Sergio Eugenio Zanotto
Sergio Eugenio Zanotto started his journey in the research world in October 2021 as a PhD candidate in Linguistics and an Independent Doctoral Fellow at the Cluster of Excellence "The Politics of Inequality", University of Konstanz. His work implies the creation and the analysis of a corpus of Italian political discourses. He would like to shed lights on which linguistic features are more relevant in attracting political preferences among different strata of the population.
Independent Project:
- Discourse analysis of Italian political discources
Doctoral Project:
"Analysis of Italian Political Discourse: How Language Influences Political Preferences"
Abstract:
My dissertation project aims at understanding how linguistic features influence political preferences among different strata of the population. In order to achieve this goal, the project is divided into three main phases.
The first phase includes the creation of a corpus of 150 Italian Political TV Interviews taken from 5 different popular TV shows during the Election period of the Italian parliamentary elections in 2018. The corpus is annotated at the textual level by using the Stanza pipeline and the Italian version of LIans. The output of this phase is a fully annotated corpus for conducting different linguistic analyses.
Thus, a series of analyses are conducted in order to identify possible linguistic features that could be meaningful in influencing political preferences. Examples of such analysis are type-token ratio, average lexical density, question classification, topic analysis, argument mining and language complexity. The output of the second phase includes a broad range of results that permits a better characterization of the language of Italian Politics. Nevertheless, these results will help to identify the proper data in the corpus to use as the dataset for the third phase.
The final phase consists of the creation of a survey to submit to participants who have the right of voting in the Italian parliamentary elections. The main idea is based on collecting judgements over arguments on a specific topic and analyzing these judgements considering the different educational, socio-economical and geographical backgrounds of the participants. The sentences are going to be selected or eventually modified in order to avoid biases given by already fixed preferences of specific individuals or parties.
Research Interests:
- Manual and automated detection of arguments in political discourses
- Impact of different linguistic features on attracting political preferences among different strata of the population