The paper presents some definitions for a corpus and the main features of modern corpora. It is shown that the computational lexicography has its own traditions in Bulgaria. At the same time not full potential of computational linguistics approaches and methods is in use.
