My completed machine learning projects. Detailed notes are in each IPython notebook.
- Spam Classification
- Binary classifcation, detects if an email is spam or not spam
- 98.6% accuracy on testing set
- Kaggle Challenge
- Negitive comment sentiment
- Multiclass, multi label classification of Wikipedia comments.
- Predicts 6 different classes of negitivity in comments: toxic, severe toxic, obscene,threat,insult and identity hate
- 96.8% accuracy on testing set
- Kaggle Challenge
- Demo App