Explainable Online Sexism Detection

# Explainable Online Sexism Detection
# Based on Kirk et al., SemEval 2023
# Paper: https://aclanthology.org/2023.semeval-1.305/
# Dataset: https://github.com/rewire-online/edos
#
# This task asks annotators to classify online posts for sexist content
# and highlight the specific expressions that are sexist. The classification
# follows a fine-grained taxonomy covering different types of sexism.
#
# Classification Labels:
# - Not Sexist: The text does not contain sexist content
# - Threats: Direct or indirect threats of harm based on gender
# - Derogation: Derogatory or demeaning language targeting gender
# - Animosity: Expression of hostility or contempt toward a gender
# - Prejudiced Discussion: Stereotyping or generalizing about gender

annotation_task_name: "Explainable Online Sexism 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: sexism_category
    description: "Classify the type of sexism present in this text"
    labels:
      - "Not Sexist"
      - "Threats"
      - "Derogation"
      - "Animosity"
      - "Prejudiced Discussion"
    keyboard_shortcuts:
      "Not Sexist": "1"
      "Threats": "2"
      "Derogation": "3"
      "Animosity": "4"
      "Prejudiced Discussion": "5"
    tooltips:
      "Not Sexist": "The text does not contain sexist content"
      "Threats": "Direct or indirect threats of harm, violence, or intimidation based on gender"
      "Derogation": "Derogatory, demeaning, or belittling language targeting a gender"
      "Animosity": "Expression of hostility, contempt, or ill-will toward a gender group"
      "Prejudiced Discussion": "Stereotyping, generalizing, or expressing prejudiced views about gender"

  - annotation_type: span
    name: sexist_expression
    description: "Highlight the specific text that contains the sexist expression"
    labels:
      - "Sexist Expression"

annotation_instructions: |
  You will see a social media post. Your task is to:
  1. Determine whether the post contains sexist content.
  2. If sexist, classify the type of sexism (threats, derogation, animosity, or prejudiced discussion).
  3. Highlight the specific text spans that contain the sexist language.
  Note: Focus on the content, not the speaker's identity. Consider context carefully.

html_layout: |
  <div style="padding: 15px; max-width: 800px; margin: auto;">
    <div style="background: #fef2f2; border: 1px solid #fecaca; border-radius: 8px; padding: 16px; margin-bottom: 16px;">
      <strong style="color: #991b1b;">Post:</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: 3
allow_skip: true
skip_reason_required: false