ogencoglu/fair_cyberbullying_detection
Source code and models for the paper "Cyberbullying Detection with Fairness Constraints". IEEE Internet Computing, 2020
Implements fairness-constrained training using TensorFlow Constrained Optimization (TFCO) to enforce group-specific false negative/positive rate bounds across identity groups during model optimization. Supports multiple datasets (Jigsaw, Twitter, Wikipedia, Gab) with pre-trained models and reproduces results via Jupyter notebooks that configure fairness deviation thresholds per dataset. Uses Proxy Lagrangian optimization with dual constraint and objective optimizers to balance detection accuracy against demographic parity in error rates.
No commits in the last 6 months.
Stars
19
Forks
5
Language
Jupyter Notebook
License
MIT
Category
Last pushed
Mar 25, 2023
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/nlp/ogencoglu/fair_cyberbullying_detection"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Related tools
KawaroX/CyberShield
Cyberbullying early warning and governance system - Uses natural language processing and machine...
tahtaciburak/CyAnnuaire
Cybersecurity event detection and analysis platform based on real time tweets.
CasuallyPassingBy/Cyberbullying_Tweet_Analysis
This project focuses on the analysis of cyberbullying tweets categorized by various...
khaoula-ghz/Cyberbullying-Detection
Cyberbullying detection for Algerian, Arabic & English text
rahul-sharma0507/Cyberbullying-Data-Analysis
An NLP-based machine learning project for Cyberbullying Severity Detection. Analyzes social...