To Do

Take a look at the timeline for George Washington.

Problem:

• There are too many items at same year, therefore all of the items cannot be displayed in the provided timeline height.
• Its titles are cropped.

Todo:
Solve it :-)

10.00 - 10.45 a.m.
Attended by: Gosse, Proscovia, Junte, and Ismail
Henk (go to Utrecht)

Junte

Progress:

• Ontology version 0.1 is finished (no major work will be carried on).
• The Sesame server is installed.

Todo:

• Use the 'person' type of the SWHi mockup as the use case to evaluate the ontology.
• Create SeRQL to populate person's information from the ontology.
• Write 'person' module using PHP.

Target:

• At the end of May (or June?) a person detail demo is delivered.

Proscovia

Progress:

• Build a corpus (a subset of the Wikipedia text) related to the domain.
  

This note reports things to do in the next two weeks.

Proscovia

• Extract events from Wikipedia XML (source: Gosse)
• Annotate Wikipedia cleaned text using GATE, and extract simple relations such as: involved in, located, birth, death..

Junte

• Consider using the SKOS as our subject classification schema.
• Classify each imprint to related subjects.
• Evaluate the performance of the ontology through Protege and/or Sesame queries.

Ismail

• Get the timeline of the Mockup linked to the ontology.
• Continue designing class diagrams (of the UML). See version 0.1
  

Meeting Notes

9 March 2006

Next meeting: Thursday 10 a.m.

Ismail:

• Mockup: use simple interface at the front, hide details for help menu.
• Continue with other mockup pages.
• Extract entity relationships
• Add into ontology

Junte:

• Add instances for: Identifier, topical, medium, publisher, language
• Verification of metadata accuracy is not necessary. Rely on historian and librarian who typed them.

Proscovia:

• Extract relationships between entities within sentences and paragraphs
• Evaluate the accuracy
  

Short reports

Meeting date: 2-3-2006

Junte

• Extract locations of publication
• Add imprints class and its sub classes to the vicodi ontology
• Extract person
• Extract time interval

Proscovia

• Annotate location semiautomatically
• Identify events
• Clean up the text

Ismail

• Install CVS
• Study web usability
• Draft HCI (mockup)

Todo for the next week

Junte

• continue populating instances from metadata
• Verify person and time interval using external knowledge (wikipedia)
  

Junte and Proscovia

Goal:

• Create simple (initial) ontology containing one class from Metadata (e.g. Person) and one class from Text (e.g. Location). Use Protege for this task.
• Make one or two competency questions on what answer the ontology will give.
• Extract few instances from Metadata and Text manually or automatically. Use Perl script and Gate for this task.
• Import the instances into the ontology.
• Evaluate the ontology and intances with the competency queries.

Ismail Fahmi

Goal:

• Install Sesame and Tomcat.
  Installed on /opt/apache-tomcat-5.5.12

Feb 6-14, 2006

Junte

• Studying previous works on the ontology development
• Preparing a piece of discussion note for next week meeting (Wednesday?)

Proscovia

• Studying previous works on IE (information extraction) from text
• Exploring GATE
• Preparing a piece of discussion note for next meeting

Ismail

• Finishing system requirements

Things to do in October 2005:

• Installing a Linux server for experiment.
  
• Install Apache Tomcat
  
• Install Sesame
  
• Give presentation in the Advanced Webtechnology class of the Alfa-Informatica RUG.

On these days, the following tasks were carried:

• Create a Perl script to convert MARC XML into MARC RDF of the Evans metadata. Done: see here
  
• Evaluate the resulted RDF using the Welkins tool. So far the graph is good. We can see the relation between the nodes/elements.
  
• Select an RDF framework from these suites/tools: KAON, Sesame, Protege. For reasoning (queries) we prefer using the Sesame.
  
• Install Apache Tomcat
  
• Install Sesame

The installation cannot be done on my Windows PC. The Apache Tomcat setup is stopped in the middle of the installation.

After installing the Drupal, the following are things to do this week:

• Learn how to enrich and refine the Evans Metadata (RDF)
  
• Read papers about historical semantic web.
  
• Investigate the Perseus Digital library, as the most successful digital library in humanities domain.
  
• Investigate how people use the Evans resources. Sources: Readex, a nice article about user's experience