Speakers
Description
The lexicographic treatment of word-formation categories that lie at the boundary between inflection and derivation remains an unresolved issue. In this paper, we introduce the process of compiling DeriVallex, a valency lexicon containing automatically generated valency frames, which provides information on the valency behavior of Czech noun and adjectival deverbal derivatives being halfway between forms and lexemes. These word-formation categories inherit their valency frames and other valency-related phenomena (as, e.g., reflexivity, reciprocity, and other semantic alternations) from their base verbs. We first outline the automatic generation of valency frames of these noun and adjectival derivatives, which resulted in 17,145 valency frames assigned to lexical units, i.e., individual senses of these derivatives, represented by 11,248 noun and adjectival lemmas. The focus of this paper in on the process of structuring of the obtained data into a valency lexicon, providing rich information on valency and derivational properties of the selected nouns and adjectives, enabling human users to search and filter the data. The resulting data may be integrated into both valency lexicons and explanatory dictionaries, thereby increasing their descriptive comprehensiveness.