LarKC WP4 Teleconference, 11.00 -12.00, July 10, 2009
The skype ID of the chair: larkc-wp4
Participants
Barry (STI), Yi (WICI), Yan (WICI), Hans (MPG), Vassil (?OntoText), Zhisheng (VUA), Emanuele (CEFRIEL), Axel (HLRS), Alex (HLRS).
Agenda
- Month18 Deliverable D4.7.1 Initial Evaluation and Revision of Plug-ins deployed in use-cases
- Month18 Deliverable D4.3.1 strategies and Design for Interleaving Reasoning and Selection of axioms
- Month18 Deliverable D4.2.2 Analysis of Human Search Strategies
D4.7.1: Initial Evaluation & Revision of plug-ins deployed in use-cases
due M18
The latex files of D4.7.1 are available at: http://svn.larkc.eu/wp4/larkc471/
Contributions by each partner and the status
- UIBK: contribute a chapter on evaluation of Rule-based reasoning, latex file: uibk.tex
- HLRS: Evaluation of parellel reasoning, latex file: hlrs.tex
- VUA: Contribute chapter on framework of evaluation and benchmarking of reasoners, latex file: framework.tex
- Ontotext: Evaluate the use-case WP7a/b req. for the plugins, latex file: ontotext.tex
- WICI: Evaluation of granular reasoner, latex file: wici.tex
- MPG: contribute some work which is based on the work for WP1 and WP2, latex file: mpg.tex
- CEFRIEL: contribute a chapter on the evaluation measures from the perspective of the case study of Urban computing, latex file: cefriel.tex
Some issues/questions on the deliverable
- dataset from the use cases for the evaluation and benchmarking
- how to make the evaluation: manual, semi-automatic,or automatic
- gold standards for the evaluation
- how to compare different reasoners
- check missing part of each chapter
- Chapter to describe the general picture of LarKC reasoners
- integration of chapters
check them with the following initial goals of the evaluation:
- Bug detection
- Robustness
- Performance analysis
- Scalability Potential
- Platform Improvement
suggested benchmarks:
- Lehigh University Benchmark (LUBM)
- IBM China Benchmark (IBMChina)
suggested dataset:
- Geonames (motivated from the use case of Urban Computing)
?PubMed (motivated from the use case of WP7a and WP7b)
Timeline of Deliverable 4.7.1:
July 1, 2009, Second Draft (passed)
July 10, 2009, telecon on the integration issue
July 25,2009, missing part due
August 1-30, 2009, Final Integration
Sept 1-9, 2009, Final Revision
Sept 10-17, 2009, Deliverable Review
Sept 18-24, 2009, Quality Control
Sept 30, 2009, Submit to the Commission.
D4.3.1: Strategies & Design for interleaving reasoning and selection of axioms
due M18
The latex files of D4.3.1 are available at: http://svn.larkc.eu/wp4/larkc431/
Contributions by each partner and the status
1 Introduction (2nd draft is done, VUA)
2 A framework of Web scale reasoning by interleaving reasoning and selection (2nd draft is done, VUA)
3 Stefan’s approach to approximate reasoning (1st draft is done, VUA)
4 Granular reasoning for the semantic Web ( 1st draft is done, WICI)
5 Selection Strategies for LarKC Reasoner Plug-ins (MPG)
6 Conclusion
Timeline of Deliverable 4.3.1:
July 1, 2009, Second Draft (passed)
July 10, 2009, telecon on the integration issue
July 25,2009, missing part due
August 1-30, 2009, Final Integration
Sept 1-9, 2009, Final Revision
Sept 10-17, 2009, Deliverable Review by Gulay Unel
Sept 18-24, 2009, Quality Control
Sept 30, 2009, Submit to the Commission.
D4.2.2 Analysis of Human Search Strategies
due M18
MPG is currently (July 09) analyzing the results of the first experiments with human participants.
MPG and WICI need to coordinate their respective contents and derive implications for the design of LarKC plugins.
Ask Gaston (VUA) if he wants to contribute to this deliverable.