Definition

The goal of the task is to detect political bias of news reporting at the article and at the media level. This is an ordinal classification task and is offered in English. It includes two subtasks:

• Subtask 3A: Given an article, classify its political leaning as left, center or right.
• Subtask 3B: Given the URL to a news outlet (e.g., www.cnn.com), predict the overall political bias of that news outlet as left, center or right leaning.

Datasets

Currently under preparation.

Evaluation

This is an ordinal classification task. We use mean absolute error as the official measure for both subtasks.

Submission

Format checker

Scorer

Baseline script

Submission site

Leaderboard

Will be available soon.

Organizers

• Giovanni Da San Martino, University of Padova
• Preslav Nakov, Mohamed bin Zayed University of Artificial Intelligence
• Firoj Alam, Qatar Computing Research Institute, HBKU
• Rabindra Nath Nandi, BJIT Limited
• Dilshod Azizov, Mohamed bin Zayed University of Artificial Intelligence

Contact

Please join on Slack channel for any queries: [https://join.slack.com/t/newworkspace-qgd1635/shared_invite/zt-1k9lnm0ys- aMeEmUtGY0tfW5HbkXHPJg](https://join.slack.com/t/newworkspace-qgd1635/ shared_invite/zt-1k9lnm0ys-aMeEmUtGY0tfW5HbkXHPJg)

Slack Channel (task3-political-bias): https://ct-23-participants.slack.com