A multilingual approach to disambiguate prepositions and case suffixes

Abstract

This paper presents preliminary experiments in the use of translation equivalences to disambiguate prepositions or case suffixes. The core of the method is to find translations of the occurrence of the target preposition or case suffix, and assign the intersection of their set of interpretations. Given a table with prepositions and their possible interpretations, the method is fully automatic. We have tested this method on the occurrences of the Basque instrumental case -z in the definitions of a Basque dictionary, looking for the translations in the definitions from 3 Spanish and 3 English dictionaries. The results have been that we are able to disambiguate with 94.5% accuracy 2.3% of those occurrences (up to 91). The ambiguity is reduced from 7 readings down to 3.1. The results are very encouraging given the simple techniques used, and show great potential for improvement.

Hindi Correspondence of Bengali Nominal Suffixes

“…A parallel corpus based approach for disambiguating the prepositions and case suffixes for Basque language has been proposed by Agirre et al [1]. This general approach could also be applied to find translation equivalences for prepositions of any language.…”

“…It is translated to Hindi suffix -0 when it is attached to a word which is the karta of a transitive verb which is not in past perfect tense. We refer both of these two uses of null-marker to null (1). The Bengali null-marker is translated to Hindi postposition "ne" when the head word, with which the marker is attached, is a karta of the verb which is in the past perfect tense.…”

“…In both the cases, the noun with which the ra-marker is attached is nonspecific and in direct (not oblique) form. We refer to both of these two uses of ra-marker as ra (1).…”

“…A Bengali te-marker may be translated to Hindi nullmarker or postpositions "sAtha", "kA", "se", "me.N", "pe", and "lie". The te-marker which is used with karta is referred to as te (1). te( 1) is translated to the Hindi suffix -0 when it is attached with the single karta (general karta) of the sentence.…”