<?xml version="1.0" standalone="yes"?>
<Paper uid="W02-0801">
  <Title>A Multilingual Approach to Disambiguate Prepositions and Case Suffixes</Title>
  <Section position="2" start_page="0" end_page="0" type="intro">
    <SectionTitle>
1 Introduction
</SectionTitle>
    <Paragraph position="0"> This paper presents some preliminary experiments in the use of translation equivalences to disambiguate the interpretations of case suffixes in Basque. Basque is an agglutinative language, and its case suffixes are more or less equivalent to prepositions, but are also used to mark the subject and objects of verbs. The method is general, and could be as easily applied to prepositions in any other language. The core of the method is to find a preposition in the translation of an occurrence of the target case suffix, and select the interpretation(s) in the intersection of both as the valid interpretation(s). At this point, we have not used additional sources for the disambiguation, e.g. governing verbs, nouns, etc., but they could complement the technique here presented.</Paragraph>
    <Paragraph position="1"> In this particular experiment, the method was tested on the definitions of a Basque monolingual dictionary, using the -z instrumental as the target case suffix. The main reason is that we are in the process of building a Lexical Knowledge Base out of dictionary definitions, and the disambiguation of case suffixes and other semantic dependencies is of great interest.</Paragraph>
    <Paragraph position="2"> The method searches for the respective definitions in English and Spanish monolingual dictionaries and tries to find a preposition that is the translation of the target case suffix. Once the preposition is found, the intersection of the set of interpretations of both the source case suffix and the translated preposition is taken, and the outcome is stored.</Paragraph>
    <Paragraph position="3"> The resources needed to perform this task are the following: lemmatizers, bilingual dictionaries and monolingual dictionaries, as well as a table of possible interpretations of prepositions and case suffixes. In our case, we have used Basque, English and Spanish lemmatizers, Basque/English and Basque/Spanish bilingual dictionaries, a target Basque monolingual dictionary, 3 Spanish and 3 English monolingual dictionaries.</Paragraph>
    <Paragraph position="4"> The method is fully automatic; the Spanish and English monolingual dictionaries are accessed from the Internet, and the rest are local, installed in our machines. The manual work has been to build the table with possible interpretations of the prepositions and case suffixes.</Paragraph>
    <Paragraph position="5"> July 2002, pp. 1-8. Association for Computational Linguistics. Disambiguation: Recent Successes and Future Directions, Philadelphia, Proceedings of the SIGLEX/SENSEVAL Workshop on Word Sense The paper is structured as follows. Section 2 presents the method for disambiguation in detail.</Paragraph>
    <Paragraph position="6"> Section 3 introduces the interpretations for the case suffix and the prepositions. The results are shown in Section 4, which are further discussed in Section 5. Finally, section 6 presents the conclusions and future work.</Paragraph>
  </Section>
class="xml-element"></Paper>