantic Web?

Trends

53

(54)

Challenge 5: Revisions & Versioning

 essential aspect of the wiki philosophy

 much harder with meta-data than only with textual content

Trends

54

(55)

Challenge 6: Reasoning

 how can reasoning support users?

 what kinds of reasoning are useful in Semantic Wikis (guess: rule-based)?

 how to deal with performance issues (needs to be close to real-time)?

Trends

55

(56)

Challenge 7: Reason Maintenance

 what rules are the justification for a triple?

 how can results of reasoning be explained to users?

 example: background turns purple because a rule says that all pages concerning “foo” should be

rendered as purple; user needs to be able to get an explanation

 example:

Amazon “why was this recommended to me”

Trends

56

(57)

Challenge 8: Permissions, Trust, Provenance

 big outstanding issue of the Semantic Web

 reputation systems can help (e.g. Community Equity by Sun)

 is metadata about metadata

Trends

57

(58)

KiWi – Knowledge in a Wiki Applications

 Software Knowledge Management: Supporting Software Engineers in sharing knowledge (Sun Microsystems)

 Project Knowledge Management: Supporting Project Managers in documenting project

knowledge (Logica)

 KiWi Showcase: “KiWi PhotoStories”, a social

networking and story and image sharing platform

Trends

58

(59)

KiWi – Knowledge in a Wiki Technology

 KiWi addresses personalisation

 KiWi allows arbitrary resources to tag other resources

 KiWi partly addresses reason maintenance

 KiWi addresses rule-based reasoning in Semantic Wikis

 KiWi has a proposal for versioning and transactions (implemented but undocumented)

Trends

59

(60)

KiWi – Knowledge in a Wiki

 website: http://www.kiwi-project.eu

 contact:

Coordinator: Sebastian Schaffert

(sebastian.schaffert@salzburgresearch.at)

Dissemination: Julia Eder

(julia.eder@salzburgresearch.at)

Trends

60

(61)

Semantic Wikis:

The Wiki Way to the Semantic Web?

Semantic Wiki Mini-Series

 1st session:

A Survey of the Landscape and State-of-Art in Semantic Wiki

Co-chairs:

– Sebastian Schaffert (Salzburg Research, Austria) – Max Völkel (AIFB-Karlsruhe)

 Thanks for listening!

61

(62)

Semantic Wiki Mini Series Plan & Dates

Session 2 scheduled on 20th November 2008 Semantic Wiki Technology (1):

An introduction to some of the Semantic Wiki Engines

Chair?

Panelists (tentative): MarkusKrotzsch and/or DennyVrendecic;

SebastianSchaffert; TobiasKuhn; MartinHepp; ...(?)

Engines (tentative): Semantic MediaWiki, IkeWiki, AceWiki, OntoWiki, ...(?)

Session 3 scheduled on 11th December 2008 Semantic Wiki Technology (2):

Semantic Wiki Extension, Add-on's and other Enhancements

Chair?

Panelists (tentative): YaronKoren; MarkGreaves and/or Thomas Schweitzer(?);

JieBao and/or LiDing; PeterYim and/or KenBaclawski; HaroldSolbrig(?), ...(?)

Engines (tentative): Semantic Forms, SMWHalo extension, blog, purple number tag (PMWX), Lex Wiki extension(?), ...

62

Planning

(63)

Semantic Wiki Mini Series Plan & Dates

Session 4 scheduled on 22th Januar 2009

Semantic Wiki Applications & Use Cases (1):

vertical applications

Panelists: HaroldSolbrig; ...; ChristophLange; MarkGreaves; ...(?)

Topics: Applications in Healthcare and Life Science, e-Science,

Mathematics, AI, Education, ... – panelists to brief the participants on the "what," "why" and "how" of their semantic wiki

project/implementations

Session 5 on Februar 2009

Semantic Wiki Applications & Use Cases (2):

horizontal applications

Panelists: SebastianSchaffert and/or PeterDolog; ...; PeterYim;

MikeDean; ...(?)

Topics: applications in Knowledge Management, software

engineering, collaboration and community support, open ontology repository, ... - panelists to brief the participants on the "what,"

"why" and "how" of their semantic wiki project/implementations

63

Planning

(64)

Semantic Wiki Mini Series Plan & Dates

 Session 6 in March 2009

The Future of Semantic Wiki:

Trends, Challenges and Outlook (Panel Discussion)

Co-chair: candidates - DeborahMcGuinness, RudiStuder, MarkMusen

Panelists: hopefully, all panelists from previous session can join us in this discussion and to answer questions as well

looking for as many panelists as we can, 5-minute briefs from each, and an extensive moderated discussion

segment

issues relating to scope, KR, Reasoning, HCI, access control, adoption, ...

Planning

Références

Documents relatifs

In the context of the project “CliWE” we used a prototype of the clinical wiki environment for the development of a guideline covering the diagnosis and therapy of sepsis.. Sepsis is

Since each concept on the SABOK has several digital assets associated (the same used to build the ontology), it can be used as a digital asset search tool as well.. The ontology

In order to make the ontology expressive enough, we have defined some sub classes of the article class (i.e., the Tourist Service one). They are in much cases defined with

Semantic wiki refactorings are transformations in the semantic structure of a se- mantic wiki knowledge base, that allow to improve the implicit ontology quality.. Refactorings

In addition, we have incorporated tools such as Semantic Forms, which allow the ontology classes to be edited as a form rather than as a free form wiki page

The rationale for developing the KIWI system based on a semantic wiki is that – as demonstrated in our two use cases – organisations in knowledge-intensive areas are nowadays more

For this the outline of the whole proof from the book [10] needs to be represented in the wiki, where the mathematical statements (including definitions, lemmas, and theorems)

The author of [12] presents a solution based on OWL Wiki Forms (a Semantic MediaWiki extensions that map Semantic Web ontologies to a Semantic Forms-based semantic wiki) and Fresnel