Abstract
Treq - Translation Equivalents Database R44 Popis anglicky Treq is a collection of bilingual Czech-foreign language dictionaries, built automatically from the InterCorp parallel corpus. The first step was to align the original and translated texts word-to-word using statistical methods provided by the GIZA++ program.
The aligned word pairs were then sorted and summarized. The results of the automatic excerption were not reviewed; however, the relative frequency of a given pair may serve as an approximate indicator of reliability.
The more often an equivalent of a given word occurs in comparison with other equivalents, the more useful it may be. In release 2.0, the queries accept regular expressions and multi-word units in addition to individual words.
We can also choose whether the result should be based on translations of the core, individual collections, or all InterCorp texts. The database is extracted from InterCorp release 9.