Morpheme Extraction Task at FIRE 2012--2013

The Morpheme Extraction task (MET) was organized for the first time in FIRE 2012 and subsequently in FIRE 2013. Participating systems were required to provide morphemes of given term lists. The track was offered in five languages viz. Bengali, Gujarati, Hindi, Odia, Marathi in 2012 and a sixth language: Tamil was added in 2013. Evaluation was done in two different ways viz., IR evaluation and Linguistic evaluation. The IR evaluation exercise comprised of retrieval experiments on given corpora with a standard set of queries with fixed relevant judgements and culminated by finding the Mean Average Precision for the runs. For Linguistic evaluation, system generated morphemes were compared against gold standard morpheme analyses. This overview paper describes the goals, data, tasks, participants, evaluation and results obtained.