EVALITA 2023 Datasets

EVALITA provides a shared framework for the evaluation of different systems and approaches on separate tasks, all for Italian.

Task list and websites: Tasks

---

Affect

EMit – Categorical Emotion Detection in Italian Social Media: Dataset

EmotivITA – Dimensional and Multi-dimensional emotion analysis: Dataset

Authorship Analysis

PoliticIT – Political Ideology Detection in Italian Texts: Dataset

GeoLingIt – Geolocation of Linguistic Variation in Italy: Dataset

LangLearn – Language Learning Development: Dataset

Computational Ethics

HaSpeeDe 3 – Political and Religious Hate Speech Detection: Dataset

HODI – Homotransphobia Detection in Italian: Dataset

MULTI-Fake-DetectiVE – MULTImodal Fake News Detection and VErification: Dataset

ACTI – Automatic Conspiracy Theory Identification: Subtask A Dataset - Subtask B Dataset

New Challenges in Long-standing Tasks

NERMuD -Named-Entities Recognition on Multi-Domain Documents: Dataset

CLinkaRT – Linking a Lab Result to its Test Event in the Clinical Domain: Dataset

WiC-ITA – Word-in-Context task for Italian: Dataset

DisCoTEX – Assessing DIScourse COherence in Italian TEXts: Dataset