Some kinds of diminutives tend to be misinterpreted by foreign language learners, even if their native language is typologically similar, as in the case of Polish learners of Czech. This is especially true about words with formal derivation suffixes but lacking any diminutive or emphatic meaning.
However, a simple type of mismatch - the presence and absence of diminutive suffixes or meanings in translational equivalents - is often accompanied by polysemy and/or homonymy and the prevalent phenomenon of false friends. Following a detailed analysis of the individual types of mismatches we present a study of translational equivalents, extracted automatically from a parallel corpus (about 12 mil. word tokens per language).
Diminutive lexemes are classifies according to their suffix. The results complement previous results about a higher number of diminutive types in Czech by detailed statistics calculated on tokens.