Machine learning projects using classification models( regularized logistic regression, CART, random forest, boosted models)
1. Sentiment analysis for Amazon product review: Use NLP to generate features from review text and create classification models which predict positive/negative sentiment from input features (review text, user profile).
2. Loan Default Prediction on lending club data: predict when a loan is likely to be risky or safe for the bank.
- Implement classification model (logistic regression model, decision trees, RF, boosting, stochastic gradient ascent) on real-world, large-scale machine learning tasks
- Handling missing data
- Evaluate models using metrics (precision and recall, F1 score, ROC, AUC)