8 Mafia Language Analysis and Detection Using Natural Language Processing Tools
Elena Morandini · 2025
The following chapter frames the study of automated language detection, focusing on Natural Language Processing (NLP) techniques to identify distinctive linguistic traits in the language of the Italian Mafia. Drawing from methods used to identify hate speech in social networks, this research explores using Machine Learning (ML) tools to detect Mafia language. An empirical, analytical blending approach was used, using Corpus Linguistics methodology for text collection and NLP standard procedures for analysis. The experiment proceeded in two phases. First, NLP tools such as AntConc, RStudio, and T-Lab identified key linguistic features of Mafia language. Second, the Weka ML supervised approach trained and tested datasets to detect Mafia language automatically. Results are discussed, highlighting their relevance both for Linguistics and law enforcement. The chapter concludes by underscoring the contribution of this piece of research to the interdisciplinary efforts of Linguistics and technology in combating organised crime.