#61 - Implement variable detection component using dkpro-tc

- added variable detection model
openminted · Dec 20, 2017 · f0cd704 · f0cd704
1 parent b06f9c7
commit f0cd704
Show file tree

Hide file tree

Showing 20 changed files with 21 additions and 0 deletions.
diff --git a/data/models/variable-detection/bipartitionThreshold.txt b/data/models/variable-detection/bipartitionThreshold.txt
@@ -0,0 +1,3 @@
+#Bipartition threshold used to train this model (only multi-label classification)
+#Tue Dec 19 17:35:02 CET 2017
+threshold=0.5
diff --git a/data/models/variable-detection/classLabels.txt b/data/models/variable-detection/classLabels.txt
@@ -0,0 +1,2 @@
+No
+Yes
diff --git a/data/models/variable-detection/classifier.ser b/data/models/variable-detection/classifier.ser
diff --git a/...sFolder/eu/openminted/uc/socialsciences/variabledetection/features/LuceneLemmaNGram.class b/...sFolder/eu/openminted/uc/socialsciences/variabledetection/features/LuceneLemmaNGram.class
diff --git a/...ction/featureClassFolder/org/dkpro/tc/features/entityrecognition/NEFeatureExtractor.class b/...ction/featureClassFolder/org/dkpro/tc/features/entityrecognition/NEFeatureExtractor.class
diff --git a/...odels/variable-detection/featureClassFolder/org/dkpro/tc/features/ngram/LuceneNGram.class b/...odels/variable-detection/featureClassFolder/org/dkpro/tc/features/ngram/LuceneNGram.class
diff --git a/...s/variable-detection/featureClassFolder/org/dkpro/tc/features/ngram/LuceneSkipNGram.class b/...s/variable-detection/featureClassFolder/org/dkpro/tc/features/ngram/LuceneSkipNGram.class
diff --git a/data/models/variable-detection/featureExtractorConfiguration.txt b/data/models/variable-detection/featureExtractorConfiguration.txt
@@ -0,0 +1,4 @@
+org.dkpro.tc.features.ngram.LuceneNGram	ngramMinN=1	uniqueFeatureExtractorName=LuceneNGram8801365502936	ngramUseTopK=50	ngramMaxN=3
+eu.openminted.uc.socialsciences.variabledetection.features.LuceneLemmaNGram	ngramMinN=2	uniqueFeatureExtractorName=LuceneLemmaNGram8801365738531	ngramUseTopK=50	ngramMaxN=3
+org.dkpro.tc.features.ngram.LuceneSkipNGram	ngramMinN=2	uniqueFeatureExtractorName=LuceneSkipNGram8801365971564	ngramUseTopK=50	ngramMaxN=3
+org.dkpro.tc.features.entityrecognition.NEFeatureExtractor	uniqueFeatureExtractorName=NEFeatureExtractor8801366308006
diff --git a/data/models/variable-detection/featureMode.txt b/data/models/variable-detection/featureMode.txt
@@ -0,0 +1,3 @@
+#Feature mode used to train this model
+#Wed Dec 20 12:03:17 CET 2017
+featureMode=document
diff --git a/data/models/variable-detection/learningMode.txt b/data/models/variable-detection/learningMode.txt
@@ -0,0 +1,3 @@
+#Learning mode used to train this model
+#Wed Dec 20 12:03:17 CET 2017
+learningMode=singleLabel
diff --git a/data/models/variable-detection/lucene/_0.cfe b/data/models/variable-detection/lucene/_0.cfe
diff --git a/data/models/variable-detection/lucene/_0.cfs b/data/models/variable-detection/lucene/_0.cfs
diff --git a/data/models/variable-detection/lucene/_0.si b/data/models/variable-detection/lucene/_0.si
diff --git a/data/models/variable-detection/lucene/segments.gen b/data/models/variable-detection/lucene/segments.gen
diff --git a/data/models/variable-detection/lucene/segments_1 b/data/models/variable-detection/lucene/segments_1
diff --git a/data/models/variable-detection/meta.txt b/data/models/variable-detection/meta.txt
@@ -0,0 +1 @@
+org.dkpro.tc.ml.weka.WekaClassificationAdapter
diff --git a/data/models/variable-detection/metaCollectorOverrides.txt b/data/models/variable-detection/metaCollectorOverrides.txt
@@ -0,0 +1 @@
+targetLocation=lucene
diff --git a/data/models/variable-detection/metaExtractorOverrides.txt b/data/models/variable-detection/metaExtractorOverrides.txt
@@ -0,0 +1 @@
+sourceLocation=lucene
diff --git a/data/models/variable-detection/tcVersion.txt b/data/models/variable-detection/tcVersion.txt
@@ -0,0 +1,3 @@
+#Version of DKPro TC used to train this model
+#Tue Dec 19 17:35:02 CET 2017
+TcVersion=0.9.0
diff --git a/data/models/variable-detection/training_data b/data/models/variable-detection/training_data