diff --git a/build.gradle b/build.gradle index 148fced..5ea54de 100644 --- a/build.gradle +++ b/build.gradle @@ -519,6 +519,7 @@ apply from: 'gradle/snowball-benchmarks.gradle' apply from: 'gradle/paicehusk-benchmarks.gradle' apply from: 'gradle/opennlp-benchmarks.gradle' apply from: 'gradle/hunspell-benchmarks.gradle' +apply from: 'gradle/cistem-benchmarks.gradle' gradle.taskGraph.whenReady { taskGraph -> def banner = """ diff --git a/docs/benchmarking.md b/docs/benchmarking.md index 9b6fa8c..7cab933 100644 --- a/docs/benchmarking.md +++ b/docs/benchmarking.md @@ -10,7 +10,7 @@ This page is the entry point for benchmark interpretation. Detailed tables and l - Radixor is the quality-oriented baseline in same-language comparisons. Its exact-root accuracy is often close to 100%, while many faster competitors are light, minimal, possessive, or aggressive rule-based stemmers with much lower root agreement. - The measured Radixor cost buys dictionary-trained stemming precision. That precision improves search quality by mapping inflected forms to intended dictionary roots instead of approximate or over-reduced stems. - Speed benchmarks process changed dictionary tokens where the surface form differs from the expected root. Accuracy benchmarks process the complete dictionary. -- Accuracy-only benchmarks intentionally use one deterministic JMH measurement iteration without warmup because repeated precision passes would duplicate the same counters. +- Accuracy tables use deterministic auxiliary counters from the current JMH reports. Repeated measurement samples duplicate the same exact-root accounting and are not interpreted as timing results. - The historical Porter performance badge is retired. Benchmark reporting now uses speed and quality tables rather than a single Porter ratio. ## Benchmark Documentation Map diff --git a/docs/benchmarks/index.md b/docs/benchmarks/index.md index 0f43071..9ba1db4 100644 --- a/docs/benchmarks/index.md +++ b/docs/benchmarks/index.md @@ -47,8 +47,8 @@ Open [Language Benchmark Pages](languages/index.md) for the complete language li The English dictionary coverage benchmark shows the current contracted-trie operating curve. With the full English dictionary, Radixor reaches `97.478%` all-token exactness and `97.197%` -changed-token exactness at `109.8 ns/token`. Even with a deterministic 10% dictionary slice, it -keeps `92.868%` all-token exactness and `76.516%` changed-token exactness at `90.9 ns/token`. +changed-token exactness at `135.8 ns/token`. Even with a deterministic 10% dictionary slice, it +keeps `92.868%` all-token exactness and `76.516%` changed-token exactness at `86.0 ns/token`. Those figures should not be reduced to a single speed badge. The professional interpretation is a quality/speed envelope: the amount and quality of dictionary knowledge affect stemming precision, diff --git a/docs/benchmarks/languages/czech.md b/docs/benchmarks/languages/czech.md index 7eefeb5..bd96031 100644 --- a/docs/benchmarks/languages/czech.md +++ b/docs/benchmarks/languages/czech.md @@ -26,11 +26,12 @@ Radixor stores the preferred transformation for each normalized dictionary word ## Accuracy -Accuracy is computed from one deterministic JMH measurement iteration without warmup. The benchmark may execute the full dictionary pass more than once inside that single timed iteration; percentages divide matching counters by evaluated counters from the same iteration. +Accuracy is computed from JMH auxiliary counters in the current report. The counters are deterministic for a fixed corpus and stemmer; percentages divide matching counters by evaluated counters from the same report and are not timing metrics. | Stemmer | All exact | Changed exact | Root preserved | Note | | --- | ---: | ---: | ---: | --- | | Radixor | 99.465% | 99.439% | 99.582% | Full Radixor dictionary patch-command stemmer. | +| Lucene HunspellStemFilter | 84.850% | 82.269% | 96.806% | Benchmark-only Czech Hunspell dictionary compared via Lucene HunspellStemFilter. | | Lucene CzechStemFilter | 16.784% | 15.538% | 22.559% | Lucene Czech suffix stemmer implemented as a TokenFilter. | ## Speed @@ -39,8 +40,9 @@ Speed uses JMH average time, 3 warmup iterations, 5 measurement iterations, 1 fo | Stemmer | Benchmark method | Score ms/op | Error ms | ns/token | Relative vs Radixor | Note | | --- | --- | ---: | ---: | ---: | ---: | --- | -| Radixor | `czechRadixor` | 3.117 | 0.454 | 66.9 | 1.000 | Full Radixor dictionary patch-command stemmer. | -| Lucene CzechStemFilter | `czechLuceneCzechStemFilter` | 2.921 | 0.202 | 62.7 | 0.937 | Czech suffix stemmer implemented as a Lucene TokenFilter. | +| Radixor | `czechRadixor` | 3.332 | 0.240 | 71.6 | 1.000 | Full Radixor dictionary patch-command stemmer. | +| Lucene HunspellStemFilter | `luceneHunspellStemFilter` | 346.819 | 3.622 | 7448.4 | 104.091 | Benchmark-only Czech Hunspell dictionary compared via Lucene HunspellStemFilter. | +| Lucene CzechStemFilter | `czechLuceneCzechStemFilter` | 3.163 | 0.253 | 67.9 | 0.949 | Czech suffix stemmer implemented as a Lucene TokenFilter. | ## Interpretation Notes diff --git a/docs/benchmarks/languages/danish.md b/docs/benchmarks/languages/danish.md index 9d09561..b1d5fe0 100644 --- a/docs/benchmarks/languages/danish.md +++ b/docs/benchmarks/languages/danish.md @@ -26,11 +26,11 @@ Radixor stores the preferred transformation for each normalized dictionary word ## Accuracy -Accuracy is computed from one deterministic JMH measurement iteration without warmup. The benchmark may execute the full dictionary pass more than once inside that single timed iteration; percentages divide matching counters by evaluated counters from the same iteration. +Accuracy is computed from JMH auxiliary counters in the current report. The counters are deterministic for a fixed corpus and stemmer; percentages divide matching counters by evaluated counters from the same report and are not timing metrics. | Stemmer | All exact | Changed exact | Root preserved | Note | | --- | ---: | ---: | ---: | --- | -| Radixor | 99.371% | 99.527% | 98.923% | Radixor baseline in the Snowball-language comparison family. | +| Radixor | 99.371% | 99.527% | 98.923% | Full Radixor dictionary patch-command stemmer. | | Lucene SnowballFilter | 55.509% | 54.159% | 59.371% | Lucene TokenFilter integration path around the Snowball algorithm. | | Official Snowball direct | 55.509% | 54.159% | 59.371% | Official Snowball generated Java stemmer; rule-based suffix algorithm. | @@ -40,9 +40,9 @@ Speed uses JMH average time, 3 warmup iterations, 5 measurement iterations, 1 fo | Stemmer | Benchmark method | Score ms/op | Error ms | ns/token | Relative vs Radixor | Note | | --- | --- | ---: | ---: | ---: | ---: | --- | -| Radixor | `radixor[DANISH]` | 1.065 | 0.019 | 44.6 | 1.000 | Radixor baseline for the Snowball-language comparison family. | -| Official Snowball direct | `snowballDirect[DANISH]` | 2.028 | 0.011 | 84.9 | 1.904 | Official Snowball generated Java stemmer; direct API. | -| Lucene SnowballFilter | `luceneSnowballFilter[DANISH]` | 2.692 | 0.028 | 112.6 | 2.527 | Lucene TokenFilter path around Snowball; includes TokenStream overhead. | +| Radixor | `radixor[DANISH]` | 1.143 | 0.017 | 47.8 | 1.000 | Full Radixor dictionary patch-command stemmer. | +| Official Snowball direct | `snowballDirect[DANISH]` | 2.168 | 0.058 | 90.7 | 1.896 | Official Snowball generated Java stemmer; direct API. | +| Lucene SnowballFilter | `luceneSnowballFilter[DANISH]` | 2.975 | 0.143 | 124.5 | 2.602 | Lucene TokenFilter path around Snowball; includes TokenStream overhead. | ## Interpretation Notes diff --git a/docs/benchmarks/languages/dutch.md b/docs/benchmarks/languages/dutch.md index 719a96d..a5ad3ea 100644 --- a/docs/benchmarks/languages/dutch.md +++ b/docs/benchmarks/languages/dutch.md @@ -26,11 +26,12 @@ Radixor stores the preferred transformation for each normalized dictionary word ## Accuracy -Accuracy is computed from one deterministic JMH measurement iteration without warmup. The benchmark may execute the full dictionary pass more than once inside that single timed iteration; percentages divide matching counters by evaluated counters from the same iteration. +Accuracy is computed from JMH auxiliary counters in the current report. The counters are deterministic for a fixed corpus and stemmer; percentages divide matching counters by evaluated counters from the same report and are not timing metrics. | Stemmer | All exact | Changed exact | Root preserved | Note | | --- | ---: | ---: | ---: | --- | -| Radixor | 99.120% | 98.711% | 100.000% | Radixor baseline in the Snowball-language comparison family. | +| Radixor | 99.120% | 98.711% | 100.000% | Full Radixor dictionary patch-command stemmer. | +| Lucene HunspellStemFilter | 46.590% | 22.718% | 97.976% | Benchmark-only Dutch Hunspell dictionary compared via Lucene HunspellStemFilter. | | Official Snowball direct | 15.954% | 8.992% | 30.939% | Official Snowball generated Java stemmer; rule-based suffix algorithm. | | Lucene SnowballFilter | 12.620% | 5.441% | 28.073% | Lucene TokenFilter integration path around the Snowball algorithm. | @@ -40,9 +41,10 @@ Speed uses JMH average time, 3 warmup iterations, 5 measurement iterations, 1 fo | Stemmer | Benchmark method | Score ms/op | Error ms | ns/token | Relative vs Radixor | Note | | --- | --- | ---: | ---: | ---: | ---: | --- | -| Radixor | `radixor[DUTCH]` | 1.262 | 0.039 | 58.7 | 1.000 | Radixor baseline for the Snowball-language comparison family. | -| Official Snowball direct | `snowballDirect[DUTCH]` | 3.968 | 0.258 | 184.7 | 3.145 | Official Snowball generated Java stemmer; direct API. | -| Lucene SnowballFilter | `luceneSnowballFilter[DUTCH]` | 6.866 | 0.337 | 319.6 | 5.441 | Lucene TokenFilter path around Snowball; includes TokenStream overhead. | +| Radixor | `radixor[DUTCH]` | 1.331 | 0.114 | 61.9 | 1.000 | Full Radixor dictionary patch-command stemmer. | +| Lucene HunspellStemFilter | `luceneHunspellStemFilter` | 22.760 | 1.387 | 1059.3 | 17.105 | Benchmark-only Dutch Hunspell dictionary compared via Lucene HunspellStemFilter. | +| Official Snowball direct | `snowballDirect[DUTCH]` | 4.146 | 0.291 | 193.0 | 3.116 | Official Snowball generated Java stemmer; direct API. | +| Lucene SnowballFilter | `luceneSnowballFilter[DUTCH]` | 7.375 | 0.595 | 343.3 | 5.543 | Lucene TokenFilter path around Snowball; includes TokenStream overhead. | ## Interpretation Notes diff --git a/docs/benchmarks/languages/english.md b/docs/benchmarks/languages/english.md index 179cb08..926b86c 100644 --- a/docs/benchmarks/languages/english.md +++ b/docs/benchmarks/languages/english.md @@ -26,13 +26,14 @@ Radixor stores the preferred transformation for each normalized dictionary word ## Accuracy -Accuracy is computed from one deterministic JMH measurement iteration without warmup. The benchmark may execute the full dictionary pass more than once inside that single timed iteration; percentages divide matching counters by evaluated counters from the same iteration. +Accuracy is computed from JMH auxiliary counters in the current report. The counters are deterministic for a fixed corpus and stemmer; percentages divide matching counters by evaluated counters from the same report and are not timing metrics. | Stemmer | All exact | Changed exact | Root preserved | Note | | --- | ---: | ---: | ---: | --- | | Radixor | 97.478% | 97.197% | 97.552% | Full Radixor dictionary patch-command stemmer. | | Lucene EnglishMinimalStemFilter | 90.981% | 65.189% | 97.820% | Minimal English plural reduction, not a full stemmer. | | Lucene KStemFilter | 80.076% | 76.608% | 80.996% | Krovetz-style English stemming TokenFilter; broader than minimal suffix reducers. | +| Lucene HunspellStemFilter | 80.243% | 12.750% | 98.139% | Benchmark-only English Hunspell dictionary compared via Lucene HunspellStemFilter. | | Lucene EnglishPossessiveFilter | 79.032% | 0.003% | 99.987% | Possessive-ending remover only, not a full stemmer. | | Snowball English / Porter2 | 40.342% | 46.296% | 38.763% | Porter2 rule-based suffix stemmer, distinct from original Porter. | | Lucene PorterStemFilter | 39.538% | 46.201% | 37.772% | Lucene TokenFilter path for Porter suffix rules; not dictionary-root equivalent. | @@ -47,16 +48,17 @@ Speed uses JMH average time, 3 warmup iterations, 5 measurement iterations, 1 fo | Stemmer | Benchmark method | Score ms/op | Error ms | ns/token | Relative vs Radixor | Note | | --- | --- | ---: | ---: | ---: | ---: | --- | -| Radixor | `radixorUsUkProfiPreferredStem` | 16.621 | 8.532 | 79.0 | 1.000 | Full dictionary patch-command stemmer using compiled patch commands. | -| Lucene EnglishPossessiveFilter | `luceneEnglishPossessiveFilter` | 23.845 | 0.833 | 113.3 | 1.435 | Possessive-ending remover only; not a full stemmer. | -| Lucene EnglishMinimalStemFilter | `luceneEnglishMinimalStemFilter` | 17.091 | 0.198 | 81.2 | 1.028 | Narrow plural reduction filter; not a full stemmer. | -| Lucene PorterStemmer direct copy | `lucenePorterStemmerCopied` | 18.598 | 10.954 | 88.4 | 1.119 | Benchmark-only generated copy of Lucene package-private Porter implementation. | -| OpenNLP PorterStemmer | `opennlpPorterStemmer` | 18.213 | 10.674 | 86.5 | 1.096 | Apache OpenNLP Porter implementation. | -| Snowball original Porter | `snowballOriginalPorter` | 32.921 | 11.520 | 156.4 | 1.981 | Classic Porter suffix-rule stemmer; historical English baseline, not a dictionary-equivalent stemmer. | -| Lucene PorterStemFilter | `lucenePorterStemFilter` | 42.874 | 1.321 | 203.7 | 2.579 | Lucene TokenFilter integration path for Porter; includes TokenStream overhead. | -| Lucene KStemFilter | `luceneKStemFilter` | 50.483 | 3.624 | 239.8 | 3.037 | Krovetz-style English TokenFilter; broader than minimal suffix filters. | -| Snowball English / Porter2 | `snowballEnglishPorter2` | 47.844 | 1.887 | 227.3 | 2.878 | Porter2 suffix-rule stemmer, distinct from original Porter. | -| Paice/Husk Lancaster | `paiceHuskLancaster` | 135.050 | 11.088 | 641.6 | 8.125 | Aggressive rule-based English stemmer. | +| Radixor | `radixorUsUkProfiPreferredStem` | 21.987 | 8.707 | 104.5 | 1.000 | Full dictionary patch-command stemmer using compiled patch commands. | +| Lucene EnglishPossessiveFilter | `luceneEnglishPossessiveFilter` | 24.539 | 1.515 | 116.6 | 1.116 | Possessive-ending remover only; not a full stemmer. | +| Lucene EnglishMinimalStemFilter | `luceneEnglishMinimalStemFilter` | 22.702 | 1.195 | 107.8 | 1.032 | Narrow plural reduction filter; not a full stemmer. | +| Lucene PorterStemmer direct copy | `lucenePorterStemmerCopied` | 24.696 | 13.235 | 117.3 | 1.123 | Benchmark-only generated copy of Lucene package-private Porter implementation. | +| OpenNLP PorterStemmer | `opennlpPorterStemmer` | 23.121 | 12.528 | 109.8 | 1.052 | Apache OpenNLP Porter implementation. | +| Snowball original Porter | `snowballOriginalPorter` | 38.904 | 10.353 | 184.8 | 1.769 | Classic Porter suffix-rule stemmer; historical English baseline, not a dictionary-equivalent stemmer. | +| Lucene PorterStemFilter | `lucenePorterStemFilter` | 37.021 | 1.196 | 175.9 | 1.684 | Lucene TokenFilter integration path for Porter; includes TokenStream overhead. | +| Lucene KStemFilter | `luceneKStemFilter` | 51.640 | 2.591 | 245.3 | 2.349 | Krovetz-style English TokenFilter; broader than minimal suffix filters. | +| Lucene HunspellStemFilter | `luceneHunspellStemFilter` | 79.785 | 1.347 | 379.0 | 3.629 | Benchmark-only English Hunspell comparison using the benchmark Hunspell corpus. | +| Snowball English / Porter2 | `snowballEnglishPorter2` | 52.437 | 0.773 | 249.1 | 2.385 | Porter2 suffix-rule stemmer, distinct from original Porter. | +| Paice/Husk Lancaster | `paiceHuskLancaster` | 141.556 | 12.324 | 672.5 | 6.438 | Aggressive rule-based English stemmer. | ## Interpretation Notes diff --git a/docs/benchmarks/languages/finnish.md b/docs/benchmarks/languages/finnish.md index 1435e88..692bac8 100644 --- a/docs/benchmarks/languages/finnish.md +++ b/docs/benchmarks/languages/finnish.md @@ -26,7 +26,7 @@ Radixor stores the preferred transformation for each normalized dictionary word ## Accuracy -Accuracy is computed from one deterministic JMH measurement iteration without warmup. The benchmark may execute the full dictionary pass more than once inside that single timed iteration; percentages divide matching counters by evaluated counters from the same iteration. +Accuracy is computed from JMH auxiliary counters in the current report. The counters are deterministic for a fixed corpus and stemmer; percentages divide matching counters by evaluated counters from the same report and are not timing metrics. | Stemmer | All exact | Changed exact | Root preserved | Note | | --- | ---: | ---: | ---: | --- | @@ -41,10 +41,10 @@ Speed uses JMH average time, 3 warmup iterations, 5 measurement iterations, 1 fo | Stemmer | Benchmark method | Score ms/op | Error ms | ns/token | Relative vs Radixor | Note | | --- | --- | ---: | ---: | ---: | ---: | --- | -| Radixor | `finnishRadixor` | 228.248 | 9.245 | 130.1 | 1.000 | Full Radixor dictionary patch-command stemmer. | -| Lucene FinnishLightStemFilter | `finnishLuceneFinnishLightStemFilter` | 175.923 | 87.177 | 100.3 | 0.771 | Light Finnish suffix stemmer. | -| Official Snowball direct | `snowballDirect[FINNISH]` | 265.579 | 95.687 | 151.4 | 1.164 | Official Snowball generated Java stemmer; direct API. | -| Lucene SnowballFilter | `luceneSnowballFilter[FINNISH]` | 338.099 | 175.101 | 192.7 | 1.481 | Lucene TokenFilter path around Snowball; includes TokenStream overhead. | +| Radixor | `finnishRadixor` | 308.076 | 15.529 | 175.6 | 1.000 | Full Radixor dictionary patch-command stemmer. | +| Lucene FinnishLightStemFilter | `finnishLuceneFinnishLightStemFilter` | 175.250 | 46.995 | 99.9 | 0.569 | Light Finnish suffix stemmer. | +| Official Snowball direct | `snowballDirect[FINNISH]` | 264.652 | 63.054 | 150.8 | 0.859 | Official Snowball generated Java stemmer; direct API. | +| Lucene SnowballFilter | `luceneSnowballFilter[FINNISH]` | 374.883 | 238.157 | 213.6 | 1.217 | Lucene TokenFilter path around Snowball; includes TokenStream overhead. | ## Interpretation Notes diff --git a/docs/benchmarks/languages/french.md b/docs/benchmarks/languages/french.md index b57b667..685bc8a 100644 --- a/docs/benchmarks/languages/french.md +++ b/docs/benchmarks/languages/french.md @@ -26,11 +26,12 @@ Radixor stores the preferred transformation for each normalized dictionary word ## Accuracy -Accuracy is computed from one deterministic JMH measurement iteration without warmup. The benchmark may execute the full dictionary pass more than once inside that single timed iteration; percentages divide matching counters by evaluated counters from the same iteration. +Accuracy is computed from JMH auxiliary counters in the current report. The counters are deterministic for a fixed corpus and stemmer; percentages divide matching counters by evaluated counters from the same report and are not timing metrics. | Stemmer | All exact | Changed exact | Root preserved | Note | | --- | ---: | ---: | ---: | --- | | Radixor | 94.831% | 94.859% | 94.734% | Full Radixor dictionary patch-command stemmer. | +| Lucene HunspellStemFilter | 68.923% | 63.617% | 86.876% | Benchmark-only French Hunspell dictionary compared via Lucene HunspellStemFilter. | | Lucene FrenchMinimalStemFilter | 11.472% | 6.236% | 29.192% | Minimal suffix reducer; narrow baseline, not a full stemmer. | | Lucene SnowballFilter | 8.551% | 5.183% | 19.952% | Lucene TokenFilter integration path around the Snowball algorithm. | | Official Snowball direct | 8.462% | 5.067% | 19.952% | Official Snowball generated Java stemmer; rule-based suffix algorithm. | @@ -42,11 +43,12 @@ Speed uses JMH average time, 3 warmup iterations, 5 measurement iterations, 1 fo | Stemmer | Benchmark method | Score ms/op | Error ms | ns/token | Relative vs Radixor | Note | | --- | --- | ---: | ---: | ---: | ---: | --- | -| Radixor | `frenchRadixor` | 38.598 | 5.425 | 105.5 | 1.000 | Full Radixor dictionary patch-command stemmer. | -| Lucene FrenchMinimalStemFilter | `frenchLuceneFrenchMinimalStemFilter` | 17.657 | 1.956 | 48.2 | 0.457 | Minimal French suffix reducer; narrow baseline. | -| Lucene FrenchLightStemFilter | `frenchLuceneFrenchLightStemFilter` | 28.742 | 2.391 | 78.5 | 0.745 | Light French suffix stemmer. | -| Official Snowball direct | `snowballDirect[FRENCH]` | 104.983 | 10.116 | 286.9 | 2.720 | Official Snowball generated Java stemmer; direct API. | -| Lucene SnowballFilter | `luceneSnowballFilter[FRENCH]` | 117.938 | 4.007 | 322.3 | 3.056 | Lucene TokenFilter path around Snowball; includes TokenStream overhead. | +| Radixor | `frenchRadixor` | 47.033 | 4.146 | 128.5 | 1.000 | Full Radixor dictionary patch-command stemmer. | +| Lucene HunspellStemFilter | `luceneHunspellStemFilter` | 1664.935 | 65.928 | 4549.4 | 35.399 | Benchmark-only French Hunspell dictionary compared via Lucene HunspellStemFilter. | +| Lucene FrenchMinimalStemFilter | `frenchLuceneFrenchMinimalStemFilter` | 19.234 | 2.098 | 52.6 | 0.409 | Minimal French suffix reducer; narrow baseline. | +| Lucene FrenchLightStemFilter | `frenchLuceneFrenchLightStemFilter` | 30.560 | 3.680 | 83.5 | 0.650 | Light French suffix stemmer. | +| Official Snowball direct | `snowballDirect[FRENCH]` | 111.057 | 8.172 | 303.5 | 2.361 | Official Snowball generated Java stemmer; direct API. | +| Lucene SnowballFilter | `luceneSnowballFilter[FRENCH]` | 123.648 | 3.500 | 337.9 | 2.629 | Lucene TokenFilter path around Snowball; includes TokenStream overhead. | ## Interpretation Notes diff --git a/docs/benchmarks/languages/german.md b/docs/benchmarks/languages/german.md index 548719e..5134c02 100644 --- a/docs/benchmarks/languages/german.md +++ b/docs/benchmarks/languages/german.md @@ -26,16 +26,18 @@ Radixor stores the preferred transformation for each normalized dictionary word ## Accuracy -Accuracy is computed from one deterministic JMH measurement iteration without warmup. The benchmark may execute the full dictionary pass more than once inside that single timed iteration; percentages divide matching counters by evaluated counters from the same iteration. +Accuracy is computed from JMH auxiliary counters in the current report. The counters are deterministic for a fixed corpus and stemmer; percentages divide matching counters by evaluated counters from the same report and are not timing metrics. | Stemmer | All exact | Changed exact | Root preserved | Note | | --- | ---: | ---: | ---: | --- | -| Radixor | 97.455% | 97.973% | 96.476% | Full Radixor dictionary patch-command stemmer. | -| Lucene GermanLightStemFilter | 38.583% | 35.800% | 43.849% | Light suffix stemmer; intentionally narrower than a dictionary-derived stemmer. | -| Lucene GermanMinimalStemFilter | 37.492% | 38.538% | 35.513% | Minimal suffix reducer; narrow baseline, not a full stemmer. | -| Lucene SnowballFilter | 33.380% | 30.939% | 37.999% | Lucene TokenFilter integration path around the Snowball algorithm. | -| Official Snowball direct | 32.863% | 31.225% | 35.963% | Official Snowball generated Java stemmer; rule-based suffix algorithm. | -| Lucene GermanStemFilter | 26.168% | 24.979% | 28.416% | German Lucene stemming TokenFilter; broader than minimal/light variants. | +| Radixor | 92.725% | 92.847% | 92.396% | Full Radixor dictionary patch-command stemmer. | +| Lucene HunspellStemFilter | 47.064% | 29.661% | 93.678% | Benchmark-only German Hunspell dictionary compared via Lucene HunspellStemFilter. | +| CISTEM (German) | 24.675% | 23.724% | 27.222% | Benchmark-only CISTEM implementation. | +| Lucene GermanLightStemFilter | 37.434% | 35.465% | 42.707% | Light suffix stemmer; intentionally narrower than a dictionary-derived stemmer. | +| Lucene GermanMinimalStemFilter | 27.640% | 24.951% | 34.844% | Minimal suffix reducer; narrow baseline, not a full stemmer. | +| Lucene SnowballFilter | 30.956% | 28.853% | 36.589% | Lucene TokenFilter integration path around the Snowball algorithm. | +| Official Snowball direct | 30.481% | 29.027% | 34.376% | Official Snowball generated Java stemmer; rule-based suffix algorithm. | +| Lucene GermanStemFilter | 21.559% | 19.312% | 27.576% | German Lucene stemming TokenFilter; broader than minimal/light variants. | ## Speed @@ -43,12 +45,14 @@ Speed uses JMH average time, 3 warmup iterations, 5 measurement iterations, 1 fo | Stemmer | Benchmark method | Score ms/op | Error ms | ns/token | Relative vs Radixor | Note | | --- | --- | ---: | ---: | ---: | ---: | --- | -| Radixor | `germanRadixor` | 9.518 | 0.338 | 68.2 | 1.000 | Full Radixor dictionary patch-command stemmer. | -| Lucene GermanMinimalStemFilter | `germanLuceneGermanMinimalStemFilter` | 12.020 | 0.460 | 86.1 | 1.263 | Minimal German suffix reduction; narrow baseline. | -| Lucene GermanLightStemFilter | `germanLuceneGermanLightStemFilter` | 12.413 | 1.059 | 88.9 | 1.304 | Light German suffix stemmer; narrower than a dictionary stemmer. | -| Lucene GermanStemFilter | `germanLuceneGermanStemFilter` | 36.644 | 5.667 | 262.4 | 3.850 | Older German stemming TokenFilter with normalization requirements. | -| Lucene SnowballFilter | `luceneSnowballFilter[GERMAN]` | 54.846 | 8.710 | 392.8 | 5.762 | Lucene TokenFilter path around Snowball; includes TokenStream overhead. | -| Official Snowball direct | `snowballDirect[GERMAN]` | 52.974 | 7.989 | 379.4 | 5.566 | Official Snowball generated Java stemmer; direct API. | +| Radixor | `germanRadixor` | 41.166 | 2.396 | 294.8 | 1.000 | Full Radixor dictionary patch-command stemmer. | +| CISTEM | `germanCistem` | 248.392 | 12.294 | 1778.8 | 6.034 | Benchmark-only CISTEM implementation. | +| Lucene HunspellStemFilter | `luceneHunspellStemFilter` | 281.322 | 3.411 | 2014.6 | 6.834 | Benchmark-only German Hunspell dictionary compared via Lucene HunspellStemFilter. | +| Lucene GermanMinimalStemFilter | `germanLuceneGermanMinimalStemFilter` | 23.562 | 0.969 | 168.7 | 0.572 | Minimal German suffix reduction; narrow baseline. | +| Lucene GermanLightStemFilter | `germanLuceneGermanLightStemFilter` | 24.410 | 1.034 | 174.8 | 0.593 | Light German suffix stemmer; narrower than a dictionary stemmer. | +| Lucene GermanStemFilter | `germanLuceneGermanStemFilter` | 71.039 | 4.443 | 508.7 | 1.726 | Older German stemming TokenFilter with normalization requirements. | +| Lucene SnowballFilter | `luceneSnowballFilter[GERMAN]` | 105.771 | 9.617 | 757.4 | 2.569 | Lucene TokenFilter path around Snowball; includes TokenStream overhead. | +| Official Snowball direct | `snowballDirect[GERMAN]` | 100.688 | 9.018 | 721.0 | 2.446 | Official Snowball generated Java stemmer; direct API. | ## Interpretation Notes diff --git a/docs/benchmarks/languages/hungarian.md b/docs/benchmarks/languages/hungarian.md index d26cb3f..19eba95 100644 --- a/docs/benchmarks/languages/hungarian.md +++ b/docs/benchmarks/languages/hungarian.md @@ -26,7 +26,7 @@ Radixor stores the preferred transformation for each normalized dictionary word ## Accuracy -Accuracy is computed from one deterministic JMH measurement iteration without warmup. The benchmark may execute the full dictionary pass more than once inside that single timed iteration; percentages divide matching counters by evaluated counters from the same iteration. +Accuracy is computed from JMH auxiliary counters in the current report. The counters are deterministic for a fixed corpus and stemmer; percentages divide matching counters by evaluated counters from the same report and are not timing metrics. | Stemmer | All exact | Changed exact | Root preserved | Note | | --- | ---: | ---: | ---: | --- | @@ -41,10 +41,10 @@ Speed uses JMH average time, 3 warmup iterations, 5 measurement iterations, 1 fo | Stemmer | Benchmark method | Score ms/op | Error ms | ns/token | Relative vs Radixor | Note | | --- | --- | ---: | ---: | ---: | ---: | --- | -| Radixor | `hungarianRadixor` | 53.844 | 5.619 | 60.0 | 1.000 | Full Radixor dictionary patch-command stemmer. | -| Lucene HungarianLightStemFilter | `hungarianLuceneHungarianLightStemFilter` | 85.802 | 4.554 | 95.7 | 1.594 | Light Hungarian suffix stemmer. | -| Official Snowball direct | `snowballDirect[HUNGARIAN]` | 161.996 | 61.038 | 180.6 | 3.009 | Official Snowball generated Java stemmer; direct API. | -| Lucene SnowballFilter | `luceneSnowballFilter[HUNGARIAN]` | 185.097 | 44.447 | 206.4 | 3.438 | Lucene TokenFilter path around Snowball; includes TokenStream overhead. | +| Radixor | `hungarianRadixor` | 62.232 | 6.412 | 69.4 | 1.000 | Full Radixor dictionary patch-command stemmer. | +| Lucene HungarianLightStemFilter | `hungarianLuceneHungarianLightStemFilter` | 92.813 | 6.929 | 103.5 | 1.491 | Light Hungarian suffix stemmer. | +| Official Snowball direct | `snowballDirect[HUNGARIAN]` | 157.765 | 13.202 | 175.9 | 2.535 | Official Snowball generated Java stemmer; direct API. | +| Lucene SnowballFilter | `luceneSnowballFilter[HUNGARIAN]` | 188.863 | 15.880 | 210.6 | 3.035 | Lucene TokenFilter path around Snowball; includes TokenStream overhead. | ## Interpretation Notes diff --git a/docs/benchmarks/languages/italian.md b/docs/benchmarks/languages/italian.md index 060a376..3506f0d 100644 --- a/docs/benchmarks/languages/italian.md +++ b/docs/benchmarks/languages/italian.md @@ -25,7 +25,7 @@ Radixor stores the preferred transformation for each normalized dictionary word ## Accuracy -Accuracy is computed from one deterministic JMH measurement iteration without warmup. The benchmark may execute the full dictionary pass more than once inside that single timed iteration; percentages divide matching counters by evaluated counters from the same iteration. +Accuracy is computed from JMH auxiliary counters in the current report. The counters are deterministic for a fixed corpus and stemmer; percentages divide matching counters by evaluated counters from the same report and are not timing metrics. | Stemmer | All exact | Changed exact | Root preserved | Note | | --- | ---: | ---: | ---: | --- | @@ -40,10 +40,10 @@ Speed uses JMH average time, 3 warmup iterations, 5 measurement iterations, 1 fo | Stemmer | Benchmark method | Score ms/op | Error ms | ns/token | Relative vs Radixor | Note | | --- | --- | ---: | ---: | ---: | ---: | --- | -| Radixor | `italianRadixor` | 23.776 | 9.977 | 74.9 | 1.000 | Full Radixor dictionary patch-command stemmer. | -| Lucene ItalianLightStemFilter | `italianLuceneItalianLightStemFilter` | 14.940 | 1.682 | 47.1 | 0.628 | Light Italian suffix stemmer. | -| Official Snowball direct | `snowballDirect[ITALIAN]` | 99.401 | 9.433 | 313.0 | 4.181 | Official Snowball generated Java stemmer; direct API. | -| Lucene SnowballFilter | `luceneSnowballFilter[ITALIAN]` | 108.462 | 8.014 | 341.6 | 4.562 | Lucene TokenFilter path around Snowball; includes TokenStream overhead. | +| Radixor | `italianRadixor` | 24.491 | 3.128 | 77.1 | 1.000 | Full Radixor dictionary patch-command stemmer. | +| Lucene ItalianLightStemFilter | `italianLuceneItalianLightStemFilter` | 15.977 | 1.041 | 50.3 | 0.652 | Light Italian suffix stemmer. | +| Official Snowball direct | `snowballDirect[ITALIAN]` | 109.526 | 12.572 | 344.9 | 4.472 | Official Snowball generated Java stemmer; direct API. | +| Lucene SnowballFilter | `luceneSnowballFilter[ITALIAN]` | 116.260 | 7.459 | 366.1 | 4.747 | Lucene TokenFilter path around Snowball; includes TokenStream overhead. | ## Interpretation Notes diff --git a/docs/benchmarks/languages/norwegian-bokmal.md b/docs/benchmarks/languages/norwegian-bokmal.md index 1221d8b..dba1874 100644 --- a/docs/benchmarks/languages/norwegian-bokmal.md +++ b/docs/benchmarks/languages/norwegian-bokmal.md @@ -26,7 +26,7 @@ Radixor stores the preferred transformation for each normalized dictionary word ## Accuracy -Accuracy is computed from one deterministic JMH measurement iteration without warmup. The benchmark may execute the full dictionary pass more than once inside that single timed iteration; percentages divide matching counters by evaluated counters from the same iteration. +Accuracy is computed from JMH auxiliary counters in the current report. The counters are deterministic for a fixed corpus and stemmer; percentages divide matching counters by evaluated counters from the same report and are not timing metrics. | Stemmer | All exact | Changed exact | Root preserved | Note | | --- | ---: | ---: | ---: | --- | @@ -42,11 +42,11 @@ Speed uses JMH average time, 3 warmup iterations, 5 measurement iterations, 1 fo | Stemmer | Benchmark method | Score ms/op | Error ms | ns/token | Relative vs Radixor | Note | | --- | --- | ---: | ---: | ---: | ---: | --- | -| Radixor | `norwegianBokmalRadixor` | 3.235 | 0.147 | 56.4 | 1.000 | Full Radixor dictionary patch-command stemmer. | -| Lucene NorwegianMinimalStemFilter | `norwegianBokmalLuceneNorwegianMinimalStemFilter` | 2.720 | 0.190 | 47.4 | 0.841 | Minimal Norwegian suffix reducer. | -| Lucene NorwegianLightStemFilter | `norwegianBokmalLuceneNorwegianLightStemFilter` | 3.189 | 0.261 | 55.6 | 0.986 | Light Norwegian suffix stemmer. | -| Official Snowball direct | `snowballDirect[NORWEGIAN_BOKMAL]` | 3.978 | 0.031 | 69.3 | 1.230 | Official Snowball generated Java stemmer; direct API. | -| Lucene SnowballFilter | `luceneSnowballFilter[NORWEGIAN_BOKMAL]` | 5.526 | 0.409 | 96.3 | 1.708 | Lucene TokenFilter path around Snowball; includes TokenStream overhead. | +| Radixor | `norwegianBokmalRadixor` | 3.631 | 1.377 | 63.3 | 1.000 | Full Radixor dictionary patch-command stemmer. | +| Lucene NorwegianMinimalStemFilter | `norwegianBokmalLuceneNorwegianMinimalStemFilter` | 2.910 | 0.177 | 50.7 | 0.801 | Minimal Norwegian suffix reducer. | +| Lucene NorwegianLightStemFilter | `norwegianBokmalLuceneNorwegianLightStemFilter` | 3.335 | 0.116 | 58.1 | 0.919 | Light Norwegian suffix stemmer. | +| Official Snowball direct | `snowballDirect[NORWEGIAN_BOKMAL]` | 4.277 | 0.082 | 74.5 | 1.178 | Official Snowball generated Java stemmer; direct API. | +| Lucene SnowballFilter | `luceneSnowballFilter[NORWEGIAN_BOKMAL]` | 6.077 | 0.208 | 105.9 | 1.674 | Lucene TokenFilter path around Snowball; includes TokenStream overhead. | ## Interpretation Notes diff --git a/docs/benchmarks/languages/norwegian-nynorsk.md b/docs/benchmarks/languages/norwegian-nynorsk.md index 914842a..372085a 100644 --- a/docs/benchmarks/languages/norwegian-nynorsk.md +++ b/docs/benchmarks/languages/norwegian-nynorsk.md @@ -26,11 +26,11 @@ Radixor stores the preferred transformation for each normalized dictionary word ## Accuracy -Accuracy is computed from one deterministic JMH measurement iteration without warmup. The benchmark may execute the full dictionary pass more than once inside that single timed iteration; percentages divide matching counters by evaluated counters from the same iteration. +Accuracy is computed from JMH auxiliary counters in the current report. The counters are deterministic for a fixed corpus and stemmer; percentages divide matching counters by evaluated counters from the same report and are not timing metrics. | Stemmer | All exact | Changed exact | Root preserved | Note | | --- | ---: | ---: | ---: | --- | -| Radixor | 93.089% | 91.395% | 96.863% | Radixor baseline in the Snowball-language comparison family. | +| Radixor | 93.089% | 91.395% | 96.863% | Full Radixor dictionary patch-command stemmer. | | Official Snowball direct | 60.974% | 60.212% | 62.670% | Official Snowball generated Java stemmer; rule-based suffix algorithm. | | Lucene SnowballFilter | 60.918% | 60.146% | 62.638% | Lucene TokenFilter integration path around the Snowball algorithm. | @@ -40,9 +40,9 @@ Speed uses JMH average time, 3 warmup iterations, 5 measurement iterations, 1 fo | Stemmer | Benchmark method | Score ms/op | Error ms | ns/token | Relative vs Radixor | Note | | --- | --- | ---: | ---: | ---: | ---: | --- | -| Radixor | `radixor[NORWEGIAN_NYNORSK]` | 0.541 | 0.021 | 39.9 | 1.000 | Radixor baseline for the Snowball-language comparison family. | -| Official Snowball direct | `snowballDirect[NORWEGIAN_NYNORSK]` | 0.855 | 0.008 | 63.0 | 1.580 | Official Snowball generated Java stemmer; direct API. | -| Lucene SnowballFilter | `luceneSnowballFilter[NORWEGIAN_NYNORSK]` | 1.221 | 0.025 | 90.0 | 2.258 | Lucene TokenFilter path around Snowball; includes TokenStream overhead. | +| Radixor | `radixor[NORWEGIAN_NYNORSK]` | 0.571 | 0.012 | 42.1 | 1.000 | Full Radixor dictionary patch-command stemmer. | +| Official Snowball direct | `snowballDirect[NORWEGIAN_NYNORSK]` | 0.919 | 0.038 | 67.7 | 1.609 | Official Snowball generated Java stemmer; direct API. | +| Lucene SnowballFilter | `luceneSnowballFilter[NORWEGIAN_NYNORSK]` | 1.309 | 0.015 | 96.5 | 2.292 | Lucene TokenFilter path around Snowball; includes TokenStream overhead. | ## Interpretation Notes diff --git a/docs/benchmarks/languages/persian.md b/docs/benchmarks/languages/persian.md index 13db80a..6601984 100644 --- a/docs/benchmarks/languages/persian.md +++ b/docs/benchmarks/languages/persian.md @@ -24,7 +24,7 @@ Radixor stores the preferred transformation for each normalized dictionary word ## Accuracy -Accuracy is computed from one deterministic JMH measurement iteration without warmup. The benchmark may execute the full dictionary pass more than once inside that single timed iteration; percentages divide matching counters by evaluated counters from the same iteration. +Accuracy is computed from JMH auxiliary counters in the current report. The counters are deterministic for a fixed corpus and stemmer; percentages divide matching counters by evaluated counters from the same report and are not timing metrics. | Stemmer | All exact | Changed exact | Root preserved | Note | | --- | ---: | ---: | ---: | --- | @@ -37,8 +37,8 @@ Speed uses JMH average time, 3 warmup iterations, 5 measurement iterations, 1 fo | Stemmer | Benchmark method | Score ms/op | Error ms | ns/token | Relative vs Radixor | Note | | --- | --- | ---: | ---: | ---: | ---: | --- | -| Radixor | `persianRadixor` | 0.231 | 0.021 | 63.6 | 1.000 | Full Radixor dictionary patch-command stemmer. | -| Lucene PersianStemFilter | `persianLucenePersianStemFilter` | 0.443 | 0.012 | 122.0 | 1.918 | Persian suffix stemmer with Lucene normalization in the measured path. | +| Radixor | `persianRadixor` | 0.245 | 0.025 | 49.0 | 1.000 | Full Radixor dictionary patch-command stemmer. | +| Lucene PersianStemFilter | `persianLucenePersianStemFilter` | 0.466 | 0.015 | 93.3 | 1.902 | Persian suffix stemmer with Lucene normalization in the measured path. | ## Interpretation Notes diff --git a/docs/benchmarks/languages/polish.md b/docs/benchmarks/languages/polish.md index 5833b53..10af68f 100644 --- a/docs/benchmarks/languages/polish.md +++ b/docs/benchmarks/languages/polish.md @@ -26,11 +26,12 @@ Radixor stores the preferred transformation for each normalized dictionary word ## Accuracy -Accuracy is computed from one deterministic JMH measurement iteration without warmup. The benchmark may execute the full dictionary pass more than once inside that single timed iteration; percentages divide matching counters by evaluated counters from the same iteration. +Accuracy is computed from JMH auxiliary counters in the current report. The counters are deterministic for a fixed corpus and stemmer; percentages divide matching counters by evaluated counters from the same report and are not timing metrics. | Stemmer | All exact | Changed exact | Root preserved | Note | | --- | ---: | ---: | ---: | --- | | Radixor | 98.837% | 98.744% | 99.359% | Full Radixor dictionary patch-command stemmer. | +| Lucene HunspellStemFilter | 89.545% | 88.272% | 96.713% | Benchmark-only Polish Hunspell dictionary compared via Lucene HunspellStemFilter. | | Lucene MorfologikFilter | 87.729% | 86.606% | 94.047% | Dictionary-based path; Morfologik can emit multiple terms. | | Lucene StempelFilter | 70.009% | 69.262% | 74.220% | Lucene TokenFilter integration path for table-driven Polish Stempel. | | Lucene StempelStemmer direct | 70.009% | 69.262% | 74.220% | Direct table-driven Polish Stempel stemmer API. | @@ -41,10 +42,11 @@ Speed uses JMH average time, 3 warmup iterations, 5 measurement iterations, 1 fo | Stemmer | Benchmark method | Score ms/op | Error ms | ns/token | Relative vs Radixor | Note | | --- | --- | ---: | ---: | ---: | ---: | --- | -| Radixor | `polishRadixor` | 7.760 | 0.240 | 69.1 | 1.000 | Full Radixor dictionary patch-command stemmer. | -| Lucene StempelStemmer direct | `polishLuceneStempelStemmerDirect` | 34.295 | 0.418 | 305.3 | 4.420 | Direct table-driven Polish Stempel stemmer API. | -| Lucene StempelFilter | `polishLuceneStempelFilter` | 39.116 | 1.717 | 348.2 | 5.041 | Lucene TokenFilter integration path for table-driven Polish Stempel. | -| Lucene MorfologikFilter | `polishLuceneMorfologikFilter` | 128.516 | 12.557 | 1143.9 | 16.562 | Dictionary-based Morfologik TokenFilter; may emit multiple terms. | +| Radixor | `polishRadixor` | 9.049 | 0.485 | 80.5 | 1.000 | Full Radixor dictionary patch-command stemmer. | +| Lucene HunspellStemFilter | `luceneHunspellStemFilter` | 483.316 | 11.455 | 4301.8 | 53.408 | Benchmark-only Polish Hunspell dictionary compared via Lucene HunspellStemFilter. | +| Lucene StempelStemmer direct | `polishLuceneStempelStemmerDirect` | 41.932 | 1.916 | 373.2 | 4.634 | Direct table-driven Polish Stempel stemmer API. | +| Lucene StempelFilter | `polishLuceneStempelFilter` | 45.277 | 13.693 | 403.0 | 5.003 | Lucene TokenFilter integration path for table-driven Polish Stempel. | +| Lucene MorfologikFilter | `polishLuceneMorfologikFilter` | 135.763 | 31.634 | 1208.4 | 15.002 | Dictionary-based Morfologik TokenFilter; may emit multiple terms. | ## Interpretation Notes diff --git a/docs/benchmarks/languages/portuguese.md b/docs/benchmarks/languages/portuguese.md index 1f4f3bb..6b2186f 100644 --- a/docs/benchmarks/languages/portuguese.md +++ b/docs/benchmarks/languages/portuguese.md @@ -26,7 +26,7 @@ Radixor stores the preferred transformation for each normalized dictionary word ## Accuracy -Accuracy is computed from one deterministic JMH measurement iteration without warmup. The benchmark may execute the full dictionary pass more than once inside that single timed iteration; percentages divide matching counters by evaluated counters from the same iteration. +Accuracy is computed from JMH auxiliary counters in the current report. The counters are deterministic for a fixed corpus and stemmer; percentages divide matching counters by evaluated counters from the same report and are not timing metrics. | Stemmer | All exact | Changed exact | Root preserved | Note | | --- | ---: | ---: | ---: | --- | @@ -43,12 +43,12 @@ Speed uses JMH average time, 3 warmup iterations, 5 measurement iterations, 1 fo | Stemmer | Benchmark method | Score ms/op | Error ms | ns/token | Relative vs Radixor | Note | | --- | --- | ---: | ---: | ---: | ---: | --- | -| Radixor | `portugueseRadixor` | 10.598 | 0.273 | 51.1 | 1.000 | Full Radixor dictionary patch-command stemmer. | -| Lucene PortugueseLightStemFilter | `portugueseLucenePortugueseLightStemFilter` | 10.101 | 0.389 | 48.7 | 0.953 | Light Portuguese suffix stemmer. | -| Lucene PortugueseMinimalStemFilter | `portugueseLucenePortugueseMinimalStemFilter` | 14.493 | 0.760 | 69.8 | 1.367 | Minimal Portuguese suffix reducer. | -| Official Snowball direct | `snowballDirect[PORTUGUESE]` | 52.508 | 6.338 | 253.1 | 4.954 | Official Snowball generated Java stemmer; direct API. | -| Lucene SnowballFilter | `luceneSnowballFilter[PORTUGUESE]` | 54.048 | 1.173 | 260.5 | 5.100 | Lucene TokenFilter path around Snowball; includes TokenStream overhead. | -| Lucene PortugueseStemFilter | `portugueseLucenePortugueseStemFilter` | 141.208 | 12.785 | 680.6 | 13.324 | Portuguese RSLP-style Lucene TokenFilter. | +| Radixor | `portugueseRadixor` | 12.109 | 0.698 | 58.4 | 1.000 | Full Radixor dictionary patch-command stemmer. | +| Lucene PortugueseLightStemFilter | `portugueseLucenePortugueseLightStemFilter` | 11.172 | 1.870 | 53.8 | 0.923 | Light Portuguese suffix stemmer. | +| Lucene PortugueseMinimalStemFilter | `portugueseLucenePortugueseMinimalStemFilter` | 16.038 | 1.752 | 77.3 | 1.325 | Minimal Portuguese suffix reducer. | +| Official Snowball direct | `snowballDirect[PORTUGUESE]` | 53.725 | 5.356 | 258.9 | 4.437 | Official Snowball generated Java stemmer; direct API. | +| Lucene SnowballFilter | `luceneSnowballFilter[PORTUGUESE]` | 57.457 | 1.182 | 276.9 | 4.745 | Lucene TokenFilter path around Snowball; includes TokenStream overhead. | +| Lucene PortugueseStemFilter | `portugueseLucenePortugueseStemFilter` | 165.447 | 40.334 | 797.4 | 13.663 | Portuguese RSLP-style Lucene TokenFilter. | ## Interpretation Notes diff --git a/docs/benchmarks/languages/russian.md b/docs/benchmarks/languages/russian.md index 4efe28b..166d89b 100644 --- a/docs/benchmarks/languages/russian.md +++ b/docs/benchmarks/languages/russian.md @@ -26,7 +26,7 @@ Radixor stores the preferred transformation for each normalized dictionary word ## Accuracy -Accuracy is computed from one deterministic JMH measurement iteration without warmup. The benchmark may execute the full dictionary pass more than once inside that single timed iteration; percentages divide matching counters by evaluated counters from the same iteration. +Accuracy is computed from JMH auxiliary counters in the current report. The counters are deterministic for a fixed corpus and stemmer; percentages divide matching counters by evaluated counters from the same report and are not timing metrics. | Stemmer | All exact | Changed exact | Root preserved | Note | | --- | ---: | ---: | ---: | --- | @@ -41,10 +41,10 @@ Speed uses JMH average time, 3 warmup iterations, 5 measurement iterations, 1 fo | Stemmer | Benchmark method | Score ms/op | Error ms | ns/token | Relative vs Radixor | Note | | --- | --- | ---: | ---: | ---: | ---: | --- | -| Radixor | `russianRadixor` | 72.970 | 15.642 | 99.8 | 1.000 | Full Radixor dictionary patch-command stemmer. | -| Lucene RussianLightStemFilter | `russianLuceneRussianLightStemFilter` | 57.900 | 4.404 | 79.2 | 0.793 | Light Russian suffix stemmer. | -| Official Snowball direct | `snowballDirect[RUSSIAN]` | 99.019 | 14.688 | 135.4 | 1.357 | Official Snowball generated Java stemmer; direct API. | -| Lucene SnowballFilter | `luceneSnowballFilter[RUSSIAN]` | 128.272 | 7.201 | 175.4 | 1.758 | Lucene TokenFilter path around Snowball; includes TokenStream overhead. | +| Radixor | `russianRadixor` | 89.671 | 3.886 | 122.6 | 1.000 | Full Radixor dictionary patch-command stemmer. | +| Lucene RussianLightStemFilter | `russianLuceneRussianLightStemFilter` | 60.522 | 5.310 | 82.7 | 0.675 | Light Russian suffix stemmer. | +| Official Snowball direct | `snowballDirect[RUSSIAN]` | 106.031 | 9.287 | 145.0 | 1.182 | Official Snowball generated Java stemmer; direct API. | +| Lucene SnowballFilter | `luceneSnowballFilter[RUSSIAN]` | 137.512 | 10.801 | 188.0 | 1.534 | Lucene TokenFilter path around Snowball; includes TokenStream overhead. | ## Interpretation Notes diff --git a/docs/benchmarks/languages/spanish.md b/docs/benchmarks/languages/spanish.md index 27eb50b..1f1cc1e 100644 --- a/docs/benchmarks/languages/spanish.md +++ b/docs/benchmarks/languages/spanish.md @@ -26,11 +26,12 @@ Radixor stores the preferred transformation for each normalized dictionary word ## Accuracy -Accuracy is computed from one deterministic JMH measurement iteration without warmup. The benchmark may execute the full dictionary pass more than once inside that single timed iteration; percentages divide matching counters by evaluated counters from the same iteration. +Accuracy is computed from JMH auxiliary counters in the current report. The counters are deterministic for a fixed corpus and stemmer; percentages divide matching counters by evaluated counters from the same report and are not timing metrics. | Stemmer | All exact | Changed exact | Root preserved | Note | | --- | ---: | ---: | ---: | --- | | Radixor | 97.459% | 97.544% | 96.891% | Full Radixor dictionary patch-command stemmer. | +| Lucene HunspellStemFilter | 49.074% | 42.656% | 92.154% | Benchmark-only Spanish Hunspell dictionary compared via Lucene HunspellStemFilter. | | Lucene SpanishMinimalStemFilter | 17.284% | 5.347% | 97.403% | Minimal suffix reducer; narrow baseline, not a full stemmer. | | Lucene SpanishPluralStemFilter | 15.140% | 5.802% | 77.820% | Plural-focused suffix reducer; narrow baseline. | | Lucene SpanishLightStemFilter | 9.577% | 7.088% | 26.279% | Light suffix stemmer; intentionally narrower than a dictionary-derived stemmer. | @@ -43,12 +44,13 @@ Speed uses JMH average time, 3 warmup iterations, 5 measurement iterations, 1 fo | Stemmer | Benchmark method | Score ms/op | Error ms | ns/token | Relative vs Radixor | Note | | --- | --- | ---: | ---: | ---: | ---: | --- | -| Radixor | `spanishRadixor` | 64.539 | 3.448 | 80.0 | 1.000 | Full Radixor dictionary patch-command stemmer. | -| Lucene SpanishMinimalStemFilter | `spanishLuceneSpanishMinimalStemFilter` | 38.288 | 2.382 | 47.5 | 0.593 | Minimal Spanish suffix reducer; narrow baseline. | -| Lucene SpanishLightStemFilter | `spanishLuceneSpanishLightStemFilter` | 40.855 | 2.407 | 50.7 | 0.633 | Light Spanish suffix stemmer. | -| Lucene SpanishPluralStemFilter | `spanishLuceneSpanishPluralStemFilter` | 91.054 | 8.822 | 112.9 | 1.411 | Plural-oriented Spanish suffix reducer. | -| Official Snowball direct | `snowballDirect[SPANISH]` | 168.813 | 32.860 | 209.4 | 2.616 | Official Snowball generated Java stemmer; direct API. | -| Lucene SnowballFilter | `luceneSnowballFilter[SPANISH]` | 185.626 | 50.297 | 230.2 | 2.876 | Lucene TokenFilter path around Snowball; includes TokenStream overhead. | +| Radixor | `spanishRadixor` | 78.919 | 7.253 | 97.9 | 1.000 | Full Radixor dictionary patch-command stemmer. | +| Lucene HunspellStemFilter | `luceneHunspellStemFilter` | 2079.041 | 193.548 | 2578.6 | 26.344 | Benchmark-only Spanish Hunspell dictionary compared via Lucene HunspellStemFilter. | +| Lucene SpanishMinimalStemFilter | `spanishLuceneSpanishMinimalStemFilter` | 45.596 | 4.639 | 56.6 | 0.578 | Minimal Spanish suffix reducer; narrow baseline. | +| Lucene SpanishLightStemFilter | `spanishLuceneSpanishLightStemFilter` | 42.003 | 1.683 | 52.1 | 0.532 | Light Spanish suffix stemmer. | +| Lucene SpanishPluralStemFilter | `spanishLuceneSpanishPluralStemFilter` | 93.734 | 6.247 | 116.3 | 1.188 | Plural-oriented Spanish suffix reducer. | +| Official Snowball direct | `snowballDirect[SPANISH]` | 171.995 | 11.035 | 213.3 | 2.179 | Official Snowball generated Java stemmer; direct API. | +| Lucene SnowballFilter | `luceneSnowballFilter[SPANISH]` | 211.138 | 17.940 | 261.9 | 2.675 | Lucene TokenFilter path around Snowball; includes TokenStream overhead. | ## Interpretation Notes diff --git a/docs/benchmarks/languages/swedish.md b/docs/benchmarks/languages/swedish.md index ea55506..7f20312 100644 --- a/docs/benchmarks/languages/swedish.md +++ b/docs/benchmarks/languages/swedish.md @@ -26,7 +26,7 @@ Radixor stores the preferred transformation for each normalized dictionary word ## Accuracy -Accuracy is computed from one deterministic JMH measurement iteration without warmup. The benchmark may execute the full dictionary pass more than once inside that single timed iteration; percentages divide matching counters by evaluated counters from the same iteration. +Accuracy is computed from JMH auxiliary counters in the current report. The counters are deterministic for a fixed corpus and stemmer; percentages divide matching counters by evaluated counters from the same report and are not timing metrics. | Stemmer | All exact | Changed exact | Root preserved | Note | | --- | ---: | ---: | ---: | --- | @@ -42,11 +42,11 @@ Speed uses JMH average time, 3 warmup iterations, 5 measurement iterations, 1 fo | Stemmer | Benchmark method | Score ms/op | Error ms | ns/token | Relative vs Radixor | Note | | --- | --- | ---: | ---: | ---: | ---: | --- | -| Radixor | `swedishRadixor` | 4.916 | 0.525 | 57.3 | 1.000 | Full Radixor dictionary patch-command stemmer. | -| Lucene SwedishMinimalStemFilter | `swedishLuceneSwedishMinimalStemFilter` | 4.453 | 0.530 | 51.9 | 0.906 | Minimal Swedish suffix reducer. | -| Lucene SwedishLightStemFilter | `swedishLuceneSwedishLightStemFilter` | 4.523 | 0.151 | 52.8 | 0.920 | Light Swedish suffix stemmer. | -| Official Snowball direct | `snowballDirect[SWEDISH]` | 7.075 | 0.541 | 82.5 | 1.439 | Official Snowball generated Java stemmer; direct API. | -| Lucene SnowballFilter | `luceneSnowballFilter[SWEDISH]` | 9.379 | 0.056 | 109.4 | 1.908 | Lucene TokenFilter path around Snowball; includes TokenStream overhead. | +| Radixor | `swedishRadixor` | 5.489 | 0.355 | 64.0 | 1.000 | Full Radixor dictionary patch-command stemmer. | +| Lucene SwedishMinimalStemFilter | `swedishLuceneSwedishMinimalStemFilter` | 4.630 | 0.130 | 54.0 | 0.843 | Minimal Swedish suffix reducer. | +| Lucene SwedishLightStemFilter | `swedishLuceneSwedishLightStemFilter` | 4.876 | 0.328 | 56.9 | 0.888 | Light Swedish suffix stemmer. | +| Official Snowball direct | `snowballDirect[SWEDISH]` | 7.517 | 0.072 | 87.7 | 1.370 | Official Snowball generated Java stemmer; direct API. | +| Lucene SnowballFilter | `luceneSnowballFilter[SWEDISH]` | 9.793 | 0.338 | 114.2 | 1.784 | Lucene TokenFilter path around Snowball; includes TokenStream overhead. | ## Interpretation Notes diff --git a/docs/benchmarks/languages/ukrainian.md b/docs/benchmarks/languages/ukrainian.md index b710ff4..f537028 100644 --- a/docs/benchmarks/languages/ukrainian.md +++ b/docs/benchmarks/languages/ukrainian.md @@ -26,11 +26,12 @@ Radixor stores the preferred transformation for each normalized dictionary word ## Accuracy -Accuracy is computed from one deterministic JMH measurement iteration without warmup. The benchmark may execute the full dictionary pass more than once inside that single timed iteration; percentages divide matching counters by evaluated counters from the same iteration. +Accuracy is computed from JMH auxiliary counters in the current report. The counters are deterministic for a fixed corpus and stemmer; percentages divide matching counters by evaluated counters from the same report and are not timing metrics. | Stemmer | All exact | Changed exact | Root preserved | Note | | --- | ---: | ---: | ---: | --- | | Radixor | 99.307% | 99.365% | 99.062% | Full Radixor dictionary patch-command stemmer. | +| Lucene HunspellStemFilter | 86.815% | 83.759% | 99.866% | Benchmark-only Ukrainian Hunspell dictionary compared via Lucene HunspellStemFilter. | | Lucene MorfologikFilter | 92.362% | 90.637% | 99.732% | Dictionary-based path; Morfologik can emit multiple terms. | | Morfologik direct | 92.362% | 90.637% | 99.732% | Direct dictionary lookup; first returned stem is used for quality when no ranking weight is exposed. | @@ -40,9 +41,10 @@ Speed uses JMH average time, 3 warmup iterations, 5 measurement iterations, 1 fo | Stemmer | Benchmark method | Score ms/op | Error ms | ns/token | Relative vs Radixor | Note | | --- | --- | ---: | ---: | ---: | ---: | --- | -| Radixor | `ukrainianRadixor` | 0.605 | 0.056 | 47.4 | 1.000 | Full Radixor dictionary patch-command stemmer. | -| Morfologik direct | `ukrainianMorfologikDirect` | 8.106 | 0.040 | 635.7 | 13.408 | Direct Morfologik dictionary lookup; first returned stem is used for quality. | -| Lucene MorfologikFilter | `ukrainianLuceneMorfologikFilter` | 14.684 | 5.214 | 1151.5 | 24.287 | Dictionary-based Morfologik TokenFilter; may emit multiple terms. | +| Radixor | `ukrainianRadixor` | 0.682 | 0.057 | 53.5 | 1.000 | Full Radixor dictionary patch-command stemmer. | +| Lucene HunspellStemFilter | `luceneHunspellStemFilter` | 43.527 | 1.207 | 3413.3 | 63.799 | Benchmark-only Ukrainian Hunspell dictionary compared via Lucene HunspellStemFilter. | +| Morfologik direct | `ukrainianMorfologikDirect` | 8.680 | 0.073 | 680.7 | 12.723 | Direct Morfologik dictionary lookup; first returned stem is used for quality. | +| Lucene MorfologikFilter | `ukrainianLuceneMorfologikFilter` | 14.575 | 0.248 | 1143.0 | 21.364 | Dictionary-based Morfologik TokenFilter; may emit multiple terms. | ## Interpretation Notes diff --git a/docs/benchmarks/languages/yiddish.md b/docs/benchmarks/languages/yiddish.md index 5637e1c..dcce465 100644 --- a/docs/benchmarks/languages/yiddish.md +++ b/docs/benchmarks/languages/yiddish.md @@ -25,11 +25,11 @@ Radixor stores the preferred transformation for each normalized dictionary word ## Accuracy -Accuracy is computed from one deterministic JMH measurement iteration without warmup. The benchmark may execute the full dictionary pass more than once inside that single timed iteration; percentages divide matching counters by evaluated counters from the same iteration. +Accuracy is computed from JMH auxiliary counters in the current report. The counters are deterministic for a fixed corpus and stemmer; percentages divide matching counters by evaluated counters from the same report and are not timing metrics. | Stemmer | All exact | Changed exact | Root preserved | Note | | --- | ---: | ---: | ---: | --- | -| Radixor | 98.930% | 98.343% | 100.000% | Radixor baseline in the Snowball-language comparison family. | +| Radixor | 98.930% | 98.343% | 100.000% | Full Radixor dictionary patch-command stemmer. | | Lucene SnowballFilter | 2.837% | 2.558% | 3.346% | Lucene TokenFilter integration path around the Snowball algorithm. | | Official Snowball direct | 2.837% | 2.558% | 3.346% | Official Snowball generated Java stemmer; rule-based suffix algorithm. | @@ -39,9 +39,9 @@ Speed uses JMH average time, 3 warmup iterations, 5 measurement iterations, 1 fo | Stemmer | Benchmark method | Score ms/op | Error ms | ns/token | Relative vs Radixor | Note | | --- | --- | ---: | ---: | ---: | ---: | --- | -| Radixor | `radixor[YIDDISH]` | 0.236 | 0.004 | 85.1 | 1.000 | Radixor baseline for the Snowball-language comparison family. | -| Official Snowball direct | `snowballDirect[YIDDISH]` | 1.432 | 0.193 | 515.7 | 6.058 | Official Snowball generated Java stemmer; direct API. | -| Lucene SnowballFilter | `luceneSnowballFilter[YIDDISH]` | 1.595 | 0.068 | 574.6 | 6.749 | Lucene TokenFilter path around Snowball; includes TokenStream overhead. | +| Radixor | `radixor[YIDDISH]` | 0.254 | 0.004 | 50.7 | 1.000 | Full Radixor dictionary patch-command stemmer. | +| Official Snowball direct | `snowballDirect[YIDDISH]` | 1.537 | 0.220 | 307.3 | 6.058 | Official Snowball generated Java stemmer; direct API. | +| Lucene SnowballFilter | `luceneSnowballFilter[YIDDISH]` | 1.714 | 0.120 | 342.8 | 6.756 | Lucene TokenFilter path around Snowball; includes TokenStream overhead. | ## Interpretation Notes diff --git a/docs/benchmarks/reference/candidates.md b/docs/benchmarks/reference/candidates.md index 667d0ac..16536a6 100644 --- a/docs/benchmarks/reference/candidates.md +++ b/docs/benchmarks/reference/candidates.md @@ -4,20 +4,28 @@ Implemented benchmark methods are documented on the per-language pages under [La ## Included Candidate Families -The current benchmark pages include Radixor baselines, Lucene language filters where the language matches a bundled Radixor resource, Lucene Stempel and Morfologik paths where applicable, official Snowball Java stemmers where same-language comparison is available, and selected English-specific non-Lucene baselines such as OpenNLP Porter and Paice/Husk Lancaster. +The current benchmark pages include Radixor baselines, Lucene language filters where the language matches a bundled Radixor resource, Lucene Stempel and Morfologik paths where applicable, official Snowball Java stemmers where same-language comparison is available, benchmark-only CISTEM German stemmer evaluation, benchmark-only Hunspell comparisons, and selected English-specific non-Lucene baselines such as OpenNLP Porter and Paice/Husk Lancaster. + +Benchmark-only Hunspell comparisons use bundled benchmark dictionaries and the Lucene HunspellStemFilter adapter over the selected language token streams. +The CISTEM candidate is implemented in `src/jmh/java/org/egothor/stemmer/benchmark/Cistem.java` and follows the original MIT-licensed upstream implementation from Leonie Weissweiler's CISTEM project. +CISTEM German gold-standard files are not vendored in this repository. The Gradle JMH resource preparation tasks download `goldstandard1.txt` and `goldstandard2.txt` from the upstream CISTEM repository into generated build resources. Direct stemmer APIs and Lucene TokenFilter paths are documented separately on language pages. TokenFilter rows include TokenStream, attribute, and required normalization overhead. Direct rows measure exposed direct APIs. +For the benchmark refresh used in this documentation build: + +- Hunspell families are included in `HunspellStemmerComparisonBenchmark` (speed) and `HunspellStemmerComparisonBenchmarkQuality` (quality for all benchmark languages in this corpus). The legacy + `EnglishHunspellStemmerComparisonBenchmarkQuality` result is retained for continuity. +- CISTEM quality is present in the published per-language results under `GERMAN_CISTEM`. CISTEM speed is present as `germanCistem` in `MultiLanguageStemmerComparisonBenchmark`. + ## Evaluated But Skipped Candidates | Candidate | Language | Link/source | Reason skipped | | --- | --- | --- | --- | | Lucene Arabic, Bulgarian, Bengali, Sorani, Greek, Galician, Hindi, Indonesian, Latvian, Telugu filters | Various | `lucene-analysis-common` | No bundled same-language Radixor resource in this repository snapshot. | | Lucene analyzer-only paths | Multiple | Lucene analyzers | Full analyzers mix tokenization, stop-word handling, and other behavior; direct filters are used where available. | -| Lucene HunspellStemFilter | Multiple | `lucene-analysis-common` | Requires external Hunspell dictionaries not resolved as benchmark-only resources here. | | Lucene StemmerOverrideFilter | Multiple | `lucene-analysis-common` | Override map facility, not a stemmer algorithm. | | Additional Snowball Lovins | English | Official Snowball Java distribution | No Lovins Java stemmer was present in the selected Snowball Java distribution. | | Lemur Project Krovetz Stemmer | English | Lemur Project | Lucene KStem represents the Krovetz-style path without adding separate dependency and license risk. | | Smile Lancaster / Paice-Husk | English | Smile NLP | Smile is large for one stemmer; Paice/Husk is included through a smaller benchmark-only generated path. | -| CISTEM German stemmer | German | `https://github.com/LeonieWeissweiler/CISTEM` | Clean benchmark-only Java integration was not completed in this phase. | | `stemmerEval` reference repository | Multiple | `https://github.com/endredy/stemmerEval` | Used only as a candidate reference; no code or data copied. | diff --git a/docs/benchmarks/reference/english-coverage.md b/docs/benchmarks/reference/english-coverage.md index ca8ffc2..8adf27d 100644 --- a/docs/benchmarks/reference/english-coverage.md +++ b/docs/benchmarks/reference/english-coverage.md @@ -6,16 +6,16 @@ This benchmark is the clearest demonstration of the Radixor quality/speed envelo | Used rows | Actual row ratio | All exact | Changed exact | Root preserved | Speed ms/op | Error ms | ns/token | | ---: | ---: | ---: | ---: | ---: | ---: | ---: | ---: | -| 100% | 100.000% | 97.478% | 97.197% | 97.552% | 23.113 | 7.065 | 109.8 | -| 90% | 90.000% | 97.047% | 94.913% | 97.613% | 21.270 | 9.914 | 101.0 | -| 80% | 80.000% | 96.635% | 92.768% | 97.661% | 19.170 | 6.609 | 91.1 | -| 70% | 70.000% | 96.209% | 90.565% | 97.705% | 20.857 | 6.734 | 99.1 | -| 60% | 60.000% | 95.750% | 88.384% | 97.703% | 14.975 | 1.215 | 71.1 | -| 50% | 50.000% | 95.262% | 86.107% | 97.690% | 15.249 | 1.078 | 72.4 | -| 40% | 40.000% | 94.753% | 83.855% | 97.643% | 15.323 | 2.340 | 72.8 | -| 30% | 30.000% | 94.208% | 81.651% | 97.537% | 16.778 | 2.643 | 79.7 | -| 20% | 20.000% | 93.633% | 79.366% | 97.416% | 18.929 | 3.241 | 89.9 | -| 10% | 10.000% | 92.868% | 76.516% | 97.204% | 19.124 | 1.883 | 90.9 | +| 100% | 100.000% | 97.478% | 97.197% | 97.552% | 28.578 | 7.571 | 135.8 | +| 90% | 90.000% | 97.047% | 94.913% | 97.613% | 26.612 | 9.227 | 126.4 | +| 80% | 80.000% | 96.635% | 92.768% | 97.661% | 23.331 | 8.106 | 110.8 | +| 70% | 70.000% | 96.209% | 90.565% | 97.705% | 22.362 | 1.957 | 106.2 | +| 60% | 60.000% | 95.750% | 88.384% | 97.703% | 16.497 | 2.026 | 78.4 | +| 50% | 50.000% | 95.262% | 86.107% | 97.690% | 16.035 | 0.986 | 76.2 | +| 40% | 40.000% | 94.753% | 83.855% | 97.643% | 16.459 | 0.664 | 78.2 | +| 30% | 30.000% | 94.208% | 81.651% | 97.537% | 19.566 | 0.758 | 92.9 | +| 20% | 20.000% | 93.633% | 79.366% | 97.416% | 14.616 | 0.487 | 69.4 | +| 10% | 10.000% | 92.868% | 76.516% | 97.204% | 18.093 | 3.147 | 86.0 | ## Column Meanings diff --git a/docs/benchmarks/reference/environment.md b/docs/benchmarks/reference/environment.md index 97b0e77..2f73301 100644 --- a/docs/benchmarks/reference/environment.md +++ b/docs/benchmarks/reference/environment.md @@ -4,11 +4,11 @@ The values below are environment-specific and must not be read as universal perf | Item | Value | | --- | --- | -| Benchmark date | 2026-07-03 | -| Focused comparison command family | JMH jar runs limited to `EnglishStemmerComparisonBenchmark`, `MultiLanguageStemmerComparisonBenchmark`, and `SnowballLanguageStemmerComparisonBenchmark`; Radixor exact-root metrics were recomputed deterministically against the same contracted loaders | +| Benchmark date | 2026-07-06 (Europe/Prague) | +| Focused comparison command family | `./gradlew jmh -Pjmh.includes='.*StemmerComparisonBenchmark.*' --no-daemon` | | English coverage command | `./gradlew jmh -Pjmh.includes='.*EnglishRadixorDictionaryCoverageBenchmark.*' --no-daemon` | -| Speed result reports | `build/reports/jmh/contracted/english-comparison.csv`, `multilanguage-speed.csv`, `snowball-language-speed.csv` | -| Accuracy result reports | Deterministic Radixor exact-root pass over bundled dictionaries; non-Radixor quality rows retained from the existing published quality suite | +| Speed result reports | `build/reports/jmh/stemmer-comparison-2026-07-06.csv`, `build/reports/jmh/stemmer-comparison-2026-07-06.txt`, `build/reports/jmh/english-coverage-2026-07-06.csv`, and `build/reports/jmh/english-coverage-2026-07-06.txt` | +| Accuracy result reports | `build/reports/jmh/stemmer-comparison-2026-07-06.csv`, `build/reports/jmh/english-coverage-2026-07-06.csv`, and deterministic Radixor exact-root accounting over the same bundled language corpora | | Final comparison JMH scope | Stemmer comparison benchmarks only; internal `FrequencyTrie*` microbenchmarks were not run | | Coverage JMH scope | English Radixor dictionary coverage benchmark only | | JMH version | 1.37 | @@ -16,15 +16,19 @@ The values below are environment-specific and must not be read as universal perf | Score unit | `ns/op` | | Speed warmup | 3 iterations, 1 s each | | Speed measurement | 5 iterations, 1 s each | -| Accuracy warmup | none for deterministic exact-root accounting | -| Accuracy measurement | 1 deterministic measurement iteration; counters only, not speed interpretation | +| Accuracy warmup | 3 JMH warmup iterations were applied by the Gradle invocation; timing scores from quality methods are not interpreted | +| Accuracy measurement | 5 JMH measurement samples; documentation uses deterministic auxiliary counter ratios from the same report | | Fork count in generated report files | 1 | | Default fork policy for accuracy-only benchmark classes | `@Fork(0)` for future default runs because accuracy counters are deterministic and not interpreted as speed | | Thread count | 1 | -| JVM reported by JMH | OpenJDK 64-Bit Server VM, 25.0.3+9 | +| JVM reported by JMH | JDK 25.0.3, OpenJDK 64-Bit Server VM, 25.0.3+9 | +| Java runtime | OpenJDK Runtime Environment, Red Hat build 25.0.3+9 | | JVM invoker | `/usr/lib/jvm/java-25-openjdk/bin/java` | -| Operating system | Linux 7.0.13-200.fc44.x86_64 | -| CPU | AMD Ryzen 5 7600 6-Core Processor | +| Operating system | Fedora Linux 44 (MATE-Compiz) | +| Kernel | Linux 7.0.12-201.fc44.x86_64 | +| Architecture | x86_64 | +| CPU | AMD Ryzen 5 8600G w/ Radeon 760M Graphics | +| Physical cores | 6 | | Logical CPUs | 12 | ## Contracted Trie Baseline @@ -35,12 +39,10 @@ All Radixor rows in the refreshed benchmark tables use contracted compiled patch Generated local report files for this benchmark update: -- `build/reports/jmh/contracted/english-comparison.csv` -- `build/reports/jmh/contracted/english-comparison.txt` -- `build/reports/jmh/contracted/multilanguage-speed.csv` -- `build/reports/jmh/contracted/multilanguage-speed.txt` -- `build/reports/jmh/contracted/snowball-language-speed.csv` -- `build/reports/jmh/contracted/snowball-language-speed.txt` +- `build/reports/jmh/stemmer-comparison-2026-07-06.csv` +- `build/reports/jmh/stemmer-comparison-2026-07-06.txt` +- `build/reports/jmh/english-coverage-2026-07-06.csv` +- `build/reports/jmh/english-coverage-2026-07-06.txt` JMH TXT and CSV reports are still published as benchmark artifacts. They are not converted into a Porter speed badge. diff --git a/docs/benchmarks/reference/methodology.md b/docs/benchmarks/reference/methodology.md index a92be84..6b13e25 100644 --- a/docs/benchmarks/reference/methodology.md +++ b/docs/benchmarks/reference/methodology.md @@ -56,4 +56,4 @@ rootPreservedPercent = rootPreservedMatches / rootEvaluatedTokens * 100 Morfologik can emit multiple terms for one input token. The quality benchmark uses the first emitted term for exact-root accounting when no ranking weight is exposed. Throughput benchmarks for Morfologik TokenFilter paths consume all emitted terms. -Quality reports intentionally use one deterministic measurement iteration without warmup, because exact-root agreement is not a timing metric and repeated precision passes would only duplicate the same counters. +Quality reports use JMH auxiliary counter rows. Exact-root accounting is deterministic for a fixed corpus and stemmer, so repeated measurement samples duplicate the same counters; documentation uses the counter ratios and does not interpret quality benchmark timing scores. diff --git a/gradle/cistem-benchmarks.gradle b/gradle/cistem-benchmarks.gradle new file mode 100644 index 0000000..1fdcead --- /dev/null +++ b/gradle/cistem-benchmarks.gradle @@ -0,0 +1,69 @@ +def cistemGoldStandardBaseUrl = 'https://raw.githubusercontent.com/LeonieWeissweiler/CISTEM/refs/heads/master/gold_standards' +def cistemGoldStandardFiles = [ + 'goldstandard1.txt', + 'goldstandard2.txt' +] +def cistemGoldStandardDownloadDirectory = layout.buildDirectory.dir('third-party/cistem-gold-standards') +def cistemGoldStandardGeneratedResourcesDirectory = layout.buildDirectory.dir('generated/resources/cistem-gold-standards') + +def cistemGoldStandardDownloadedFiles = cistemGoldStandardFiles.collect { String fileName -> + cistemGoldStandardDownloadDirectory.map { it.file(fileName) } +} + +tasks.register('downloadCistemGoldStandards') { + group = 'build setup' + description = 'Downloads benchmark-only CISTEM German gold standards.' + + outputs.files(cistemGoldStandardDownloadedFiles) + + doLast { + cistemGoldStandardFiles.each { String fileName -> + final File targetFile = cistemGoldStandardDownloadDirectory.get().file(fileName).asFile + targetFile.parentFile.mkdirs() + + if (!targetFile.exists()) { + final URL sourceUrl = new URL("${cistemGoldStandardBaseUrl}/${fileName}") + try { + sourceUrl.withInputStream { inputStream -> + targetFile.withOutputStream { outputStream -> + outputStream << inputStream + } + } + } catch (FileNotFoundException exception) { + throw new GradleException( + "Unable to download CISTEM gold standard ${fileName} from ${sourceUrl}.", + exception) + } + } + + if (targetFile.length() <= 0L) { + throw new GradleException("Downloaded CISTEM gold standard ${fileName} was empty.") + } + } + } +} + +tasks.register('prepareCistemGoldStandardResources', Copy) { + group = 'build setup' + description = 'Copies benchmark-only CISTEM German gold standards into the JMH resource output.' + + dependsOn(tasks.named('downloadCistemGoldStandards')) + + from(cistemGoldStandardDownloadDirectory) { + include 'goldstandard1.txt' + include 'goldstandard2.txt' + } + into(cistemGoldStandardGeneratedResourcesDirectory) +} + +sourceSets { + jmh { + resources { + srcDir(cistemGoldStandardGeneratedResourcesDirectory) + } + } +} + +tasks.named('processJmhResources') { + dependsOn(tasks.named('prepareCistemGoldStandardResources')) +} diff --git a/src/jmh/java/org/egothor/stemmer/benchmark/Cistem.java b/src/jmh/java/org/egothor/stemmer/benchmark/Cistem.java new file mode 100644 index 0000000..7d02b59 --- /dev/null +++ b/src/jmh/java/org/egothor/stemmer/benchmark/Cistem.java @@ -0,0 +1,178 @@ +/******************************************************************************* + * MIT License + * + * Copyright (c) 2017 Leonie Weißweiler + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + * + * Source: CISTEM German stemmer + * Authors: Leonie Weissweiler, Alexander Fraser + * https://github.com/LeonieWeissweiler/CISTEM + * https://www.cis.lmu.de/~weissweiler/cistem/ + ******************************************************************************/ +package org.egothor.stemmer.benchmark; + +import java.util.regex.Pattern; + +public final class Cistem { + + private static final Pattern GE_PATTERN = Pattern.compile("^ge(.{4,})"); + private static final Pattern DOLLAR1_PATTERN = Pattern.compile("(.)\\1"); + private static final Pattern ND_PATTERN = Pattern.compile("nd$"); + private static final Pattern EMR_PATTERN = Pattern.compile("e[mr]$"); + private static final Pattern T_PATTERN = Pattern.compile("t$"); + private static final Pattern ESN_PATTERN = Pattern.compile("[esn]$"); + private static final Pattern STAR_PATTERN = Pattern.compile("(.)\\*"); + + private Cistem() { + } + + public static String stem(final String word) { + return stem(word, false); + } + + public static String stem(final String word, final boolean caseInsensitive) { + if (word.isEmpty()) { + return word; + } + + String normalized = word; + normalized = normalized.replace("Ü", "U"); + normalized = normalized.replace("Ö", "O"); + normalized = normalized.replace("Ä", "A"); + normalized = normalized.replace("ü", "u"); + normalized = normalized.replace("ö", "o"); + normalized = normalized.replace("ä", "a"); + + final boolean uppercase = Character.isUpperCase(normalized.charAt(0)); + + normalized = normalized.toLowerCase(); + normalized = normalized.replace("ß", "ss"); + normalized = GE_PATTERN.matcher(normalized).replaceAll("$1"); + normalized = normalized.replace("sch", "$"); + normalized = normalized.replace("ei", "%"); + normalized = normalized.replace("ie", "&"); + normalized = DOLLAR1_PATTERN.matcher(normalized).replaceAll("$1*"); + + while (normalized.length() > 3) { + if (normalized.length() > 5) { + String newWord = EMR_PATTERN.matcher(normalized).replaceAll(""); + if (!normalized.equals(newWord)) { + normalized = newWord; + continue; + } + + newWord = ND_PATTERN.matcher(normalized).replaceAll(""); + if (!normalized.equals(newWord)) { + normalized = newWord; + continue; + } + } + + if (!uppercase || caseInsensitive) { + final String newWord = T_PATTERN.matcher(normalized).replaceAll(""); + if (!normalized.equals(newWord)) { + normalized = newWord; + continue; + } + } + + final String newWord = ESN_PATTERN.matcher(normalized).replaceAll(""); + if (!normalized.equals(newWord)) { + normalized = newWord; + } else { + break; + } + } + + normalized = STAR_PATTERN.matcher(normalized).replaceAll("$1$1"); + normalized = normalized.replace("&", "ie"); + normalized = normalized.replace("%", "ei"); + normalized = normalized.replace("$", "sch"); + + return normalized; + } + + public static String[] segment(final String word) { + return segment(word, false); + } + + public static String[] segment(final String word, final boolean caseInsensitive) { + if (word.isEmpty()) { + return new String[] {"", ""}; + } + + int restLength = 0; + final boolean uppercase = Character.isUpperCase(word.charAt(0)); + String normalized = word.toLowerCase(); + final String original = new String(normalized); + + normalized = normalized.replace("sch", "$"); + normalized = normalized.replace("ei", "%"); + normalized = normalized.replace("ie", "&"); + normalized = DOLLAR1_PATTERN.matcher(normalized).replaceAll("$1*"); + + while (normalized.length() > 3) { + if (normalized.length() > 5) { + String newWord = normalized.replaceAll("e[mr]$", ""); + if (!normalized.equals(newWord)) { + restLength += 2; + normalized = newWord; + continue; + } + + newWord = normalized.replaceAll("nd$", ""); + if (!normalized.equals(newWord)) { + restLength += 2; + normalized = newWord; + continue; + } + } + + if (!uppercase || caseInsensitive) { + final String newWord = normalized.replaceAll("t$", ""); + if (!normalized.equals(newWord)) { + restLength += 1; + normalized = newWord; + continue; + } + } + + final String newWord = normalized.replaceAll("[esn]$", ""); + if (!normalized.equals(newWord)) { + restLength += 1; + normalized = newWord; + } else { + break; + } + } + + normalized = normalized.replaceAll("(.)\\*", "$1$1"); + normalized = normalized.replace("&", "ie"); + normalized = normalized.replace("%", "ei"); + normalized = normalized.replace("$", "sch"); + + String rest = ""; + if (restLength != 0) { + rest = original.substring(original.length() - restLength); + } + + return new String[] {normalized, rest}; + } +} diff --git a/src/jmh/java/org/egothor/stemmer/benchmark/GermanGoldstandardStemmerComparisonBenchmark.java b/src/jmh/java/org/egothor/stemmer/benchmark/GermanGoldstandardStemmerComparisonBenchmark.java new file mode 100644 index 0000000..ff787a9 --- /dev/null +++ b/src/jmh/java/org/egothor/stemmer/benchmark/GermanGoldstandardStemmerComparisonBenchmark.java @@ -0,0 +1,579 @@ +/******************************************************************************* + * Copyright (C) 2026, Leo Galambos + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright notice, + * this list of conditions and the following disclaimer. + * + * 2. Redistributions in binary form must reproduce the above copyright notice, + * this list of conditions and the following disclaimer in the documentation + * and/or other materials provided with the distribution. + * + * 3. Neither the name of the copyright holder nor the names of its contributors + * may be used to endorse or promote products derived from this software + * without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + ******************************************************************************/ +package org.egothor.stemmer.benchmark; + +import java.io.BufferedReader; +import java.io.IOException; +import java.io.InputStream; +import java.io.InputStreamReader; +import java.nio.charset.StandardCharsets; +import java.util.ArrayList; +import java.util.LinkedHashMap; +import java.util.List; +import java.util.Map; +import java.util.Objects; +import java.util.function.Function; + +import org.apache.lucene.analysis.LowerCaseFilter; +import org.apache.lucene.analysis.TokenStream; +import org.apache.lucene.analysis.de.GermanLightStemFilter; +import org.apache.lucene.analysis.de.GermanMinimalStemFilter; +import org.apache.lucene.analysis.de.GermanNormalizationFilter; +import org.apache.lucene.analysis.de.GermanStemFilter; +import org.apache.lucene.analysis.snowball.SnowballFilter; +import org.apache.lucene.analysis.tokenattributes.CharTermAttribute; +import org.apache.lucene.analysis.tokenattributes.PositionIncrementAttribute; +import org.openjdk.jmh.annotations.AuxCounters; +import org.openjdk.jmh.annotations.Benchmark; +import org.openjdk.jmh.annotations.BenchmarkMode; +import org.openjdk.jmh.annotations.Fork; +import org.openjdk.jmh.annotations.Level; +import org.openjdk.jmh.annotations.Measurement; +import org.openjdk.jmh.annotations.Mode; +import org.openjdk.jmh.annotations.OutputTimeUnit; +import org.openjdk.jmh.annotations.Param; +import org.openjdk.jmh.annotations.Scope; +import org.openjdk.jmh.annotations.Setup; +import org.openjdk.jmh.annotations.State; +import org.openjdk.jmh.annotations.Warmup; +import org.openjdk.jmh.infra.Blackhole; + +import org.egothor.stemmer.FrequencyTrie; +import org.egothor.stemmer.ReductionMode; +import org.egothor.stemmer.StemmerPatchTrieLoader; + +/** + * German-only stemmer comparison on CISTEM gold standards. + * + *

+ * Each benchmark operation is fed by one cluster file. The same candidate set is + * evaluated twice, once per file, to produce one precision/recall/f-measure + * table for each gold standard. + *

+ */ +@BenchmarkMode(Mode.AverageTime) +@OutputTimeUnit(java.util.concurrent.TimeUnit.NANOSECONDS) +@Warmup(iterations = 3, time = 1, timeUnit = java.util.concurrent.TimeUnit.SECONDS) +@Measurement(iterations = 5, time = 1, timeUnit = java.util.concurrent.TimeUnit.SECONDS) +@Fork(1) +public class GermanGoldstandardStemmerComparisonBenchmark { + + /** + * Shared German benchmark state for one dataset and one candidate. + */ + @State(Scope.Benchmark) + public static class SharedState { + + /** + * Gold standard dataset. + */ + @Param({"goldstandard1.txt", "goldstandard2.txt"}) + public String goldStandardFileName; + + /** + * Candidate stemmer. + */ + @Param({ + "GERMAN_RADIXOR", + "GERMAN_LUCENE_GERMAN_STEM_FILTER", + "GERMAN_LUCENE_GERMAN_LIGHT_STEM_FILTER", + "GERMAN_LUCENE_GERMAN_MINIMAL_STEM_FILTER", + "GERMAN_CISTEM", + "SNOWBALL_GERMAN_DIRECT", + "SNOWBALL_GERMAN_LUCENE_FILTER" + }) + public String candidateName; + + /** + * Parsed gold standard corpus. + */ + private GermanGoldstandardCorpus corpus; + + /** + * Gold standard words flattened by cluster order. + */ + private String[] allTokens; + + /** + * Candidate evaluator. + */ + private GoldstandardStemmer stemmer; + + /** + * Initializes one candidate on one gold standard corpus. + * + * @throws IOException when the corpus cannot be loaded + */ + @Setup(Level.Trial) + public void setUp() throws IOException { + this.corpus = loadCorpus(this.goldStandardFileName); + this.allTokens = flattenCorpusTokens(this.corpus); + this.stemmer = GermanCandidate.valueOf(this.candidateName).createEvaluator(); + } + } + + /** + * JMH auxiliary counters for CISTEM-style cluster accounting. + */ + @State(Scope.Thread) + @AuxCounters(AuxCounters.Type.EVENTS) + public static class GoldstandardQualityCounters { + + /** + * True positives across clusters. + */ + public long truePositives; + + /** + * False positives across clusters. + */ + public long falsePositives; + + /** + * False negatives across clusters. + */ + public long falseNegatives; + + /** + * Evaluated clusters. + */ + public long evaluatedClusters; + + /** + * Evaluated tokens. + */ + public long evaluatedTokens; + + /** + * Resets counters before each measured iteration. + */ + @Setup(Level.Iteration) + public void reset() { + this.truePositives = 0L; + this.falsePositives = 0L; + this.falseNegatives = 0L; + this.evaluatedClusters = 0L; + this.evaluatedTokens = 0L; + } + } + + /** + * Evaluates CISTEM-style precision, recall, and F1-relevant counts. + * + * @param state shared benchmark state + * @param counters quality counters + * @param blackhole result sink + * @return evaluated token count for this operation + * @throws IOException if token filtering cannot run + */ + @Benchmark + @Warmup(iterations = 0) + @Measurement(iterations = 1, time = 1, timeUnit = java.util.concurrent.TimeUnit.MILLISECONDS) + @Fork(0) + public long cistemStyleQuality(final SharedState state, final GoldstandardQualityCounters counters, + final Blackhole blackhole) throws IOException { + final GoldstandardResult result = evaluateCistemStyle(state.corpus, state.allTokens, state.stemmer, blackhole); + counters.truePositives += result.truePositives(); + counters.falsePositives += result.falsePositives(); + counters.falseNegatives += result.falseNegatives(); + counters.evaluatedClusters += result.evaluatedClusters(); + counters.evaluatedTokens += result.evaluatedTokens(); + return result.evaluatedTokens(); + } + + /** + * Benchmarks candidate throughput over the selected gold standard. + * + * @param state shared benchmark state + * @param blackhole result sink + * @throws IOException if token filtering cannot run + */ + @Benchmark + public void cistemStyleSpeed(final SharedState state, final Blackhole blackhole) throws IOException { + state.stemmer.stem(state.allTokens, blackhole); + } + + /** + * Named German candidates used for the CISTEM gold-standard comparison. + */ + private enum GermanCandidate { + GERMAN_RADIXOR, + GERMAN_LUCENE_GERMAN_STEM_FILTER, + GERMAN_LUCENE_GERMAN_LIGHT_STEM_FILTER, + GERMAN_LUCENE_GERMAN_MINIMAL_STEM_FILTER, + GERMAN_CISTEM, + SNOWBALL_GERMAN_DIRECT, + SNOWBALL_GERMAN_LUCENE_FILTER; + + /** + * Creates a candidate evaluator. + * + * @return stemmer evaluator + * @throws IOException if trie resources cannot be loaded + */ + GoldstandardStemmer createEvaluator() throws IOException { + return switch (this) { + case GERMAN_RADIXOR -> direct(createGermanRadixorStemmer()); + case GERMAN_LUCENE_GERMAN_STEM_FILTER -> + tokenFilter(input -> new GermanStemFilter(lowercase(input))); + case GERMAN_LUCENE_GERMAN_LIGHT_STEM_FILTER -> + tokenFilter(input -> new GermanLightStemFilter(germanNormalize(input))); + case GERMAN_LUCENE_GERMAN_MINIMAL_STEM_FILTER -> + tokenFilter(input -> new GermanMinimalStemFilter(germanNormalize(input))); + case GERMAN_CISTEM -> direct(Cistem::stem); + case SNOWBALL_GERMAN_DIRECT -> direct(SnowballLanguageCase.GERMAN.createDirectStemmer()::stem); + case SNOWBALL_GERMAN_LUCENE_FILTER -> + tokenFilter(input -> new SnowballFilter(new LowerCaseFilter(input), + SnowballLanguageCase.GERMAN.luceneSnowballName())); + }; + } + } + + /** + * Evaluates one full corpus through CISTEM-style cluster scoring. + * + *

+ * For each cluster, the most frequent predicted stem is considered the + * cluster main stem. TP are cluster words mapped to this stem, FN are + * words mapped elsewhere inside the same cluster, and FP are words from + * other clusters mapped to the same main stem. + *

+ * + * @param corpus parsed gold standard corpus + * @param allTokens flattened token sequence + * @param stemmer candidate stemmer + * @param blackhole result sink + * @return aggregated TP/FP/FN counters and token metrics + * @throws IOException when token filtering cannot run + */ + private static GoldstandardResult evaluateCistemStyle(final GermanGoldstandardCorpus corpus, + final String[] allTokens, final GoldstandardStemmer stemmer, final Blackhole blackhole) throws IOException { + final String[] predicted = stemmer.stem(allTokens, blackhole); + final Map globalPredictions = new LinkedHashMap<>(); + for (int index = 0; index < allTokens.length; index++) { + final String prediction = normalizePrediction(predicted[index], allTokens[index]); + globalPredictions.put(prediction, globalPredictions.getOrDefault(prediction, 0) + 1); + } + + long truePositives = 0L; + long falsePositives = 0L; + long falseNegatives = 0L; + int tokenOffset = 0; + for (final String[] cluster : corpus.clusters()) { + if (cluster.length == 0) { + continue; + } + + final Map localPredictions = new LinkedHashMap<>(); + for (int index = 0; index < cluster.length; index++) { + final int tokenIndex = tokenOffset + index; + final String word = allTokens[tokenIndex]; + final String prediction = normalizePrediction(predicted[tokenIndex], word); + localPredictions.put(prediction, localPredictions.getOrDefault(prediction, 0) + 1); + } + + final String mainStem = mostFrequent(localPredictions); + final int predictedAsMain = localPredictions.get(mainStem); + final int clusterSize = cluster.length; + final int falseNegative = clusterSize - predictedAsMain; + final int falsePositive = globalPredictions.get(mainStem) - predictedAsMain; + + truePositives += predictedAsMain; + falseNegatives += falseNegative; + falsePositives += falsePositive; + tokenOffset += clusterSize; + } + + return new GoldstandardResult(truePositives, falsePositives, falseNegatives, corpus.clusters().length, + allTokens.length); + } + + /** + * Creates a direct evaluator. + * + * @param stemmer direct word stemmer + * @return evaluator + */ + private static GoldstandardStemmer direct(final Stemmer stemmer) { + Objects.requireNonNull(stemmer, "stemmer"); + return (tokens, blackhole) -> { + final String[] outputs = new String[tokens.length]; + for (int index = 0; index < tokens.length; index++) { + final String output = stemmer.stem(tokens[index]); + outputs[index] = output; + blackhole.consume(output); + } + return outputs; + }; + } + + /** + * Creates a TokenFilter evaluator. + * + * @param factory filter stream factory + * @return evaluator + */ + private static GoldstandardStemmer tokenFilter(final Function factory) { + Objects.requireNonNull(factory, "factory"); + return (tokens, blackhole) -> firstTokenFilterOutputs(tokens, factory, blackhole); + } + + /** + * Loads and parses one gold standard file from generated JMH resources. + * + * @param resourceName gold standard file name + * @return parsed corpus + * @throws IOException if reading fails + */ + private static GermanGoldstandardCorpus loadCorpus(final String resourceName) throws IOException { + final ClassLoader classLoader = GermanGoldstandardStemmerComparisonBenchmark.class.getClassLoader(); + final InputStream resourceStream = classLoader.getResourceAsStream(resourceName); + if (resourceStream == null) { + throw new IllegalStateException("Missing generated CISTEM gold standard resource: " + resourceName + + ". Run the Gradle JMH resource preparation task to download benchmark-only inputs."); + } + try (InputStream input = resourceStream) { + return parseCorpus(new BufferedReader(new InputStreamReader(input, StandardCharsets.UTF_8))); + } + } + + /** + * Parses CISTEM gold standard format into clustered candidates. + * + * @param reader UTF-8 reader + * @return parsed corpus + * @throws IOException if input cannot be read + */ + private static GermanGoldstandardCorpus parseCorpus(final BufferedReader reader) throws IOException { + final List clusters = new ArrayList<>(); + String line = reader.readLine(); + while (line != null) { + final String trimmed = line.trim(); + if (!trimmed.isEmpty()) { + final String[] words = trimmed.split("\\s+"); + if (words.length > 0) { + clusters.add(words); + } + } + line = reader.readLine(); + } + return new GermanGoldstandardCorpus(clusters.toArray(String[][]::new)); + } + + /** + * Flattens the corpus in deterministic cluster order. + * + * @param corpus corpus to flatten + * @return flattened token array + */ + private static String[] flattenCorpusTokens(final GermanGoldstandardCorpus corpus) { + int total = 0; + for (final String[] cluster : corpus.clusters()) { + total += cluster.length; + } + final String[] tokens = new String[total]; + int index = 0; + for (final String[] cluster : corpus.clusters()) { + System.arraycopy(cluster, 0, tokens, index, cluster.length); + index += cluster.length; + } + return tokens; + } + + /** + * Returns the most frequent key; insertion order is preserved on ties. + * + * @param frequencies predicted stem frequencies + * @return most frequent stem + */ + private static String mostFrequent(final Map frequencies) { + String best = null; + int bestCount = -1; + for (final Map.Entry entry : frequencies.entrySet()) { + if (entry.getValue() > bestCount) { + best = entry.getKey(); + bestCount = entry.getValue(); + } + } + return best; + } + + /** + * Normalizes a null/empty prediction using the input token as fallback. + * + * @param prediction stemmed token + * @param fallback fallback token + * @return safe prediction + */ + private static String normalizePrediction(final String prediction, final String fallback) { + if (prediction == null || prediction.isEmpty()) { + return fallback; + } + return prediction; + } + + /** + * Applies one TokenFilter to all input tokens and returns the first emitted term + * for each input token. + * + * @param tokens input token corpus + * @param factory TokenFilter factory + * @param blackhole result sink + * @return first emitted term per input token + * @throws IOException if token streaming fails + */ + private static String[] firstTokenFilterOutputs(final String[] tokens, final Function factory, + final Blackhole blackhole) throws IOException { + final String[] outputs = new String[tokens.length]; + final BenchmarkTokenStream input = new BenchmarkTokenStream(tokens); + final TokenStream output = factory.apply(input); + final CharTermAttribute termAttribute = output.addAttribute(CharTermAttribute.class); + final PositionIncrementAttribute positionAttribute = output.addAttribute(PositionIncrementAttribute.class); + + int inputIndex = -1; + boolean recordedForPosition = false; + output.reset(); + while (output.incrementToken()) { + final int positionIncrement = positionAttribute.getPositionIncrement(); + if (positionIncrement > 0) { + inputIndex += positionIncrement; + recordedForPosition = false; + } + if (inputIndex >= 0 && inputIndex < outputs.length && !recordedForPosition) { + outputs[inputIndex] = termAttribute.toString(); + blackhole.consume(termAttribute); + recordedForPosition = true; + } + } + output.end(); + output.close(); + + for (int index = 0; index < outputs.length; index++) { + if (outputs[index] == null) { + outputs[index] = tokens[index]; + } + } + return outputs; + } + + /** + * Creates a direct Radixor evaluator using the contracted dictionary trie. + * + * @return direct Radixor stemmer + * @throws IOException if the trie cannot be loaded + */ + private static Stemmer createGermanRadixorStemmer() throws IOException { + return new RadixorBenchmarkStemmer(StemmerPatchTrieLoader.loadCompiled( + StemmerPatchTrieLoader.Language.DE_DE, true, + ReductionMode.MERGE_SUBTREES_WITH_EQUIVALENT_RANKED_GET_ALL_RESULTS))::stem; + } + + /** + * Adds Lucene lower-case normalization. + * + * @param input token stream + * @return normalized token stream + */ + private static TokenStream lowercase(final TokenStream input) { + return new LowerCaseFilter(input); + } + + /** + * Adds Lucene German normalization for light and minimal filters. + * + * @param input token stream + * @return normalized token stream + */ + private static TokenStream germanNormalize(final TokenStream input) { + return new GermanNormalizationFilter(lowercase(input)); + } + + /** + * Direct or filter stemmer adapter used by this benchmark. + */ + @FunctionalInterface + private interface GoldstandardStemmer { + + /** + * Runs one complete token list. + * + * @param tokens input tokens + * @param blackhole result sink + * @return per-token outputs + * @throws IOException if filter processing fails + */ + String[] stem(String[] tokens, Blackhole blackhole) throws IOException; + } + + /** + * Deterministic direct word stem function. + */ + @FunctionalInterface + private interface Stemmer { + + /** + * Stems one token. + * + * @param token input token + * @return stemmed token + */ + String stem(String token); + } + + /** + * Immutable parsed CISTEM gold standard corpus. + */ + private static final class GermanGoldstandardCorpus { + + private final String[][] clusters; + + GermanGoldstandardCorpus(final String[][] clusters) { + this.clusters = clusters; + } + + String[][] clusters() { + return this.clusters; + } + } + + /** + * Aggregated quality result for one benchmark operation. + * + * @param truePositives true positives + * @param falsePositives false positives + * @param falseNegatives false negatives + * @param evaluatedClusters evaluated clusters + * @param evaluatedTokens evaluated tokens + */ + private record GoldstandardResult(long truePositives, long falsePositives, long falseNegatives, + long evaluatedClusters, long evaluatedTokens) { + } +} diff --git a/src/jmh/java/org/egothor/stemmer/benchmark/HunspellStemmerComparisonBenchmarkQuality.java b/src/jmh/java/org/egothor/stemmer/benchmark/HunspellStemmerComparisonBenchmarkQuality.java new file mode 100644 index 0000000..789746a --- /dev/null +++ b/src/jmh/java/org/egothor/stemmer/benchmark/HunspellStemmerComparisonBenchmarkQuality.java @@ -0,0 +1,387 @@ +/******************************************************************************* + * Copyright (C) 2026, Leo Galambos + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright notice, + * this list of conditions and the following disclaimer. + * + * 2. Redistributions in binary form must reproduce the above copyright notice, + * this list of conditions and the following disclaimer in the documentation + * and/or other materials provided with the distribution. + * + * 3. Neither the name of the copyright holder nor the names of its contributors + * may be used to endorse or promote products derived from this software + * without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + ******************************************************************************/ +package org.egothor.stemmer.benchmark; + +import java.io.IOException; +import java.io.InputStream; +import java.text.ParseException; +import java.util.List; +import java.util.Locale; +import java.util.Objects; +import java.util.concurrent.TimeUnit; + +import org.apache.lucene.analysis.LowerCaseFilter; +import org.apache.lucene.analysis.TokenStream; +import org.apache.lucene.analysis.hunspell.Dictionary; +import org.apache.lucene.analysis.hunspell.HunspellStemFilter; +import org.apache.lucene.analysis.hunspell.SortingStrategy; +import org.apache.lucene.analysis.tokenattributes.CharTermAttribute; +import org.apache.lucene.analysis.tokenattributes.PositionIncrementAttribute; +import org.egothor.stemmer.StemmerPatchTrieLoader; +import org.openjdk.jmh.annotations.AuxCounters; +import org.openjdk.jmh.annotations.Benchmark; +import org.openjdk.jmh.annotations.BenchmarkMode; +import org.openjdk.jmh.annotations.Fork; +import org.openjdk.jmh.annotations.Level; +import org.openjdk.jmh.annotations.Measurement; +import org.openjdk.jmh.annotations.Mode; +import org.openjdk.jmh.annotations.OutputTimeUnit; +import org.openjdk.jmh.annotations.Param; +import org.openjdk.jmh.annotations.Scope; +import org.openjdk.jmh.annotations.Setup; +import org.openjdk.jmh.annotations.State; +import org.openjdk.jmh.annotations.Warmup; +import org.openjdk.jmh.infra.Blackhole; + +/** + * Emits exact-root agreement metrics for the benchmark-only Hunspell comparisons. + * + *

+ * This class mirrors the existing Hunspell throughput setup but adds + * quality-style accuracy counters for every Hunspell language dictionary used + * in benchmark-only throughput comparisons. + *

+ */ +@BenchmarkMode(Mode.AverageTime) +@OutputTimeUnit(TimeUnit.NANOSECONDS) +@Warmup(iterations = 0) +@Measurement(iterations = 1, time = 1, timeUnit = TimeUnit.MILLISECONDS) +@Fork(0) +public class HunspellStemmerComparisonBenchmarkQuality { + + /** + * Shared quality corpus and Hunspell dictionary for a selected language. + */ + @State(Scope.Benchmark) + public static class SharedState { + + /** + * Selected language case. + */ + @Param({ "ENGLISH", "CZECH", "GERMAN", "SPANISH", "FRENCH", "DUTCH", "POLISH", "UKRAINIAN" }) + public String languageCaseName; + + /** + * Selected language descriptor. + */ + private HunspellLanguageCase languageCase; + + /** + * Complete language dictionary corpus and expected roots. + */ + private LanguageBenchmarkCorpus.Corpus corpus; + + /** + * Parsed benchmark-only Hunspell dictionary. + */ + private Dictionary dictionary; + + /** + * Initializes quality resources. + * + * @throws IOException if corpus or dictionary loading fails + * @throws ParseException if the Hunspell dictionary cannot be parsed + */ + @Setup(Level.Trial) + public void setUp() throws IOException, ParseException { + this.languageCase = HunspellLanguageCase.valueOf(this.languageCaseName); + this.corpus = LanguageBenchmarkCorpus.createFullCorpus(this.languageCase.radixorLanguage()); + this.dictionary = loadDictionary(this.languageCase); + } + } + + /** + * JMH auxiliary counters for exact-root agreement. + */ + @State(Scope.Thread) + @AuxCounters(AuxCounters.Type.EVENTS) + public static class AccuracyCounters { + + /** + * Number of exact-root matches. + */ + public long correctMatches; + + /** + * Number of evaluated tokens. + */ + public long evaluatedTokens; + + /** + * Number of exact-root matches where the input token differs from the + * expected root. + */ + public long changedCorrectMatches; + + /** + * Number of evaluated tokens where the input token differs from the expected + * root. + */ + public long changedEvaluatedTokens; + + /** + * Number of exact-root matches where the input token is already the expected + * root. + */ + public long rootPreservedMatches; + + /** + * Number of evaluated tokens where the input token is already the expected + * root. + */ + public long rootEvaluatedTokens; + + /** + * Resets counters before each measured iteration. + */ + @Setup(Level.Iteration) + public void reset() { + this.correctMatches = 0L; + this.evaluatedTokens = 0L; + this.changedCorrectMatches = 0L; + this.changedEvaluatedTokens = 0L; + this.rootPreservedMatches = 0L; + this.rootEvaluatedTokens = 0L; + } + } + + /** + * Evaluates exact-root agreement for the selected Hunspell dictionary. + * + * @param sharedState shared quality state + * @param counters JMH auxiliary counters + * @param blackhole result sink + * @return exact-root match count + * @throws IOException if Lucene token streaming fails + */ + @Benchmark + public int luceneHunspellStemFilterAccuracy(final SharedState sharedState, final AccuracyCounters counters, + final Blackhole blackhole) throws IOException { + final String[] actualStems = firstHunspellOutputs(sharedState.corpus.tokens(), sharedState.dictionary, + blackhole); + final String[] tokens = sharedState.corpus.tokens(); + final String[] expectedRoots = sharedState.corpus.expectedRoots(); + + int correct = 0; + int changedCorrect = 0; + int changedEvaluated = 0; + int rootPreserved = 0; + int rootEvaluated = 0; + for (int index = 0; index < actualStems.length; index++) { + final String token = tokens[index]; + final String expectedRoot = expectedRoots[index]; + final boolean exact = Objects.equals(expectedRoot, actualStems[index]); + if (exact) { + correct++; + } + if (Objects.equals(token, expectedRoot)) { + rootEvaluated++; + if (exact) { + rootPreserved++; + } + } else { + changedEvaluated++; + if (exact) { + changedCorrect++; + } + } + } + + counters.correctMatches += correct; + counters.evaluatedTokens += actualStems.length; + counters.changedCorrectMatches += changedCorrect; + counters.changedEvaluatedTokens += changedEvaluated; + counters.rootPreservedMatches += rootPreserved; + counters.rootEvaluatedTokens += rootEvaluated; + return correct; + } + + /** + * Extracts the first emitted Hunspell stem for each input token. + * + * @param tokens token corpus + * @param dictionary Hunspell dictionary + * @param blackhole result sink + * @return first emitted term per input token + * @throws IOException if Lucene streaming fails + */ + private static String[] firstHunspellOutputs(final String[] tokens, final Dictionary dictionary, + final Blackhole blackhole) throws IOException { + final String[] outputs = new String[tokens.length]; + final BenchmarkTokenStream input = new BenchmarkTokenStream(tokens); + final TokenStream output = new HunspellStemFilter(new LowerCaseFilter(input), dictionary, true); + final CharTermAttribute termAttribute = output.addAttribute(CharTermAttribute.class); + final PositionIncrementAttribute positionAttribute = output.addAttribute(PositionIncrementAttribute.class); + int inputIndex = -1; + boolean recordedForPosition = false; + + output.reset(); + while (output.incrementToken()) { + final int positionIncrement = positionAttribute.getPositionIncrement(); + if (positionIncrement > 0) { + inputIndex += positionIncrement; + recordedForPosition = false; + } + if (inputIndex >= 0 && inputIndex < outputs.length && !recordedForPosition) { + outputs[inputIndex] = termAttribute.toString(); + recordedForPosition = true; + } + blackhole.consume(termAttribute); + } + output.end(); + output.close(); + + for (int index = 0; index < outputs.length; index++) { + if (outputs[index] == null) { + outputs[index] = tokens[index]; + } + } + return outputs; + } + + /** + * Loads a benchmark-only Hunspell dictionary from generated resources. + * + * @param languageCase selected language case + * @return parsed dictionary + * @throws IOException if dictionary resources cannot be read + * @throws ParseException if dictionary parsing fails + */ + private static Dictionary loadDictionary(final HunspellLanguageCase languageCase) throws IOException, + ParseException { + final ClassLoader classLoader = HunspellStemmerComparisonBenchmarkQuality.class.getClassLoader(); + final String basePath = "hunspell/" + languageCase.hunspellResourceCode() + "/index."; + try (InputStream affixStream = openRequiredResource(classLoader, basePath + "aff"); + InputStream dictionaryStream = openRequiredResource(classLoader, basePath + "dic")) { + return new Dictionary(affixStream, List.of(dictionaryStream), true, SortingStrategy.inMemory()); + } + } + + /** + * Opens a required classpath resource. + * + * @param classLoader class loader + * @param path resource path + * @return resource stream + */ + private static InputStream openRequiredResource(final ClassLoader classLoader, final String path) { + final InputStream stream = classLoader.getResourceAsStream(path); + if (stream == null) { + throw new IllegalStateException("Missing benchmark-only Hunspell resource: " + path); + } + return stream; + } + + /** + * Benchmark language mapping. + */ + private enum HunspellLanguageCase { + + /** + * English Hunspell dictionary over the Radixor English corpus. + */ + ENGLISH("en", StemmerPatchTrieLoader.Language.US_UK), + + /** + * Czech Hunspell dictionary over the Radixor Czech corpus. + */ + CZECH("cs", StemmerPatchTrieLoader.Language.CS_CZ), + + /** + * German Hunspell dictionary over the Radixor German corpus. + */ + GERMAN("de", StemmerPatchTrieLoader.Language.DE_DE), + + /** + * Spanish Hunspell dictionary over the Radixor Spanish corpus. + */ + SPANISH("es", StemmerPatchTrieLoader.Language.ES_ES), + + /** + * French Hunspell dictionary over the Radixor French corpus. + */ + FRENCH("fr", StemmerPatchTrieLoader.Language.FR_FR), + + /** + * Dutch Hunspell dictionary over the Radixor Dutch corpus. + */ + DUTCH("nl", StemmerPatchTrieLoader.Language.NL_NL), + + /** + * Polish Hunspell dictionary over the Radixor Polish corpus. + */ + POLISH("pl", StemmerPatchTrieLoader.Language.PL_PL), + + /** + * Ukrainian Hunspell dictionary over the Radixor Ukrainian corpus. + */ + UKRAINIAN("uk", StemmerPatchTrieLoader.Language.UK_UA); + + /** + * Wooorm/dictionaries resource code. + */ + private final String hunspellResourceCode; + + /** + * Matching Radixor language. + */ + private final StemmerPatchTrieLoader.Language radixorLanguage; + + /** + * Creates a language mapping. + * + * @param hunspellResourceCode Hunspell resource code + * @param radixorLanguage Radixor language + */ + HunspellLanguageCase(final String hunspellResourceCode, final StemmerPatchTrieLoader.Language radixorLanguage) { + this.hunspellResourceCode = hunspellResourceCode.toLowerCase(Locale.ROOT); + this.radixorLanguage = radixorLanguage; + } + + /** + * Returns the Hunspell resource code. + * + * @return resource code + */ + String hunspellResourceCode() { + return this.hunspellResourceCode; + } + + /** + * Returns the matching Radixor language. + * + * @return Radixor language + */ + StemmerPatchTrieLoader.Language radixorLanguage() { + return this.radixorLanguage; + } + } +} diff --git a/src/jmh/java/org/egothor/stemmer/benchmark/MultiLanguageStemmerComparisonBenchmark.java b/src/jmh/java/org/egothor/stemmer/benchmark/MultiLanguageStemmerComparisonBenchmark.java index d3be956..f4f78bf 100644 --- a/src/jmh/java/org/egothor/stemmer/benchmark/MultiLanguageStemmerComparisonBenchmark.java +++ b/src/jmh/java/org/egothor/stemmer/benchmark/MultiLanguageStemmerComparisonBenchmark.java @@ -473,6 +473,19 @@ public class MultiLanguageStemmerComparisonBenchmark { filterState.germanMinimalStem.run(sharedState.german.tokens, blackhole); } + /** + * Runs CISTEM directly over the German corpus. + * + * @param sharedState shared benchmark state + * @param blackhole result sink + */ + @Benchmark + public void germanCistem(final SharedState sharedState, final Blackhole blackhole) { + for (final String token : sharedState.german.tokens) { + blackhole.consume(Cistem.stem(token)); + } + } + /** * Runs Radixor over the Spanish corpus. * diff --git a/src/jmh/java/org/egothor/stemmer/benchmark/StemmerComparisonBenchmarkQuality.java b/src/jmh/java/org/egothor/stemmer/benchmark/StemmerComparisonBenchmarkQuality.java index f9978db..6d845ad 100644 --- a/src/jmh/java/org/egothor/stemmer/benchmark/StemmerComparisonBenchmarkQuality.java +++ b/src/jmh/java/org/egothor/stemmer/benchmark/StemmerComparisonBenchmarkQuality.java @@ -144,6 +144,7 @@ public class StemmerComparisonBenchmarkQuality { "GERMAN_LUCENE_GERMAN_STEM_FILTER", "GERMAN_LUCENE_GERMAN_LIGHT_STEM_FILTER", "GERMAN_LUCENE_GERMAN_MINIMAL_STEM_FILTER", + "GERMAN_CISTEM", "SPANISH_RADIXOR", "SPANISH_LUCENE_SPANISH_LIGHT_STEM_FILTER", "SPANISH_LUCENE_SPANISH_MINIMAL_STEM_FILTER", @@ -178,46 +179,32 @@ public class StemmerComparisonBenchmarkQuality { "UKRAINIAN_RADIXOR", "UKRAINIAN_MORFOLOGIK_DIRECT", "UKRAINIAN_LUCENE_MORFOLOGIK_FILTER", - "SNOWBALL_DANISH_RADIXOR", "SNOWBALL_DANISH_DIRECT", "SNOWBALL_DANISH_LUCENE_FILTER", - "SNOWBALL_DUTCH_RADIXOR", "SNOWBALL_DUTCH_DIRECT", "SNOWBALL_DUTCH_LUCENE_FILTER", - "SNOWBALL_FINNISH_RADIXOR", "SNOWBALL_FINNISH_DIRECT", "SNOWBALL_FINNISH_LUCENE_FILTER", - "SNOWBALL_FRENCH_RADIXOR", "SNOWBALL_FRENCH_DIRECT", "SNOWBALL_FRENCH_LUCENE_FILTER", - "SNOWBALL_GERMAN_RADIXOR", "SNOWBALL_GERMAN_DIRECT", "SNOWBALL_GERMAN_LUCENE_FILTER", - "SNOWBALL_HUNGARIAN_RADIXOR", "SNOWBALL_HUNGARIAN_DIRECT", "SNOWBALL_HUNGARIAN_LUCENE_FILTER", - "SNOWBALL_ITALIAN_RADIXOR", "SNOWBALL_ITALIAN_DIRECT", "SNOWBALL_ITALIAN_LUCENE_FILTER", - "SNOWBALL_NORWEGIAN_BOKMAL_RADIXOR", "SNOWBALL_NORWEGIAN_BOKMAL_DIRECT", "SNOWBALL_NORWEGIAN_BOKMAL_LUCENE_FILTER", - "SNOWBALL_NORWEGIAN_NYNORSK_RADIXOR", "SNOWBALL_NORWEGIAN_NYNORSK_DIRECT", "SNOWBALL_NORWEGIAN_NYNORSK_LUCENE_FILTER", - "SNOWBALL_PORTUGUESE_RADIXOR", "SNOWBALL_PORTUGUESE_DIRECT", "SNOWBALL_PORTUGUESE_LUCENE_FILTER", - "SNOWBALL_RUSSIAN_RADIXOR", "SNOWBALL_RUSSIAN_DIRECT", "SNOWBALL_RUSSIAN_LUCENE_FILTER", - "SNOWBALL_SPANISH_RADIXOR", "SNOWBALL_SPANISH_DIRECT", "SNOWBALL_SPANISH_LUCENE_FILTER", - "SNOWBALL_SWEDISH_RADIXOR", "SNOWBALL_SWEDISH_DIRECT", "SNOWBALL_SWEDISH_LUCENE_FILTER", - "SNOWBALL_YIDDISH_RADIXOR", "SNOWBALL_YIDDISH_DIRECT", "SNOWBALL_YIDDISH_LUCENE_FILTER" }) @@ -343,6 +330,7 @@ public class StemmerComparisonBenchmarkQuality { GERMAN_LUCENE_GERMAN_STEM_FILTER(StemmerPatchTrieLoader.Language.DE_DE), GERMAN_LUCENE_GERMAN_LIGHT_STEM_FILTER(StemmerPatchTrieLoader.Language.DE_DE), GERMAN_LUCENE_GERMAN_MINIMAL_STEM_FILTER(StemmerPatchTrieLoader.Language.DE_DE), + GERMAN_CISTEM(StemmerPatchTrieLoader.Language.DE_DE), SPANISH_RADIXOR(StemmerPatchTrieLoader.Language.ES_ES), SPANISH_LUCENE_SPANISH_LIGHT_STEM_FILTER(StemmerPatchTrieLoader.Language.ES_ES), SPANISH_LUCENE_SPANISH_MINIMAL_STEM_FILTER(StemmerPatchTrieLoader.Language.ES_ES), @@ -377,52 +365,36 @@ public class StemmerComparisonBenchmarkQuality { UKRAINIAN_RADIXOR(StemmerPatchTrieLoader.Language.UK_UA), UKRAINIAN_MORFOLOGIK_DIRECT(StemmerPatchTrieLoader.Language.UK_UA), UKRAINIAN_LUCENE_MORFOLOGIK_FILTER(StemmerPatchTrieLoader.Language.UK_UA), - SNOWBALL_DANISH_RADIXOR(StemmerPatchTrieLoader.Language.DA_DK, SnowballLanguageCase.DANISH), SNOWBALL_DANISH_DIRECT(StemmerPatchTrieLoader.Language.DA_DK, SnowballLanguageCase.DANISH), SNOWBALL_DANISH_LUCENE_FILTER(StemmerPatchTrieLoader.Language.DA_DK, SnowballLanguageCase.DANISH), - SNOWBALL_DUTCH_RADIXOR(StemmerPatchTrieLoader.Language.NL_NL, SnowballLanguageCase.DUTCH), SNOWBALL_DUTCH_DIRECT(StemmerPatchTrieLoader.Language.NL_NL, SnowballLanguageCase.DUTCH), SNOWBALL_DUTCH_LUCENE_FILTER(StemmerPatchTrieLoader.Language.NL_NL, SnowballLanguageCase.DUTCH), - SNOWBALL_FINNISH_RADIXOR(StemmerPatchTrieLoader.Language.FI_FI, SnowballLanguageCase.FINNISH), SNOWBALL_FINNISH_DIRECT(StemmerPatchTrieLoader.Language.FI_FI, SnowballLanguageCase.FINNISH), SNOWBALL_FINNISH_LUCENE_FILTER(StemmerPatchTrieLoader.Language.FI_FI, SnowballLanguageCase.FINNISH), - SNOWBALL_FRENCH_RADIXOR(StemmerPatchTrieLoader.Language.FR_FR, SnowballLanguageCase.FRENCH), SNOWBALL_FRENCH_DIRECT(StemmerPatchTrieLoader.Language.FR_FR, SnowballLanguageCase.FRENCH), SNOWBALL_FRENCH_LUCENE_FILTER(StemmerPatchTrieLoader.Language.FR_FR, SnowballLanguageCase.FRENCH), - SNOWBALL_GERMAN_RADIXOR(StemmerPatchTrieLoader.Language.DE_DE, SnowballLanguageCase.GERMAN), SNOWBALL_GERMAN_DIRECT(StemmerPatchTrieLoader.Language.DE_DE, SnowballLanguageCase.GERMAN), SNOWBALL_GERMAN_LUCENE_FILTER(StemmerPatchTrieLoader.Language.DE_DE, SnowballLanguageCase.GERMAN), - SNOWBALL_HUNGARIAN_RADIXOR(StemmerPatchTrieLoader.Language.HU_HU, SnowballLanguageCase.HUNGARIAN), SNOWBALL_HUNGARIAN_DIRECT(StemmerPatchTrieLoader.Language.HU_HU, SnowballLanguageCase.HUNGARIAN), SNOWBALL_HUNGARIAN_LUCENE_FILTER(StemmerPatchTrieLoader.Language.HU_HU, SnowballLanguageCase.HUNGARIAN), - SNOWBALL_ITALIAN_RADIXOR(StemmerPatchTrieLoader.Language.IT_IT, SnowballLanguageCase.ITALIAN), SNOWBALL_ITALIAN_DIRECT(StemmerPatchTrieLoader.Language.IT_IT, SnowballLanguageCase.ITALIAN), SNOWBALL_ITALIAN_LUCENE_FILTER(StemmerPatchTrieLoader.Language.IT_IT, SnowballLanguageCase.ITALIAN), - SNOWBALL_NORWEGIAN_BOKMAL_RADIXOR(StemmerPatchTrieLoader.Language.NB_NO, - SnowballLanguageCase.NORWEGIAN_BOKMAL), SNOWBALL_NORWEGIAN_BOKMAL_DIRECT(StemmerPatchTrieLoader.Language.NB_NO, SnowballLanguageCase.NORWEGIAN_BOKMAL), SNOWBALL_NORWEGIAN_BOKMAL_LUCENE_FILTER(StemmerPatchTrieLoader.Language.NB_NO, SnowballLanguageCase.NORWEGIAN_BOKMAL), - SNOWBALL_NORWEGIAN_NYNORSK_RADIXOR(StemmerPatchTrieLoader.Language.NN_NO, - SnowballLanguageCase.NORWEGIAN_NYNORSK), SNOWBALL_NORWEGIAN_NYNORSK_DIRECT(StemmerPatchTrieLoader.Language.NN_NO, SnowballLanguageCase.NORWEGIAN_NYNORSK), SNOWBALL_NORWEGIAN_NYNORSK_LUCENE_FILTER(StemmerPatchTrieLoader.Language.NN_NO, SnowballLanguageCase.NORWEGIAN_NYNORSK), - SNOWBALL_PORTUGUESE_RADIXOR(StemmerPatchTrieLoader.Language.PT_PT, SnowballLanguageCase.PORTUGUESE), SNOWBALL_PORTUGUESE_DIRECT(StemmerPatchTrieLoader.Language.PT_PT, SnowballLanguageCase.PORTUGUESE), SNOWBALL_PORTUGUESE_LUCENE_FILTER(StemmerPatchTrieLoader.Language.PT_PT, SnowballLanguageCase.PORTUGUESE), - SNOWBALL_RUSSIAN_RADIXOR(StemmerPatchTrieLoader.Language.RU_RU, SnowballLanguageCase.RUSSIAN), SNOWBALL_RUSSIAN_DIRECT(StemmerPatchTrieLoader.Language.RU_RU, SnowballLanguageCase.RUSSIAN), SNOWBALL_RUSSIAN_LUCENE_FILTER(StemmerPatchTrieLoader.Language.RU_RU, SnowballLanguageCase.RUSSIAN), - SNOWBALL_SPANISH_RADIXOR(StemmerPatchTrieLoader.Language.ES_ES, SnowballLanguageCase.SPANISH), SNOWBALL_SPANISH_DIRECT(StemmerPatchTrieLoader.Language.ES_ES, SnowballLanguageCase.SPANISH), SNOWBALL_SPANISH_LUCENE_FILTER(StemmerPatchTrieLoader.Language.ES_ES, SnowballLanguageCase.SPANISH), - SNOWBALL_SWEDISH_RADIXOR(StemmerPatchTrieLoader.Language.SV_SE, SnowballLanguageCase.SWEDISH), SNOWBALL_SWEDISH_DIRECT(StemmerPatchTrieLoader.Language.SV_SE, SnowballLanguageCase.SWEDISH), SNOWBALL_SWEDISH_LUCENE_FILTER(StemmerPatchTrieLoader.Language.SV_SE, SnowballLanguageCase.SWEDISH), - SNOWBALL_YIDDISH_RADIXOR(StemmerPatchTrieLoader.Language.YI, SnowballLanguageCase.YIDDISH), SNOWBALL_YIDDISH_DIRECT(StemmerPatchTrieLoader.Language.YI, SnowballLanguageCase.YIDDISH), SNOWBALL_YIDDISH_LUCENE_FILTER(StemmerPatchTrieLoader.Language.YI, SnowballLanguageCase.YIDDISH); @@ -504,6 +476,7 @@ public class StemmerComparisonBenchmarkQuality { tokenFilter(input -> new GermanLightStemFilter(germanNormalize(input))); case GERMAN_LUCENE_GERMAN_MINIMAL_STEM_FILTER -> tokenFilter(input -> new GermanMinimalStemFilter(germanNormalize(input))); + case GERMAN_CISTEM -> direct(createGermanCistemStemmer()); case SPANISH_LUCENE_SPANISH_LIGHT_STEM_FILTER -> tokenFilter(input -> new SpanishLightStemFilter(lowercase(input))); case SPANISH_LUCENE_SPANISH_MINIMAL_STEM_FILTER -> @@ -561,6 +534,15 @@ public class StemmerComparisonBenchmarkQuality { } } + /** + * Creates a CISTEM stemmer adapter. + * + * @return German stem function + */ + private static Stemmer createGermanCistemStemmer() { + return Cistem::stem; + } + /** * Direct stemmer function. */ diff --git a/src/main/resources/de_de/stemmer.gz b/src/main/resources/de_de/stemmer.gz index dc9e057..a02c8b9 100644 Binary files a/src/main/resources/de_de/stemmer.gz and b/src/main/resources/de_de/stemmer.gz differ