Skip to content
Harsha Apoorv
All projects
CAPSTONE · 2022–23AI/ML

Detecting cyberbullying in tweets with classic NLP

My B.Tech capstone. I led a team of five to build a text classifier that flags cyberbullying in tweets.

Project report (opens in a new tab)

  • 20K+tweets

  • 4models compared

  • 92%accuracy (Random Forest)

01

Preview

Cyberbullying detection project
02

How it works

  • Cleaned and tokenized 20K+ tweets with NLTK
  • Turned text into features with TF-IDF
  • Trained and compared 4 classifiers; Random Forest reached 92% accuracy
03

What's next

Automatically reporting flagged content was scoped as future work; the project delivered the detection model.

Got a role, a project or a question?

Send a message right from the site. It lands straight in my inbox.

Say hi