Claudia Roberta Combei

University of Pavia | UNIPV

CorAIt is a non-native speech database for Italian, which is freely accessible online for academic research purposes. It was especially designed to meet the requirements of a larger research project focused on foreign accented Italian speech. The corpus is aimed at providing a uniform collection of speech samples uttered by non-native speakers of I...

Il presente contributo si colloca nell’ambito dell’analisi del contenuto e del discorso e, attraverso strumenti specifici, mira a esplorare la dimensione lessicale della politica populista italiana. Ci interroghiamo sulle scelte lessicali adoperate su Twitter dai politici di orientamento populista, per diffondere contenuti, definire valori e progra...

This paper investigates the acquisition of stressed vowels in L2 Italian by 6 German and 6 Romanian adult learners, who had lived in Italy for at least 6 months at the time they were recorded. In order to ensure comparability and make inferences, we examined them in comparison to 6 native Italian speakers in the same age range. The first result of...

This methodological study deploys hybrid techniques to investigate how femicide is framed in media. Results are consistent with ISTAT data and with the literature, and also offer novel insights. We find a tendency of not holding offenders accountable; that most femicides are perpetrated by men that victims know well; and that mediatic discourse aro...

Although Natural Language Processing (NLP) is at the core of many tools young people use in their everyday life, high school cur-ricula (in Italy) do not include any computational linguistics education. This lack of exposure makes the use of such tools less responsible than it could be and makes choosing computational linguistics as a university de...

We describe and make available the game-based material developed for a laboratory run at several Italian science festivals to popularize NLP among young students.

Although Natural Language Processing (NLP) is at the core of many tools young people use in their everyday life, high school curricula (in Italy) do not include any computational linguistics education. This lack of exposure makes the use of such tools less responsible than it could be and makes choosing computational linguistics as a university deg...

We describe and make available the game-based material developed for a laboratory run at several Italian science festivals to popularize NLP among young students.

Using a corpus-based and cross-modal approach, this study explores mediatic linguistic representations of gendered violence. Specifically, we analyse a journalistic corpus (WItNECS, Women in Italian Newspaper Crime Sections) and a multimedia dataset of the Italian TV program Amore Criminale (AC). The corpus is explored via collocational analysis of...

As testified by recent elections in Europe, populism has experienced significant social and political resonance. The term ‘populism’ itself has gained ground in the public opinion even if its conceptual understating remains merely intuitive. The present study treats populism as an approach to politics built around the people’s will (Mudde 2004). Th...

This study inductively explores the policy and populism dimensions of Italian political discourse on Twitter. Starting from a corpus of 25,000 tweets posted by a number of Italian political actors throughout a year (March 4, 2018 – March 4, 2019), we apply the Structural Topic Model to uncover the policy content underlying their political communica...

This is a popular science article published by Linguisticamente. The article is available here in open access: https://www.linguisticamente.org/cosa-sono-i-corpora/

In this article we perform an analysis of the immigration issue in the public discourse of a number of Italian political actors in the period 2018-2019. We select Twitter as a main arena of political communication, as the literature shows it has become an important part of the Italian information ecology. The novelty of our study lies in the applic...

Following similar research on non-native speech, in this paper we assessed the degree of perceived foreign-accented Italian. This is an extensive study on 6 accents (i.e. English, German, French, Romanian, Spanish, and Russian, in comparison to L1 Italian), supported by a database of 42 speakers whose performance was evaluated by 288 Italian listen...

Relying on linguistic cues obtained by means of various statistical models as well as descriptive lexical analyses, this study contributes to the general understanding of the Twitter users’ response to the annual Italian budget law approved at the end of December 2018. Some topics contained in the dataset of tweets are procedural or generic, but be...

Metodi quantitativi e qualitativi per la rappresentazione lessicale della violenza di genere nelle principali testate giornalistiche (La Stampa, Corriere della Sera, Il Fatto Quotidiano, Repubblica) e nel linguaggio televisivo della docu-fiction (Amore Criminale).