Tübingen at PoliticIT: Exploring SVMs, Pretrained Language Models, and Linguistic Transfer for Ideology Detection in Social Media
Much like other personal characteristics, such as gender, age, or native language, political ideology represents a useful psychographic trait that can help to understand individual and social behavior.
This repository hosts the data and the solutions we submitted for the EVALITA 2023 Task - PoliticIT: Political ideology detection in Italian texts. The goal of the task is to extract gender and political ideology from a set of Italian tweets. In this setting political ideology is considered both as a binary as well as a multiclass feature.
In our solutions we experiment both with transformer-based models as well as traditional SVM classifiers and this is reflected in the way the present repository is organized.