Developments of the Semantic Web at the Museum of the History of Science - PowerPoint PPT Presentation

About This Presentation
Title:

Developments of the Semantic Web at the Museum of the History of Science

Description:

How we can make an ontology. Mesmuses experience. Beyond Mesmuses ... e.g. Brunelleschi MAKER Dome. Maker. Dome. Brun. Florence, 16 June 2003. 11. Semantic web at IMSS ... – PowerPoint PPT presentation

Number of Views:50
Avg rating:3.0/5.0
Slides: 16
Provided by: marco83
Category:

less

Transcript and Presenter's Notes

Title: Developments of the Semantic Web at the Museum of the History of Science


1
Developments of the Semantic Web at the Museum of
the History of Science
  • Florence, 16 June 2003
  • Marco Berni

2
Main question
  • How can we go ahead on Semantic Web and
    ontologies at IMSS?

3
Try to answer
  • Evaluation of the achieved outcome
  • How we can make an ontology
  • Mesmuses experience
  • Beyond Mesmuses
  • Resources to make the ontology run
  • Critical points
  • Hypothesis on achievable results

4
Evaluation of the achieved outcome
  • Mesmuses prototypes are very useful samples of
    Semantic Web application
  • Allows data browsing with intelligent search
  • Opportunity to make searches not only on words,
    but on the meaning of words too (semantic search)
  • Test semantic search on a history of science
    domain

5
How we can make an ontologyMesmuses experience 1
  • First try
  • Theoretical approach
  • Ontology was designed in order to model domain of
    historic science knowledge
  • No data were available. Need for a totally manual
    data input
  • Schema and instances were very subjective
  • Useful for itineraries, not for search

6
How we can make an ontologyMesmuses experience 2
  • Second try
  • Bottom-up approach
  • Ontology was designed just looking to available
    databases
  • Deep analysis of databases structure in order to
    map relational structure on ontology
    concept/property
  • Data input from relational databases with little
    manual attendance
  • Schema and instances are quite objective, but
    at the moment dont cover all the history of
    science domain
  • Useful for search

7
How we can make an ontologyBeyond Mesmuses 1
IMSS Database
ARAM Multimedia Archive
STS Scientific Instruments
ICONA Ichonographic Archive
IMSS Itineraries
BIBLIO Catalogue of the Library
ARMU Historical archives
GT_at_ GalileoTek_at_
etc.
8
How we can make an ontologyBeyond Mesmuses 2
IMSS Database
ARAM Multimedia Archive
STS Scientific Instruments
DCTitleDCCreator Creator InventorDCSubjectD
CDescriptionDCPublisherDcContributor.PersonalNam
eDcContributor.CorporatelName User Receiver
Giver Purchaser IsDedicated.....
DCrelation IsVersionOf HasVersion
IsReplacedBy IsRequiredBy Requires
IsPartOf HasPart IsReferencedBy References
IsFormatOf HasFormat.....
ICONA Ichonographic Archive
BIBLIO Books Catalogue
ARMU Historical archives
GT_at_ GalileoTek_at_
etc.
9
How we can make an ontologyBeyond Mesmuses 3
  • Scientific Places in Tuscany
  • Old version
  • Plain hypertext
  • Links have not any semantics

10
How we can make an ontologyBeyond Mesmuses 4
  • Scientific Places in Tuscany
  • Forthcoming version
  • Structured hypertext
  • Links have a semantics
  • e.g. Brunelleschi MAKER Dome

Dome
Maker
Brun.
11
How we can make an ontologyBeyond Mesmuses 5
12
How we can make an ontologyBeyond Mesmuses 6
  • Start to re-design database for ontology and
    semantic search
  • Mapping of existing database schemas into classes
    and properties
  • Extended Dublin Core
  • IMSS specifics
  • Partial automatic indexing of data according to
    the mapping rules
  • Manual actions required to complete indexing
    process
  • Design an ontology valid for the whole history of
    science community

13
Resources to make the ontology run
  • Domain experts to map database to ontology
  • Software for data harvesting
  • Operators to manually adjust data
  • Portal
  • Schema / data loader
  • Browsing / searching engine
  • Graphical user interface
  • Set up a community-web and coordinate it
  • Produce (and agree on) a shared ontology
  • Share resources

14
Critical points
  • Ontology production
  • Integration between ontology and itineraries
  • Iterative construction of new ontology to make a
    better coverage of domain and data
  • User interface usability
  • Test of existing interfaces
  • Requirements to improve interaction effectiveness
  • Iterates many times this step
  • Agreement on a community web ontology for history
    of science
  • Impact of a schema modification on existing data
  • Have mapping of data to be redone?

15
Hypothesis on achievable results
  • Strongly improve access to IMSS documents and
    database
  • Provide a framework for history of science
    e-learning
  • Share navigable data with the community
  • Create an access point to the history of science
    resources on the web
  • Create a virtual museum of history of science on
    the web independently from the physical location
    of the objects
Write a Comment
User Comments (0)
About PowerShow.com