A Corpus-Based Analysis of the Language Used by Defendants of Homicide in Court

Abstract

In this study we present the updated version of the Greek Corpus of Defendants' Testimonies, GCDT and a series of new evaluations that have been carried out on the defendants' speech. Using criteria, such as lexical richness, lexical density, part-of-speech frequencies, word and sentence length, we look for linguistic features which could characterize the stylometric profile of the defendants. We also present GCWT, a reference corpus that has been constructed similar to GCWT stylistic features. GCWT contains witnesses' testimonies collected in the court.