Morpho Challenge - Evaluation of algorithms for unsupervised learning of morphology in various tasks and languages

After the release of the open sou rce softw are implementation of M orfessor alg orithm, a series of several open evalu ations has b een org aniz ed for u nsu pervised morpheme analy sis and morpheme-b ased speech recog nition and information retrieval. T he u nsu pervised morpheme analy sis is a particu larly attractive approach for speech and lang u ag e technolog y for the morpholog ically complex lang u ag es. W hen the amou nt of distinct w ord forms b ecomes prohib itive for the constru ction of a su ffi cient lex icon, it is important that the w ords can b e seg mented into smaller meaning fu l lang u ag e modeling u nits. In this presentation w e w ill demonstrate the resu lts of the evalu ations, the b aseline sy stems b u ilt u sing the open sou rce tools, and invite research g rou ps to participate in the nex t evalu ation w here the task is to enhance statistical machine translation b y morpheme analy sis.