I demonstrate a variety of machine learning modeling techniques including multiple linear regression and random forests to predict life expectancy across countries. Furthermore I show the benefits of using nested models and hypertuning to improve model accuracy.

I demonstrated hierarchical clustering on a food wholesaler dataset to analyze customer segmentation and different spending patterns. These insights will be more effective in targeting specific customer groups.

Practice hypertuning using h2o

We are going to look at anonymized patients who have been diagnosed with heart disease. Patients with similar characteristics might respond to the same treatments, and doctors would benefit from learning about the outcomes of patients similar to those they are treating. The data we are analyzing comes from the V.A. Medical Center in Long Beach, CA.

I explore crime statistics collected between 1975 and 2016 from the State of Maryland. I’ll see if linear trends exist in crime rates across Maryland. This data comes from the Maryland Statistical Analysis Center.

I analyze ten economic and social indicators collected for each province. Because these indicators are highly correlated, I will use principal component analysis (PCA) to reduce redundancies and highlight patterns that are not apparent in the raw data. After visualizing the patterns, I will use k-means clustering to partition the provinces into groups with similar development levels. These results can be used to plan public policy by helping allocate resources to develop infrastructure, education, and welfare programs.