Bilingual lexicons are fundamental resources. Modern automated lexicon generation methods usually require parallel corpora, which are not available for most language pairs. Lexicons can be generated using non-parallel corpora or a pivot language, but such lexicons are noisy. We present an algorithm for generating a high quality lexicon from a noisy one, which only requires an independent corpus for each language. Our algorithm introduces non-aligned signatures (NAS), a cross-lingual word context similarity score that avoids the over-constrained and inefficient nature of alignment-based methods. We use NAS to eliminate incorrect translations from the generated lexicon. We evaluate our method by improving the quality of noisy Spanish-Hebrew lexicons generated from two pivot English lexicons. Our algorithm substantially outperforms other lexicon generation methods. 1

Previous work on bilingual lexicon extraction from comparable corpora aimed at finding a good representation for the usage patterns of source and target words and at comparing these patterns efficiently. In this paper, we try to work it out in another way: improving the quality of the comparable corpus from which the bilingual lexicon has to be extracted. To do so, we propose a measure of comparability and a strategy to improve the quality of a given corpus through an iterative construction process. Our approach, being general, can be used with any existing bilingual lexicon extraction method. We show here that it leads to a significant improvement over standard bilingual lexicon extraction methods. 1