Aplica?ii web semantice

1
Semantic Web Applications
Introduction Vlad Posea vladposea_at_yahoo.com
2

• http//www.ted.com/talks/tim_berners_lee_on_the_ne
  xt_web.html

3
On the next web

• Web 1.0 a way to reframe the way we use
  information
• Overcome the need for systems to be interoperable
  in order for people to better access each others
  documents
• gt TBL was frustrated ?
• gt he built a system to overcome that
• Great idea linking documents - hypertext

4
Which is the problem that still frustrates people
and TBL?

• Turns out that there is still huge unlocked
  potential. There is still a huge frustration that
  people have because we haven't got data on the
  web as data.
• Which is the difference between document and
  data?
• On the web we have links between documents not
  between data sets

5
Web evolution

• Web today
• So, imagine that that link could have gone to
  virtually any document you could imagine.
• Web of documents

• Web tomorrow
• but I want to think about a world where
  everybody has put data on the web and so
  virtually everything you can imagine is on the
  web. and then calling that linked data.
• Web of data

6
how do we get to the next web?

•  those things that start with http --we're using
  them not just for documents now,we're using them
  for things that the documents are about. 

• Semantic web
• URIs used to identify data (resources)

• Web 1.0
• URLs used to identify documents

7
url that identifies a data resource
relations
8
how do we get to the next web?

• TBLs rules
• All kinds of conceptual things, they have names
  now that start with HTTP.
• I get important information back about the data
  Im looking up using a standard format
• when I get back that information it's got
  relationships

9
How do we create the semantic web

• write a program to take the data, extract it from
  Wikipedia, and put it into a blob of linked
  data on the web, which he called dbpedia.

10
(No Transcript)
11
How do we create the semantic web

• write a program to take the data, extract it from
  Wikipedia, and put it into a blob of linked
  data on the web, which he called dbpedia.
• American government data would be available on
  the Internet in accessible formats (englands
  data already is)

12
(No Transcript)
13
How do we create the semantic web

• write a program to take the data, extract it from
  Wikipedia, and put it into a blob of linked
  data on the web, which he called dbpedia.
• American government data would be available on
  the Internet in accessible formats (englands
  data already is)
• data is about our lives. You just -- you log on
  to your social networking site, your favorite
  one, you say, "This is my friend." Bing!
  Relationship. Data. You say, "This photograph,
  it's about -- it depicts this person. "

14
(No Transcript)
15
How do we create the semantic data?

• Extract data from web sites and publish it in
  semantic formats
• When you develop a web site make sure the data is
  also available in semantic formats
• Fetch semantic data from social web sites

16
What is it good for?

• Answer questions based on relations between
  different data concepts
• What proteins are involved in signal
  transduction and also related to pyramidal
  neurons?
• All soccer players, who played as goalkeeper
  for a club that has a stadium with more than
  40.000 seats and who are born in a country with
  more than 10 million inhabitants (dbpedia
  example)
• Query the web like you would query a database!

17
Web 2.0 Semantic Web

• gt Social Semantic Web
•  social semantic web is all about everybody doing
  their bit and it creates an incredible
  resource because everybody else does theirs. And
  that is what linked data is all about. It's about
  people doing their bit to produce a little bit,
  and it all connecting. 

18
What I want you to do

• Create linked data and make it available on the
  web
• Choose a domain that youre interested in
• Make an ontology of the domain (linking it to
  existing ontologies)
• Extract data from the web and create a semantic
  repository
• Make the data available for search by other
  applications
• Create a web application that uses the semantic
  data that you have

19
What others want you to do

• http//challenge.semanticweb.org
• Challenge at the biggest international conference
• The requirements represent practically the
  guidelines for the development of the Semantic
  Web
• Minimal requirements also needed for our
  project
• Additional Desirable Features what you will
  actually need for a commercial product.

20
Minimal requirements

• The application has to be an end-user application
  
• The information sources used
• should be under diverse ownership or control
• should be heterogeneous (syntactically,
  structurally, and semantically)
• should contain substantial quantities of real
  world data
• The meaning of data has to play a central role.
• Meaning must be represented using Semantic Web
  technologies.
• Data must be manipulated/processed in interesting
  ways to derive useful information and
• this semantic information processing has to play
  a central role in achieving things that
  alternative technologies cannot do as well, or at
  all

21
Additional Desirable Features

• The application provides an attractive and
  functional Web interface (for human users)
• The application should be scalable (in terms of
  the amount of data used and in terms of
  distributed components working together).
  Ideally, the application should use all data that
  is currently published on the Semantic Web.
• Rigorous evaluations have taken place that
  demonstrate the benefits of semantic
  technologies, or validate the results obtained.
• Novelty, in applying semantic technology to a
  domain or task that have not been considered
  before
• Italicinteresting from the research point of
  view
• Bold interesting from the commercial point of
  view

22
Additional Desirable Features

• Functionality is different from or goes beyond
  pure information retrieval
• The application has clear commercial potential
  and/or large existing user base
• Contextual information is used for ratings or
  rankings
• Multimedia documents are used in some way
• There is a use of dynamic data (e.g. workflows),
  perhaps in combination with static information
• The results should be as accurate as possible
  (e.g. use a ranking of results according to
  context)

23
Examples Semantic Web Challenge winners - 2010

• NCBO Resource Index Ontology-Based Search and
  Mining of Biomedical Resources winner 2010
• http//www.cs.vu.nl/pmika/swc/submissions/swc2010
  _submission_4.pdf
• Uses ontologies on BioPortal to annotate
  resources with concepts from the ontologies
• Semantics used in finding synonyms, autocomplete
  on search, identify hierarchies of concepts

24
Examples Semantic Web Challenge winners - 2010

• Linking Open Government Data http//logd.tw.rpi.ed
  u/
• http//www.cs.vu.nl/pmika/swc/submissions/swc2010
  _submission_16.pdf
• The TWC LOGD Portal is a semantic web application
  dedicated to publishing Linked Data versions of
  OGD and sharing tools, services and expertise
  supporting an OGD ecosystem. It serves data users
  ranging from informed citizens, to domain
  experts, to developers consuming and creating
  novel applications enriched by government data.

25
Examples Semantic Web Challenge winners - 2010

• shortipedia.org
• http//www.cs.vu.nl/pmika/swc/submissions/swc2010
  _submission_18.pdf
• Aggregating semantic data curating by volunteer
  users (just like wikipedia)

26
Examples Semantic Web Challenge winners - 2009

• http//www.cs.vu.nl/pmika/swc/documents/TrialX-he
  althx-iswc09-challenge.pdf - matching patients
  with clinical trials based on patient records
• http//www.cs.vu.nl/pmika/swc/documents/VisiNav-p
  aper.pdf Exploration of web datasets
• http//www.cs.vu.nl/pmika/swc/documents/Sig.ma2
  0Live20views20on20the20web20of20data-sigma.p
  df Visualisation of the web of data

27
Examples Semantic Web Challenge winners - 2008

• paggr.com creates mashups from semantic data
• DBPediaMobile integrates DBPedia data with
  Google Maps
• HealthFinland - single national entry-point for
  health information, health promotion and
  health-related news

28
Evolution of the domain
2008
2009
2010
Output data in semantic formats Aggregate data on
a small scale Visualize data Develop tools
Develop tools (visualisation) Aggregating data on
larger scale from a small number of sources
Aggregate data on a large scale Complex
applications that take advantage of all the
existing data and tools
Evolution from development of tools, small
applications, output of semantic data from a
single domain to complex applications aggregating
large number of data sources providing complex
services
29
Example

• Domain Romanian Tourism
• Ontology concepts Hotel, Resort, Activity, City,
  Camera,
• Data sources tourism web sites
• Semantic interogations
• Hotel at least 2 stars where you can practice
  archery
• Hostel in the Apuseni mountains with a place
  where the children can play
• Hotel that the girlfriend would like)

30
Nume
Rating (nr stele)
Contact foafPerson
Facilita?i
31
How do you gain points

• 5 points exam
• 5 points project described in the previous
  slides

32
http//www.ted.com/talks/tim_berners_lee_the_year_
open_data_went_worldwide.html
33
References

• http//www.scientificamerican.com/article.cfm?idt
  he-semantic-web
• http//www.ted.com/talks/tim_berners_lee_on_the_ne
  xt_web.html