The MONDIAL Database

F-Logic Oracle XML RDF/OWL Datalog

The MONDIAL database has been compiled from geographical Web data sources listed below:

CIA World Factbook,
a predecessor of Global Statistics which has been collected by Johan van der Heijden.
additional textual sources for coordinates,
the International Atlas by Kümmerly & Frey, Rand McNally, and Westermann,
and some geographical data of the Karlsruhe TERRA database.

The generation of the MONDIAL database is a case study for information extraction and integration:

The data extraction and integration process using the F-Logic system FLORID is described here.
The integration in XML using the LoPiX system is described here.

The Mondial database is available in several formats:

MONDIAL in F-Logic

Sources in F-Logic Representation
- Data from Global Statistics "mondial-gs-facts.flp"
- Country Data from CIA World Factbook (1996) "mondial-cia-facts.flp"
- Organizations Data from CIA World Factbook (1996) "mondial-orgs-facts.flp"
- Data from qiblih (Geographical Coordinates) "mondial-qiblih-facts.flp"
- Country Codes etc. "mondial-codes-facts.flp"
- the original TERRA database "mondial-terra-facts.flp"

Integrated Database
- Cleaned data in an OO schema "mondial-oo-facts.flp"
- Cleaned data in Florid Frame Format "mondial-frames.flp"

Relational MONDIAL (Datalog, Oracle)

The Database training "Praktikum: Datenbankprogrammierung in SQL/Oracle" at the IFI uses the relational version of the MONDIAL database:

Relational data (Datalog) "mondial-rel-facts.flp"

Generating the Database under Oracle
- Database schema "mondial-schema.sql"
- Input statements "mondial-inputs.sql"

Generating the Database under MySQL
- Database schema for mysql "mondial-schema-mysql.sql"
- Input statements for mysql "mondial-inputs-mysql.sql"
- short description what has been changed for MySQL.

Documentation
- the ER diagram,
- the relational schema,
- the referential dependency diagram,

The Oracle instance of Mondial can be queried by that form.

MONDIAL in XML

The Mondial database provides a comprehensive example for XML, e.g., for use in teaching.

The individual sources in XML format
- CIA Factbook: Country Data cia.xml cia.dtd
- CIA Factbook: Organizations orgs.xml orgs.dtd
- Global Statistics: Countries, Cities and Provinces gs.xml gs.dtd
- Cities: Geographic Coordinates qiblih.xml qiblih.dtd
- the Original TERRA database terra.xml terra.dtd
- Country codes and names codes.xml codes.dtd

The MONDIAL case study on information integration in XML continues here.

A flat XML model
- flat DTD "mondial-flat.dtd"
- flat XML file "mondial-flat.xml"
  - excerpt: europe "mondial-europe-flat.xml"

A more sophisticated hierarchical XML model
- scalar and multivalued reference attributes (city/@capital; country/@memberships)
- cross-references in both directions (organization/member/@country, country/@memberships)
- a "boolean"/flag attribute: city/@is_country_cap
- reference attributes with more than one target class (river/to, references rivers, lakes, and seas)
- XSL(T) transformation script "mondial.xsl" (needs mondial-flat.xml)
- hierarchical DTD "mondial.dtd"
- XML Schema "mondial.xsd"
- hierarchical XML file "mondial.xml"
  - excerpt: europe "mondial-europe.xml"
(note that some browsers to not *show* XML and DTD files correctly. Download the file(s) and load them into an editor)

A version of Mondial that is distributed over several files, using XLink
- A file that contains all continents: continents.xml, continents.dtd
- A file that contains all countries: countries.xml, countries.dtd
- for each country, a file that contains its provinces: e.g., provs-D.xml, provinces.dtd
- for each country, a file that contains its cities: e.g., cities-D.xml, e.g., cities.dtd
- A file that contains all organizations: organizations.xml, organizations.dtd
- A file that contains all rivers, seas, lakes, mountains, islands, and deserts: geo.xml, geo.dtd
- A file that contains all memberships (as arcs): memberships.xml, memberships.dtd
- A file that contains all schema information for the distributed mondial files: mondial-distributed.xsd
- The XSLT script that generates the splitted version from mondial: split-dbxlink-mondial.xsl
- The completed distributed database in compressed form: Mondial-Distributed.zip

xml2tex.pl: A perl script for producing a LaTeX documentation from XML/XSL files.

MONDIAL in RDF

Files are available in N3 format and in RDF/XML format. For human readers, the N3 is better readable, but as different tools have different expectations what "valid" N3 syntax is, there is also the RDF/XML variant [in the browser, use "show source"].
(subject to change - minor change in namespace 5.6.2008)

The complete Mondial database: mondial.n3 (N3/Turtle), mondial.rdf (RDF/XML)
The N3 data for Europe only: mondial-europe.n3 (N3/Turtle), mondial-europe.rdf (RDF/XML)
Mondial RDFS/OWL metadata mondial-meta.n3 (N3/Turtle), mondial-meta.rdf (RDF/XML)
an example query in SPARQL: mondial-query.sparql

The MONDIAL database has been stimulated by the TERRA database and the SQL training of the Institut für Programmstrukturen und Datenorganisation der Universität Karlsruhe. Using Florid, a new database has been generated in 1998 from the above data sources.
MONDIAL and the DBIS Oracle training is freely available for research and educational purposes under the condition that the origin of MONDIAL is mentioned in all publications and documentation.
Recommended citation is the technical report with a reference to the Web page:

@TechReport{may-MONDIAL-report-99,
  author =	 {Wolfgang May},
  title =	 {Information Extraction and Integration with \textsc{Florid}:
        The \textsc{Mondial} Case Study},
  institution =  {Universit\"at Freiburg, Institut f\"ur Informatik},
  year = 	 1999,
  number =	 131,
  note    =     "Available from
    \url{http://dbis.informatik.uni-goettingen.de/Mondial}"
}

Mondial is used as an example in several books, lectures, courses etc. Search e.g. google for