Interactive Data Visualization: Causes of Mortality

I love letting the data tell the story. This interactive data visualization lets the user explore the top 215 causes of mortality from 2000 to 2016, and as the gif shows, it can tell powerful stories very quickly. I used Python's Pandas library to clean and reshape the data, and I used d3.js to build this bespoke interactive visualization. 

WHO_causes_of_mortality_small3.gif

Machine Learning to Identify and Predict Heart Disease

Heart disease is the cause of 1 out of every 4 deaths in the United States, but it (heart disease, not death) can be reversed if detected and treated. In this project, I used Python's Pandas and Seaborn libraries to preprocess and visually explore the data, and the Scikit-Learn library to explore 10 different machine learning algorithms and achieve a classification accuracy of 85%.  

Heart_Disease_cfn_mtx.PNG

Exploratory Data Analysis: Prison Populations by State

In Louisiana and Oregon, conviction for violent crimes only requires votes from 10 of 12 jurors. I discovered this after plotting state prison populations relative to total state population and wondering why Louisiana had a significantly higher imprisonment rate than any other state. I used Python's Pandas, Bokeh, Matplotlib, and Seaborn libraries to process and explore this data set.

crime_bokeh_male_choropleth.png