The TExSIS project aims at the automatic extraction of mono‐ and multilingual company specific terminology on the basis of a company’s document streams. These term lists are crucial in every language based man‐machine communication: in machine translation, computer‐assisted translation and in monolingual and multilingual document management.