Customer Segmentation
Built K-Means clustering models to segment households that fear they’ll be unable to get credit
Reduced high-dimensional data using principal component analysis
Bankruptcy Prediction
Implemented gradient boosting & random forests algorithms to predict bankruptcy
Earthquake Damage Prediction
Assessed SQLite database, 2015 Nepal Earthquake, and manipulated data with SQL
Built decision trees & logistic regression models for classification to predict building damage
Air Quality Index Prediction
Preprocessed air quality data with MongoDB and prepared time series data for analyses
Built ARMA & ARIMA time series models to predict PM 2.5 readings throughout the day
Housing Price Predictor
Vectorized real estate data and built linear regression & ridge models to predict housing prices
Built data pipelines for imputing missing values and encoding categorical features
Credit Card Fraud Detection
Examined credit card transactions by European card holders
Created machine learning models (logistic regression, random forests, and feed-forward neural network) to classify a given transaction as fraudulent or non-fraudulent
Wine Class Classification
Implemented logistic regression algorithm to classify quality red wine
Time Series Forecasting using Facebook Prophet
Trained time-series prediction models to predict future avocado prices