Hyperpartisan News Detection

Binary classification of news articles as hyperpartisan or not, detecting extreme one-sided political bias. Based on SemEval-2019 Task 4 (Hyperpartisan News Detection).

Archivo de configuraciónconfig.yaml

# Hyperpartisan News Detection
# Based on Kiesel et al., SemEval 2019
# Paper: https://aclanthology.org/S19-2145/
# Dataset: https://zenodo.org/record/1489920
#
# This task asks annotators to determine whether a news article exhibits
# hyperpartisan (extreme one-sided) political bias. Annotators read the
# article text and the source publication to make their judgment.
#
# Classification Labels:
# - Hyperpartisan: The article shows extreme one-sided political bias
# - Not Hyperpartisan: The article is balanced or does not show extreme bias

annotation_task_name: "Hyperpartisan News Detection"
task_dir: "."

data_files:
  - sample-data.json

item_properties:
  id_key: "id"
  text_key: "text"

output_annotation_dir: "annotation_output/"
output_annotation_format: "json"

port: 8000
server_name: localhost

annotation_schemes:
  - annotation_type: radio
    name: hyperpartisan_judgment
    description: "Does this news article exhibit hyperpartisan (extreme one-sided) political bias?"
    labels:
      - "Hyperpartisan"
      - "Not Hyperpartisan"
    keyboard_shortcuts:
      "Hyperpartisan": "1"
      "Not Hyperpartisan": "2"
    tooltips:
      "Hyperpartisan": "The article shows extreme one-sided political bias, presenting only one perspective"
      "Not Hyperpartisan": "The article is relatively balanced or does not display extreme partisan bias"

annotation_instructions: |
  You will be shown a news article along with its source publication. Your task is to
  determine whether the article is hyperpartisan -- that is, whether it exhibits extreme
  one-sided political bias. Consider the language, framing, and presentation of facts.

html_layout: |
  <div style="padding: 15px; max-width: 800px; margin: auto;">
    <div style="background: #fefce8; border: 1px solid #fde68a; border-radius: 8px; padding: 12px; margin-bottom: 12px;">
      <strong style="color: #a16207;">Source:</strong>
      <span style="font-size: 15px;">{{source}}</span>
    </div>
    <div style="background: #f0f9ff; border: 1px solid #bae6fd; border-radius: 8px; padding: 16px; margin-bottom: 16px;">
      <strong style="color: #0369a1;">Article Text:</strong>
      <p style="font-size: 16px; line-height: 1.7; margin: 8px 0 0 0;">{{text}}</p>
    </div>
  </div>

allow_all_users: true
instances_per_annotator: 50
annotation_per_instance: 2
allow_skip: true
skip_reason_required: false

Datos de ejemplosample-data.json

[
  {
    "id": "hyperpartisan_001",
    "text": "The administration's latest policy is nothing short of a disaster for working families. Once again, the radical left has pushed through legislation that will bankrupt small businesses and destroy the American dream. Every single Democrat voted for this monstrosity.",
    "source": "The Daily Patriot"
  },
  {
    "id": "hyperpartisan_002",
    "text": "The Senate voted 52-48 along party lines to approve the new infrastructure bill, which allocates $1.2 trillion for roads, bridges, and broadband internet. Supporters say the investment is overdue, while critics argue the spending is excessive.",
    "source": "Associated Press"
  }
]

// ... and 8 more items

Obtener este diseño

View on GitHub

Clone or download from the repository

Inicio rápido:

git clone https://github.com/davidjurgens/potato-showcase.git
cd potato-showcase/semeval/2019/task04-hyperpartisan-news
potato start config.yaml

Detalles

Tipos de anotación

radio

Dominio

SemEvalNLPMedia AnalysisPolitical Bias

Casos de uso

Bias DetectionMedia AnalysisNews Classification

Etiquetas

semevalsemeval-2019shared-taskhyperpartisannewsbias-detectionmedia

¿Encontró un problema o desea mejorar este diseño?

Abrir un issue

Diseños relacionados

Clickbait Spoiling

Classification and extraction of spoilers for clickbait posts, including spoiler type identification and span-level spoiler detection. Based on SemEval-2023 Task 5 (Hagen et al.).

textradio

Patronizing and Condescending Language Detection

Detection and span-level identification of patronizing and condescending language directed toward vulnerable communities in news media. Based on SemEval-2022 Task 4 (Perez-Almendros et al.).

radiospan

ADMIRE - Multimodal Idiomaticity Recognition

Multimodal idiomaticity detection task requiring annotators to identify whether expressions are used idiomatically or literally, with supporting cue analysis. Based on SemEval-2025 Task 1 (ADMIRE).

radiomultiselect