Identifying and Classifying Toxic Wikipedia Comments as part of an online Kaggle challenge which required me to build machine learning classifiers that were capable of detecting different types of of toxicity like threats, obscenity, insults, and identity-based hate.
Attached is also a PDF outlining my approach, which includes an exploratory data analysis, challenges, methodology (data gathering, data exploration, data preperation, model building), an evaluation of models used (SGD classifier, Logistic regression and Multinomial Naive Bayes), as well as a conclusion and future works section.