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

Competition 3

Finnish


Table: The obtained BLEU scores in the machine translation task for the submitted analysis in Finnish. Methods without authors are reference methods by the organizers. The types are: S = semi-supervised algorithm, P = unsupervised algorithm with supervised parameter tuning, U = fully unsupervised algorithm.
AUTHOR METHOD BLEU TYPE
MBR combination with word-based model
- Morfessor Baseline 0.2665 U
- Morfessor CatMAP 0.2634 U
- grammatical 0.2623 -
Lignos Base Inference 0.2619 U
Lignos Iterative Compounding 0.2605 U
Lignos Aggressive Compounding 0.2597 U
Kohonen Morfessor S+W 0.2594 S
Spiegler DEAP MDL-CAT 0.2590 S
Spiegler DEAP PROB-CAT 0.2589 S
Kohonen Morfessor S+W+L 0.2582 S
Spiegler Promodes-E 0.2579 P
Spiegler Promodes 0.2564 P
Spiegler DEAP MDL-NOCAT 0.2562 S
Kohonen Morfessor U+W 0.2561 P
Spiegler DEAP PROB-NOCAT 0.2555 S
Spiegler Promodes-H 0.2535 P
Golenia MAGIP 0.2533 S
Individual systems
- Morfessor Baseline 0.2614 U
- words 0.2543 -
Lignos Base Inference 0.2504 U
- Morfessor CatMAP 0.2493 U
- grammatical 0.2492 -
Lignos Iterative Compounding 0.2463 U
Kohonen Morfessor U+W 0.2456 P
Lignos Aggressive Compounding 0.2444 U
Spiegler DEAP MDL-NOCAT 0.2404 S
Spiegler DEAP PROB-NOCAT 0.2383 S
Kohonen Morfessor S+W 0.2376 S
Kohonen Morfessor S+W+L 0.2339 S
Spiegler DEAP PROB-CAT 0.2316 S
Spiegler DEAP MDL-CAT 0.2303 S
Spiegler Promodes-E 0.2251 P
Spiegler Promodes 0.2192 P
Spiegler Promodes-H 0.2107 P
Golenia MAGIP 0.2027 S

German


Table: The obtained BLEU scores in the machine translation task for the submitted analysis in German. Methods without authors are reference methods by the organizers. The types are: S = semi-supervised algorithm, P = unsupervised algorithm with supervised parameter tuning, U = fully unsupervised algorithm.
AUTHOR METHOD BLEU TYPE
MBR combination with word-based model
- Morfessor CatMAP 0.3008 U
- Morfessor Baseline 0.2994 U
Kohonen Morfessor U+W 0.2986 P
Nicolas MorphAcq 0.2986 U
Lignos Iterative Compounding 0.2985 U
Lignos Aggressive Compounding 0.2980 U
- grammatical 0.2970 -
Lignos Base Inference 0.2952 U
Individual systems
- Morfessor Baseline 0.2963 U
- words 0.2955 -
- Morfessor CatMAP 0.2862 U
Kohonen Morfessor U+W 0.2817 P
Lignos Base Inference 0.2803 U
Lignos Aggressive Compounding 0.2788 U
- grammatical 0.2782 -
Lignos Iterative Compounding 0.2770 U
Nicolas MorphAcq 0.2751 U

Return to the result page

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

You are at: CIS → Unsupervised Morpheme Analysis -- Morpho Challenge 2010

Page maintained by webmaster at cis.hut.fi, last updated Tuesday, 17-Aug-2010 16:36:25 EEST