feat(benchmarks): expand multilingual stemming quality evaluation

* cover all Radixor dictionary languages
* add PRIMARY_OUTPUT, ANY_CANDIDATE, and ALL_CANDIDATES policies
* measure pairwise over-stemming and under-stemming
* add balanced accuracy and complementary quality metrics
* compare single-output and multi-output stemmers fairly
* improve result validation, reporting, and documentation
* move stemming quality tests into the standard test source set
* preserve the existing JMH benchmark structure and badge output
This commit is contained in:
2026-07-20 23:20:17 +02:00
parent 6d35f01303
commit 05f3855b99
62 changed files with 11472 additions and 75 deletions

View File

@@ -13,7 +13,7 @@ net.jqwik:jqwik-time:1.9.3=jmhRuntimeClasspath,testCompileClasspath,testRuntimeC
net.jqwik:jqwik-web:1.9.3=jmhRuntimeClasspath,testCompileClasspath,testRuntimeClasspath
net.jqwik:jqwik:1.9.3=jmhRuntimeClasspath,testCompileClasspath,testRuntimeClasspath
net.sf.jopt-simple:jopt-simple:4.9=pitest
net.sf.jopt-simple:jopt-simple:5.0.4=jmh,jmhCompileClasspath,jmhRuntimeClasspath
net.sf.jopt-simple:jopt-simple:5.0.4=jmh,jmhCompileClasspath,jmhRuntimeClasspath,stemmingQualityJmhRuntime
net.sf.saxon:Saxon-HE:12.9=pmd
net.sourceforge.pmd:pmd-ant:7.20.0=pmd
net.sourceforge.pmd:pmd-core:7.20.0=pmd
@@ -22,17 +22,17 @@ org.antlr:antlr4-runtime:4.9.3=pmd
org.antlr:stringtemplate:3.2.1=pitest
org.apache.commons:commons-lang3:3.18.0=pitest
org.apache.commons:commons-lang3:3.20.0=pmd
org.apache.commons:commons-math3:3.6.1=jmh,jmhCompileClasspath,jmhRuntimeClasspath
org.apache.commons:commons-math3:3.6.1=jmh,jmhCompileClasspath,jmhRuntimeClasspath,stemmingQualityJmhRuntime
org.apache.commons:commons-text:1.14.0=pitest
org.apache.lucene:lucene-analysis-common:10.5.0=jmhCompileClasspath,jmhRuntimeClasspath
org.apache.lucene:lucene-analysis-morfologik:10.5.0=jmhCompileClasspath,jmhRuntimeClasspath
org.apache.lucene:lucene-analysis-stempel:10.5.0=jmhCompileClasspath,jmhRuntimeClasspath
org.apache.lucene:lucene-core:10.5.0=jmhCompileClasspath,jmhRuntimeClasspath
org.apache.opennlp:opennlp-tools:2.5.4=jmhCompileClasspath,jmhRuntimeClasspath
org.apache.lucene:lucene-analysis-common:10.5.0=jmhCompileClasspath,jmhRuntimeClasspath,stemmingQualityJmhRuntime
org.apache.lucene:lucene-analysis-morfologik:10.5.0=jmhCompileClasspath,jmhRuntimeClasspath,stemmingQualityJmhRuntime
org.apache.lucene:lucene-analysis-stempel:10.5.0=jmhCompileClasspath,jmhRuntimeClasspath,stemmingQualityJmhRuntime
org.apache.lucene:lucene-core:10.5.0=jmhCompileClasspath,jmhRuntimeClasspath,stemmingQualityJmhRuntime
org.apache.opennlp:opennlp-tools:2.5.4=jmhCompileClasspath,jmhRuntimeClasspath,stemmingQualityJmhRuntime
org.apiguardian:apiguardian-api:1.1.2=jmhRuntimeClasspath,testCompileClasspath,testRuntimeClasspath
org.carrot2:morfologik-fsa:2.1.9=jmhCompileClasspath,jmhRuntimeClasspath
org.carrot2:morfologik-polish:2.1.9=jmhRuntimeClasspath
org.carrot2:morfologik-stemming:2.1.9=jmhCompileClasspath,jmhRuntimeClasspath
org.carrot2:morfologik-fsa:2.1.9=jmhCompileClasspath,jmhRuntimeClasspath,stemmingQualityJmhRuntime
org.carrot2:morfologik-polish:2.1.9=jmhRuntimeClasspath,stemmingQualityJmhRuntime
org.carrot2:morfologik-stemming:2.1.9=jmhCompileClasspath,jmhRuntimeClasspath,stemmingQualityJmhRuntime
org.checkerframework:checker-qual:3.52.1=pmd
org.jacoco:org.jacoco.agent:0.8.14=jacocoAgent,jacocoAnt
org.jacoco:org.jacoco.ant:0.8.14=jacocoAnt
@@ -49,10 +49,10 @@ org.junit:junit-bom:5.14.3=jmhRuntimeClasspath,testCompileClasspath,testRuntimeC
org.mockito:mockito-core:5.23.0=jmhRuntimeClasspath,mockitoAgent,testCompileClasspath,testRuntimeClasspath
org.mockito:mockito-junit-jupiter:5.23.0=jmhRuntimeClasspath,testCompileClasspath,testRuntimeClasspath
org.objenesis:objenesis:3.3=jmhRuntimeClasspath,testRuntimeClasspath
org.openjdk.jmh:jmh-core:1.37=jmh,jmhCompileClasspath,jmhRuntimeClasspath
org.openjdk.jmh:jmh-generator-asm:1.37=jmh,jmhCompileClasspath,jmhRuntimeClasspath
org.openjdk.jmh:jmh-generator-bytecode:1.37=jmh,jmhCompileClasspath,jmhRuntimeClasspath
org.openjdk.jmh:jmh-generator-reflection:1.37=jmh,jmhCompileClasspath,jmhRuntimeClasspath
org.openjdk.jmh:jmh-core:1.37=jmh,jmhCompileClasspath,jmhRuntimeClasspath,stemmingQualityJmhRuntime
org.openjdk.jmh:jmh-generator-asm:1.37=jmh,jmhCompileClasspath,jmhRuntimeClasspath,stemmingQualityJmhRuntime
org.openjdk.jmh:jmh-generator-bytecode:1.37=jmh,jmhCompileClasspath,jmhRuntimeClasspath,stemmingQualityJmhRuntime
org.openjdk.jmh:jmh-generator-reflection:1.37=jmh,jmhCompileClasspath,jmhRuntimeClasspath,stemmingQualityJmhRuntime
org.opentest4j:opentest4j:1.3.0=jmhRuntimeClasspath,testCompileClasspath,testRuntimeClasspath
org.ow2.asm:asm-analysis:9.9.1=pitest
org.ow2.asm:asm-commons:9.9=jacocoAnt
@@ -60,7 +60,7 @@ org.ow2.asm:asm-commons:9.9.1=pitest
org.ow2.asm:asm-tree:9.9=jacocoAnt
org.ow2.asm:asm-tree:9.9.1=pitest
org.ow2.asm:asm-util:9.9.1=pitest
org.ow2.asm:asm:9.0=jmh,jmhCompileClasspath,jmhRuntimeClasspath
org.ow2.asm:asm:9.0=jmh,jmhCompileClasspath,jmhRuntimeClasspath,stemmingQualityJmhRuntime
org.ow2.asm:asm:9.9=jacocoAnt
org.ow2.asm:asm:9.9.1=pitest,pmd
org.pcollections:pcollections:4.0.2=pmd
@@ -70,7 +70,7 @@ org.pitest:pitest-html-report:1.22.1=pitest
org.pitest:pitest-junit5-plugin:1.2.3=pitest
org.pitest:pitest:1.22.1=pitest
org.slf4j:jul-to-slf4j:1.7.36=pmd
org.slf4j:slf4j-api:2.0.17=jmhCompileClasspath,jmhRuntimeClasspath
org.slf4j:slf4j-api:2.0.17=jmhCompileClasspath,jmhRuntimeClasspath,stemmingQualityJmhRuntime
org.xmlresolver:xmlresolver:5.3.3=pmd
ua.net.nlp:morfologik-ukrainian-search:4.9.1=jmhRuntimeClasspath
ua.net.nlp:morfologik-ukrainian-search:4.9.1=jmhRuntimeClasspath,stemmingQualityJmhRuntime
empty=annotationProcessor,compileClasspath,cyclonedxBom,jmhAnnotationProcessor,mainPmdAuxClasspath,runtimeClasspath,testAnnotationProcessor