Exploratory Data Analysis

The adult dataset is from the 1994 Census database. It is also known as “Census Income” dataset. Details of this dataset can be found at UCI Machine Learning Reposity

Below is a bar plot for counts of income categories by different race and gender. This is an interactive plot. You can change to display by gender group, and/or have two income categories stacked or side by side, or just display one income category. Hover mouse over the bars to display the exact count for that specific race, gender, and income combination.

Next, I use a stacked bar plot to visualize the relationship between years of education and income, in-group proportions are calculated as well.

It is not hard to notice that the in group proportion of making greater than $50,000 a year increase as the years of education increases. For those who don’t have any forms of college education (less than or equal to 8 years of education), less than 10% have an annual income of greater than $50,000. While for those with doctorate degrees, nearly 3 out of 4 makes greater than $50,000 a year.

Spatial Data

Here, the first interactive map shows my office location and home location. Of course, as a safety practice, exacty home location is not marked on map.

Next, I create a simple web app that gives the best route between two locations, with added waypoints as an option. Traffic and road condition is not considered here. Turn by turn navigation is also provided. Let’s say I am picking up my fiancee from Denver International Airport. We decide to have a dinner in downtown Denver before heading home. This app gives the best route from DEN to Boulder, with a stop at downtown Denver.

What if we decide to go to the movie theater in Golden after having dinner? How do we make change to the traveling plan? Just drag the route and add a waypoint at Goldan!