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

Competition 2: Finnish


Table: The obtained average precision (AP%) in the information retrieval task for the submitted segmentations in Finnish (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 75,000 (right).
Tfidf BM25 weighting for all morphemes Okapi BM25 weighting and 75,000 stoplist
METHOD WORDLIST AP% METHOD WORDLIST AP%
morfessor baseline withnew 0.4105 Bernhard 2 withnew 0.4915
Bernhard 1 withoutnew 0.4016 Bernhard 1 withnew 0.4681
grammatical first withoutnew 0.3995 Bernhard 2 withoutnew 0.4425
Bernhard 2 withoutnew 0.3984 morfessor baseline withnew 0.4412
morfessor baseline withoutnew 0.3978 morfessor catmap withnew 0.4353
grammatical all withoutnew 0.3952 Bordag 5a withnew 0.4309
morfessor catmap withnew 0.3913 Bordag 5 withnew 0.4308
Bernhard 1 withnew 0.3896 grammatical all withoutnew 0.4307
Bordag 5 withnew 0.3831 grammatical first withoutnew 0.4216
morfessor catmap withoutnew 0.3814 Bernhard 1 withoutnew 0.4183
Bernhard 2 withnew 0.3811 grammatical first withnew 0.4176
Bordag 5 withoutnew 0.3802 Bordag 5a withoutnew 0.4147
grammatical first withnew 0.3760 Bordag 5 withoutnew 0.4095
grammatical all withnew 0.3734 grammatical all withnew 0.4066
Bordag 5a withoutnew 0.3721 morfessor baseline withoutnew 0.3820
Bordag 5a withnew 0.3673 McNamee 5 withnew 0.3684
McNamee 5 withoutnew 0.3646 morfessor catmap withoutnew 0.3632
McNamee 5 withnew 0.3618 McNamee 5 withoutnew 0.3620
porter withnew 0.3566 McNamee 4 withnew 0.3603
dummy withnew 0.3559 McNamee 4 withoutnew 0.3567
McNamee 4 withoutnew 0.3518 porter withnew 0.3517
McNamee 4 withnew 0.3257 McNamee 3 withoutnew 0.3386
McNamee 3 withoutnew 0.2941 dummy withnew 0.3274
Zeman withoutnew 0.2494 McNamee 3 withnew 0.3243
McNamee 3 withnew 0.2182 Zeman withoutnew 0.2813

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 Wednesday, 08-Aug-2007 13:10:33 EEST