Frequency distribution of nominal inflection in Czech

Employing methods of usage-based linguistic approaches, this paper tests the claim that differerences in frequential distributions of cases of nominals in Czech can be explained with the animacy hierarchy. Grammatical profiles consisting of information about gender and number are extracted from SYN2015, a balanced corpus of contemporary written texts, and analysed by hierarchical clustering which groups the grammatical profiles according to similarities of relative frequential distribution of cases. The cluster analysis and subsequential conditional inference tree modelling that animacy divides the sample into two groups. Key words: frequency analysis, animacy hierarchy, cluster analysis, corpus metho