An English-Italian MWE dictionary

The translation of Multiword Expressions
(MWEs) requires the knowledge of
the correct equivalent in the target language
which is hardly ever the result of a literal
translation. This paper is based on the assumption
that the proper treatment of MWEs
in Natural Language Processing (NLP) applications
and in particular in Machine Translation
and Translation technologies calls for a
computational approach which must be, at
least partially, knowledge-based, and in particular
should be grounded on an explicit linguistic
description of MWEs, both using an
electronic dictionary and a set of rules. The
hypothesis is that a linguistic approach can
complement probabilistic methodologies to
help identify and translate MWEs correctly
since hand-crafted and linguisticallymotivated
resources, in the form of electronic
dictionaries and local grammars, obtain accurate
and reliable results for NLP purposes.
The methodology adopted for this research work is based on (i) Nooj, an NLP environment
which allows the development and testing
of the linguistic resources, (ii) an electronic
English-Italian MWE dictionary, (iii) a set
of local grammars. The dictionary mainly
consists of English phrasal verbs, support verb
constructions, idiomatic expressions and collocations
together with their translation in Italian
and contains different types of MWE POS
patterns