Experiment 20140131-NOOR¶

This is the first experiment that was ever run. No automation was in place, no result files were preserved.

Hence, this notebook, in Sake.re style, has been regenerated from the initial notebook, but does not compute anything.

Experiment design¶

Date: 2014-01-31

Hypotheses: Experimenting 3 agents performing revision of a network of ontology.

Experimental setting: The ontologies are generated from a subset of feature space. The alignments are randomly generated with 4 equivalence correspondences.

Variables¶

controled variables: []

dependent variables: ['srate', 'inc', 'fmeas', 'conv']

Experiment¶

Date: 2014-01-31

Performer: Jérôme Euzenat (INRIA)

Lazy lavender hash: ea0c5b9dac885d541fc0bd86a7a9f007f2025949

Classpath: lib/lazylav/ll.jar:lib/slf4j/logback-classic-1.2.3.jar:lib/slf4j/logback-core-1.2.3.jar:.

Parameter file: params.sh This parameter file has been generated post hoc. However, it is used for generating the skeleton of this notebook.

Executed command: There is no script.sh available for this test This was likely java -cp lib/lazylav/ll.jar fr.inria.exmo.lazylavender.engine.Monitor

Class used: NOOEnvironment, AlignmentAdjustingAgent, AlignmentRevisionExperiment.

Execution environment: MacBook Pro 15, C2D 2.66GHz, 8GB - MacOS X 6.8 - Java 1.6.0 HotSpot

Raw results¶

Raw results are not available for this experiment.

Result exploration¶

Here is the first generated plot of 5 runs:

• Each color corresponds to one of the 5 runs.

Analysis¶

This is the first ever working experiment with Lazy lavender, the week of its creation.

It shows that the system seems to work and that the population of agents, playing game at random and starting with random alignments, converge quickly to a stable state.

Moreover the success rates seem very similar.