A corpus-based local context analysis for spoken dialogues
Yasuhiro Sobashima; Osamu Furuse; Hitoshi Iida
Speech Communication 15(3-4): 205-212
1994
ISSN/ISBN: 0167-6393 DOI: 10.1016/0167-6393(94)90072-8Document Number: 257626
Grammatical and semantic constraints are effective for interpreting or understanding linguistic expressions. However, they appear to be inadequate for selecting among several candidates, all of which may be relatively correct or inadequate grammatically or semantically. Clearly, we humans interpret a linguistic expression contextually even if there are many potential interpretations. This paper introduces an example-based local context analysis method using tagged corpora to deal with contextual selection of linguistic expressions, taking into account the cohesive nature of spoken dialogues. This method performs calculations for similarity scores between linguistic expressions and for likelihood scores to select the most suitable expression. Both illocutionary force-based and morpho-syntactic classifications are considered, along with the frequencies of existing sets of neighboring linguistic expressions stored in an example database. An experimental processing unit which performs such local context analysis has been implemented in a bidirectional (English and Japanese) translation prototype system, and has shown its applicability to the selection of context-dependent translation candidates. This local context analysis mechanism can be used with conventional translation systems without contextual processing to raise translation accuracy.