Skip to content

datali-ch/detect-toxic-content

Repository files navigation

Detect Toxic Content

Discussing things you care about can be difficult. The threat of abuse and harassment online means that many people stop expressing themselves and give up on seeking different opinions. Platforms struggle to effectively facilitate conversations, leading many communities to limit or completely shut down user comments.

The challenge is to build a multi-headed model that’s capable of detecting different types of of toxicity like threats, obscenity, insults, and identity-based hate. The model(s) will hopefully help online discussion become more productive and respectful.

SETUP:

jupyter notebook exlopratory_data_analysis.ipynb
  • To fit models, run main script from command line:
python main.py --choose-model=MODEL

where MODEL is

  1. for Bag of Words
  2. for Latent Dirichlet Allocation
  3. for Long Short-Term Memory

About

Use NLP to detect abusive content

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published