Towards an NLP-based approach for measuring syntactic complexity: preliminary experiments with Italian texts from different registers

In this paper, we explore how NLP can be used to automatically identify relevant syntactic complexity features in texts with the aim of assessing their correlation with specific linguistic registers. Our final goal is twofold. On the one hand, we demonstrate that automatic morpho-syntactic and syntactic annotation of texts provides sufficiently accurate output for use in the automatic extraction and measurement of syntactic complexity features. On the other hand, we identify the set of syntactic features strongly correlating with considered linguistic registers.