20180915-NOOR

Experiment design

Designer: Jérôme Euzenat (INRIA) (2018-08-31)

Revision of networks of ontologies with expansion, relaxation and generation starting with realistic alignment size (4 agents; 10 runs; 10000 games; delete/replace/refine/add/addjoin/refadd)

Hypotheses: Starting with realistic alignment size should get closer results to starting empty than non realistic size (expansion dominance)

Experimental setting: Same as 20180601-NOOR replaying the same runs as 20180308-NOOR with realistic alignment size.

Experiment

Experimenter: Jérôme Euzenat (INRIA) (2018-09-15)

Date: 2018-09-15

Lazy lavender hash: 759ff097b96520c12aa84f3749927f9a22022e62

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

OS: stretch

Variation of: 20180601-NOOR

Parameters: params.sh

Command line (script.sh):

. params.sh

for op in delete replace refine add addjoin refadd
do
  bash scripts/runexp.sh -p ${OUTPUT} -d ${DIRPREF}-${op}-clever-nr-${POSTFIX} java -Dlog.level=INFO -cp ${CLASSPATH} fr.inria.exmo.lazylavender.engine.Monitor ${OPT} ${LOADOPT} -DrevisionModality=${op} -DexpandAlignments=clever -DnonRedundancy ${PARAMS}
  bash scripts/runexp.sh -p ${OUTPUT} -d ${DIRPREF}-${op}-clever-nr-im80-${POSTFIX} java -Dlog.level=INFO -cp ${CLASSPATH} fr.inria.exmo.lazylavender.engine.Monitor ${OPT} ${LOADOPT} -DrevisionModality=${op} -DexpandAlignments=clever -DnonRedundancy -DimmediateRatio=80 ${PARAMS}
done

Class used: NOOEnvironment, AlignmentAdjustingAgent, AlignmentRevisionExperiment, ActionLogger, AverageLogger, Monitor

Execution environment: 24 * Intel(R) Xeon(R) CPU E5-2420 0 @ 1.90GHz with 20GB RAM / Linux ProxMox 2 / Linux 4.15.17-1-pve / Java Java(TM) SE Runtime Environment 1.8.0_151 with 4.33G max heap size

Raw results

4-10000-add-clever-nr-gen-real.tsv 4-10000-add-clever-nr-gen-real.txt 4-10000-add-clever-nr-im80-gen-real.tsv 4-10000-add-clever-nr-im80-gen-real.txt 4-10000-addjoin-clever-nr-gen-real.tsv 4-10000-addjoin-clever-nr-gen-real.txt 4-10000-addjoin-clever-nr-im80-gen-real.tsv 4-10000-addjoin-clever-nr-im80-gen-real.txt 4-10000-delete-clever-nr-gen-real.tsv 4-10000-delete-clever-nr-gen-real.txt 4-10000-delete-clever-nr-im80-gen-real.tsv 4-10000-delete-clever-nr-im80-gen-real.txt 4-10000-refadd-clever-nr-gen-real.tsv 4-10000-refadd-clever-nr-gen-real.txt 4-10000-refadd-clever-nr-im80-gen-real.tsv 4-10000-refadd-clever-nr-im80-gen-real.txt 4-10000-refine-clever-nr-gen-real.tsv 4-10000-refine-clever-nr-gen-real.txt 4-10000-refine-clever-nr-im80-gen-real.tsv 4-10000-refine-clever-nr-im80-gen-real.txt 4-10000-replace-clever-nr-gen-real.tsv 4-10000-replace-clever-nr-gen-real.txt 4-10000-replace-clever-nr-im80-gen-real.tsv 4-10000-replace-clever-nr-im80-gen-real.txt

Result exploration

The first plot compare the size of four versions: generative+realistic (plain), realistic (dotted), generative (dashed), empty+generative (dashdotted):

The other plots compare simply generative+realistic (plain) to realistic (dotted), generative (dashed):

test success
rate
network
size
incoherence
degree
semantic
precision
semantic
F-measure
semantic
recall
maximum
convergence
4-10000-delete-clever-nr-gen-real 0.95 71 0.17 0.80 0.70 0.62 8575
4-10000-replace-clever-nr-gen-real 0.94 70 0.15 0.81 0.69 0.61 8988
4-10000-refine-clever-nr-gen-real 0.94 79 0.17 0.80 0.76 0.72 6254
4-10000-add-clever-nr-gen-real 0.93 91 0.25 0.72 0.79 0.88 7316
4-10000-addjoin-clever-nr-gen-real 0.96 92 0.25 0.72 0.79 0.88 4160
4-10000-refadd-clever-nr-gen-real 0.95 93 0.25 0.71 0.79 0.89 5743
4-10000-delete-clever-nr-im80-gen-real 0.94 61 0.00 1.00 0.63 0.46 8222
4-10000-replace-clever-nr-im80-gen-real 0.94 60 0.00 1.00 0.61 0.44 8098
4-10000-refine-clever-nr-im80-gen-real 0.93 70 0.00 1.00 0.74 0.59 8527
4-10000-add-clever-nr-im80-gen-real 0.92 79 0.00 1.00 0.79 0.66 8380
4-10000-addjoin-clever-nr-im80-gen-real 0.95 79 0.00 1.00 0.81 0.68 7068
4-10000-refadd-clever-nr-im80-gen-real 0.94 85 0.00 1.00 0.85 0.74 8585

Analysis

Analyst: Jérôme Euzenat (INRIA) (2018-09-18)

Key points:

Further experiments: none


This file can be retrieved from URL https://sake.re/20180915-NOOR

It is possible to check out the repository by cloning https://felapton.inrialpes.fr/cakes/20180915-NOOR.git

This experiment has been transferred from its initial location at https://gforge.inria.fr.

The original, unaltered associated zip file can be obtained from https://files.inria.fr/sakere/gforge/20180915-NOOR.zip