Laboratory of Computer and Information Science / Neural Networks Research Centre CIS Lab Helsinki University of Technology

Competition 2: German


Table: The obtained average precision (AP%) in the information retrieval task for the submitted segmentations in German (Competition 2 participants in bold and reference methods in normal font). Indexing is performed using Okapi (BM25) weighting for all morphemes except the most common ones (stoplist) with frequency higher than 150,000.
AUTHOR METHOD AP%
Monson Paramor+Morfessor 0.4734
Monson Morfessor 0.4671
Morfessor baseline 0.4656
Morfessor MAP 0.4642
McNamee four 0.4388
McNamee five 0.4331
snowball german 0.3865
Monson Paramor 0.3631
dummy - 0.3509
grammatical first 0.3353
McNamee lcn5 0.3276
grammatical all 0.3014

Return to the result page

Last year's results (Morpho Challenge 2007)


Table: For reference, the results for Morpho Challenge 2007. There were some changes in the reference methods in 2008. The obtained average precision (AP%) in the information retrieval task for the submitted segmentations in German (Competition 2 participants in bold and reference methods in normal font). Indexing is performed using Tfidf (BM25) weighting for all morphemes (left) and Okapi (BM25) weighting for all morphemes except the most common ones (stoplist) with frequency higher than 150,000 (right).
Tfidf BM25 weighting for all morphemes Okapi BM25 weighting and 150,000 stoplist
METHOD WORDLIST AP% METHOD WORDLIST AP%
morfessor baseline withnew 0.3874 Bernhard 1 withnew 0.4729
morfessor baseline withoutnew 0.3826 Bernhard 2 withoutnew 0.4676
Bernhard 1 withoutnew 0.3777 Bernhard 2 withnew 0.4625
Bernhard 2 withoutnew 0.3731 Bernhard 1 withoutnew 0.4611
porter withnew 0.3725 Monson morfessor withnew 0.4602
Bernhard 1 withnew 0.3720 morfessor catmap withnew 0.4571
Bernhard 2 withnew 0.3703 morfessor baseline withnew 0.4486
Monson morfessor withnew 0.3520 Monson morfessor withoutnew 0.4481
Monson morfessor withoutnew 0.3502 morfessor catmap withoutnew 0.4447
dummy withnew 0.3496 morfessor baseline withoutnew 0.4417
Bordag 5a withnew 0.3496 Bordag 5 withnew 0.4308
morfessor catmap withoutnew 0.3480 Bordag 5 withoutnew 0.4303
McNamee 5 withoutnew 0.3442 Bordag 5a withnew 0.4259
morfessor catmap withnew 0.3397 Bordag 5a withoutnew 0.4257
McNamee 5 withnew 0.3327 Monson paramor-m. withnew 0.4012
Bordag 5 withoutnew 0.3273 Monson paramor-m. withoutnew 0.3989
grammatical first withoutnew 0.3223 porter withnew 0.3866
Monson paramor-m. withnew 0.3200 McNamee 5 withoutnew 0.3617
grammatical first withnew 0.3196 McNamee 5 withnew 0.3527
Monson paramor-m. withoutnew 0.3184 grammatical first withoutnew 0.3467
Bordag 5 withnew 0.3156 McNamee 4 withoutnew 0.3453
grammatical all withnew 0.3128 grammatical first withnew 0.3445
grammatical all withoutnew 0.3126 McNamee 4 withnew 0.3351
McNamee 4 withoutnew 0.3091 Monson paramor withnew 0.3241
McNamee 4 withnew 0.3049 dummy withnew 0.3228
Monson paramor withnew 0.2887 Monson paramor withoutnew 0.3224
Monson paramor withoutnew 0.2861 grammatical all withnew 0.3004
Zeman withoutnew 0.2828 McNamee 3 withoutnew 0.2953
McNamee 3 withoutnew 0.2023 grammatical all withoutnew 0.2926
McNamee 3 withnew 0.1945 McNamee 3 withnew 0.2868
      Zeman withoutnew 0.2568

Return to the result page

HOME | RULES | SCHEDULE | DATASETS | EVALUATION | WORKSHOP | RESULTS | FAQ | CONTACT

You are at: CIS → Unsupervised Morpheme Analysis -- Morpho Challenge 2007

Page maintained by webmaster at cis.hut.fi, last updated Thursday, 10-Jul-2008 09:29:07 EEST