Abstract
In this paper, we describe an algorithm that employs linguistic and statistical analyses to extract bilingual collocations from a parallel corpus. Preferred bilingual syntactic patterns of collocation are obtained from idioms and collocations in the machine readable dictionary. Phrases matching the patterns are extract from aligned sentences in a parallel corpus. Those phrases are subsequently matched up based on cross linguistic statistical association. Statistical associations between the whole collocations as well as words in collocations are used jointly to link a collocation and its counterpart collocation in the other language. We experimented with an implementation of the proposed method on a very large Chinese-English parallel corpus with satisfactory results.