摘要:The aim of this paper is to investigate morphological and syntactical levels of sentences of Croatian Language. Morphological grammatical level is represented by 16 categories and 47 features with theirs value. Syntactical level is represented with constituent structure trees of sentences. Database has 1001 grammatical sentences of Croatian literature. Database of sentences is the result of theoretical research in computational linguistics. This study extends the analysis of the linguistic data in anthropology in Croatian. Access to database is through the Internet on the address: http://infoz.ffzg. hr/tepes.
关键词:linguistic database; probabilistic context-free grammar; Croatian language