期刊名称：LAHOR: journal for Croatian as mother, second and foreign lanugage
印刷版ISSN：1846-2197
电子版ISSN：1848-4972
出版年度：2013
卷号：2
期号：16
页码：189-205
语种：Croatian
出版社：Crotian Philological Society
摘要：Nowadays language corpora are recognised as valuable and informative sources of linguistic information. However, retrieving the available data can be demanding and complex, therefore sometime not suitable for all users that could benefit from it. The only existing Croatian corpus of spoken language is the Croatian Corpus of Child Language (CCCL --- Kovacevic, 2002). Speech samples were taken from three children, in equable time periods, from the onset of speech to three years of age. Samples were transcribed according the rules of CHAT, using the computer programme CLAN. CCCL is available on-line in the CHILDES (Child Language Data Exchange System --- ). It is designed to provide data about lexical and grammatical development in language acquisition. Consequently, a Croatian frequency dictionary of child language (CFDCL) has been designed to enable easier data retrieval form CCCL. It allows the analyses of most frequent lemmas in all three sub-corpora according to frequency, alphabetic ordering, time of appearance, and part-of-speech. Furthermore, it preserves the morphological encoding of types, and number of types and tokens. Therefore it incorporates a larger amount of information than traditional corpora of written language, enabling users to extract relevant information about child language development such as type/token ratio, lexical diversity, morphological diversity, etc.
关键词：Croatian corpus of child language; Croatian Frequency Dictionary of Child Language; CHILDES; lemmatization; tagging of corpus; structure of CFDCLP
