tackles cyberbullying detection on Twitter using NLP for binary classification.
Identify toxic tweets using NLP techniques. Analyze diverse cyberbullying data from social media. Build models for binary classification with F1-score evaluation. Explore, preprocess, and deploy for real-time detection. Explore the codebase and contribute!
Project Description: This project focuses on the automatic detection of cyberbullying in text data, specifically Twitter tweets. The dataset, sourced from various platforms including Kaggle, Twitter, Wikipedia Talk pages, and YouTube, contains labeled text examples categorized as either bullying or non-bullying. The different types of cyberbullying include hate speech, aggression, insults, and toxicity.
Dataset Information: Dataset Name: twitter_parsed_tweets Target Variable: oh-label (Binary classification: Toxic or Not) Evaluation Metric: F1-score