Context: This tests the influence of the number of initial correspondences on the results
Hypothesis: Increasing the number of initial correspondences improves the performance of agents (because it leaves them more chances to find correct correspondences)
Experimental setting: Same as 20140305-NOOR with
Experimenter: Jérôme Euzenat (INRIA)
Date: 2014-04-07
Lazy lavender hash: 89f0656ffc28a6c4560a7a55cf1508d866528a96
Parameters: params.sh
Command line (script.sh):
java -Xms500M -Xmx1G -cp lib/lazylav/ll.jar:lib/slf4j/logback-classic-1.0.9.jar:lib/slf4j/logback-core-1.0.9.jar:. fr.inria.exmo.lazylavender.engine.Monitor -DrevisionModality=add -DnbRuns=10 -DnbAgents=4 -DnbIterations=2000 -DnbCorresp=6 > /tmp/20140407-add4-2000-6.txt
Class used: NOOEnvironment, AlignmentAdjustingAgent, AlignmentRevisionExperiment, ActionLogger, AverageLogger, Monitor.
Execution environment: MacBook Pro 15, C2D 2.66GHz, 8GB - MacOS X 6.8 - Java 1.6.0 HotSpot
20140407-NOOR-add4-2000-10.tsv 20140407-NOOR-add4-2000-10.txt 20140407-NOOR-add4-2000-11.tsv 20140407-NOOR-add4-2000-11.txt 20140407-NOOR-add4-2000-12.tsv 20140407-NOOR-add4-2000-12.txt 20140407-NOOR-add4-2000-6.tsv 20140407-NOOR-add4-2000-6.txt 20140407-NOOR-add4-2000-8.tsv 20140407-NOOR-add4-2000-8.txt
nbCorresp | size | success | Incoherence degree | F-measure | Alcomo | LogMap |
---|---|---|---|---|---|---|
6 | 42 | 91% | 0.00 | 0.2 / 11 | 0.27 / 24 | 0.26 / 31 |
8 | 54 | 89% | 0.00 | 0.23 / 16 | 0.25 / 29 | 0.25 / 34 |
10 | 66 | 86% | 0.00 | 0.23 / 17 | 0.24 / 32 | 0.24 / 43 |
11 | 72 | 84% | 0.00 | 0.23 / 17 | 0.23 / 34 | 0.23 / 43 |
12 | 78 | 83% | 0.00 | 0.25 / 21 | 0.23 / 35 | 0.25 / 47 |
The size of the reference alignment is 70 (globally, which sets it between 10 and 12 initial correspondences. After 12, it is not possible to generate more correspondences without being redundant (this is forbidden by the program). The test performed before (eg 20140305-NOOR), are with 54 correspondences, hence nbCorresp=8.
Key points: The increase of the number of correspondences augments the chances to retain a correct correspondences and thus increases the quality of alignments provided by agents.
This is also measured with respect to LogMap and Alcomo which have, on their side, more chances to retain an incorrect correspondence. So their F-measure decreases while that of the agents increases.
This file can be retrieved from URL https://sake.re/20140407-NOOR
It is possible to check out the repository by cloning https://felapton.inrialpes.fr/cakes/20140407-NOOR.git
This experiment has been transferred from its initial location at https://gforge.inria.fr (not available any more)