Data Analytics & Viz

Visit my personal data science/viz (and other stuff) blog.

I completed Data Science, a 9-course specialization by Johns Hopkins University on Coursera. 
Specialization Certificate earned on May 5, 2015
Some things, you just need to experience.


PyPI Dependency Network. Data were collected from the Python Package Index site (thanks to Ed David for writing a Python script to extract the necessary information). I used Python to build and visualize the network. The necessary network package was graph-tool.

My Twitter Network. A network of my followers and their followers. Used Python and Gephi to process everything. This reminds me of the Friendship Paradox: Most people have fewer friends than their friends have, on average.

OFW Demand Tracker. Learning how to map connections using great circles using R. Data taken from the POEA website ( Unfortunately, the POEA report only provides the Top 10 OFW destinations (2011). It would have been nicer if the data had more points. The colors of the arcs correspond to the number of hires/rehires of OFW in the specific countries. We have the most OFWS in Saudi Arabia, then UAE, then Singapore. I also created a simple pie chart of the different occupational groups and categories of our workers for 2011. 

OFWs in the Philippines. Learning how to place pie-charts on maps using R. I got the data online from the NSO Data Archive website ( The heat map on the left shows the relative number of OFWs across the 17 administrative regions of the Philippines. We have the most number of OFWs coming from the CALABARZON Region.

On the right, the pie-charts show the distributions of the male (blue) and female (pink) OFWs coming from each region. I find it interesting that in the northernmost and southernmost parts (in particular), the OFWs are mostly women. The other regions have mostly male OFWs. I conjecture that most of our foreign domestic workers come from those areas. It's worth a look. :)