Word-for-word glossing with contextually similar words
Patrick Pantel, Dekang Lin · 2000
Many corpus-based machine translation systems require parallel corpora. In this paper, we present a word-for-word glossing algorithm that requires only a source language corpus. To gloss a word, we first identify its similar words that occurred in the same context in a large corpus. We then determine the gloss by maximizing the similarity between the set of contextually similar words and the different translations of the word in a bilingual thesaurus. 1 Introduction Word-for-word glossing is the process of directly translating each word or term in a document without considering the word order. Automating this process would benefit many NLP applications. For example, in cross-language information retrieval, glossing a document often provides a sufficient translation for humans to comprehend the key concepts. Furthermore, a glossing algorithm can be used for lexical selection in a fullfledged machine translation (MT) system. Many corpus-based MT systems require parallel corpora [1][2][3...