CombiNet. A Corpus-based Online Database of Italian Word Combinations

This paper introduces CombiNet dictionary, an on line corpus-based lexicographic tool representing combinatorial properties of Italian lexemes, developed by Roma Tre University, University of Pisa and University of Bologna. The lexicographic layout of CombiNet is designed to include different sets of information, such as i) syntactic configurations and ii) syntactic function of word combinations, iii) degree of lexical variation associated with specific types of multiword units. In fact, CombiNet records word combinations showing different degrees of lexicalizations and paradigmatic variability, which is a novelty in lexicography. This investigation intends to tackle several issues associated with CombiNet, and in particular it aims at a) showing procedures and methods used to create and compile CombiNet's entries, b) describing particular types of combinatorial phenomena emerged from the analysis of corpus-based data, c) illustrating the lexicographic layout that has been elaborated for word combinations representation, d) describing the advanced research tool CombiNet is equipped with, a useful device for lexicographic investigations as well as for lexicological analysis.