Abstract
Measuring the relative compositionality of Multi-word expressions (MWEs) is crucial to Natural Language Processing. Hindi contains a rich set of Noun+Verb MWEs and hence, it is very important to handle them. Very limited work was done previously towards characterizing the MWEs in Hindi of Noun+Verb type. Also, various statistical measures which are used to measure the compositionality of different kinds of collocations in English cannot be applied straight-away to Hindi due to insufficient corpus and resources. In this paper, we analyze in detail the types of Noun+Verb expressions in Hindi. We then propose an approach to measure their relative compositionality automatically using maximum entropy model (MaxEnt). MaxEnt integrates various measures representing the properties of the Noun+Verb expressions in Hindi. Some of the measures used by the MaxEnt are computed by mapping them to Verb-Noun expressions in English.
Users
Please
log in to take part in the discussion (add own reviews or comments).