"The capacity to learn is a gift; the ability to learn is a skill; the willingness to learn is a choice." - Brian Herbert

My Fav Photos

Here are some photos I've taken over the years:…

Political Sense From The Census

Tricky Presidential Elections Presidential elections, contrary to popular belief, are not decided by the majority. And in this sense, the United States is not a traditional "democracy" (like in old school Greece) as many mistakingly and often purposely, but not deceptively, taught to us as kids. The electoral college is…

Github Streaking

While I haven't use Github too much in the past, I've begun to really enjoy using it as my primary backup and means of saving progress. This means that a lot of my code is public, even if I don't necessarily want it to be. But, I've also gotten over…

Not So Naive Bayes Classification

Naive Bayes is a surprisingly powerful and efficient classifier. The reason why it's called naive is because it makes the assumption that the features are independent; and this is often false or impossible to prove. But despite this false assumption, Naive Bayes still works so well. So why? Harry Zhang…

The More You Do, The More There Is To Do!

While working on my last project, one big realization I came to was that the more I added to my project, the more things I could add to it. Basically the more I did, the more there was to do. With each feature that I added there were more possibilities.…

The Metis Glass is Half Full

It's been a crazy 6 weeks so far at Metis! We just finished up our McNulty projects and our section on supervised machine learning. Next week we move onto unsupervised learning and natural language processing. I find it surprisingly hard to reflect on this last month and half, because I've…

Interesting Data Sources!

I found this great source of interesting data sets the other day. Thank you @robbseaton for putting this together, it is definitely incredibly useful and just plane cool to look through! Some sample datasets include: 1. Survey of Inmates in State and Federal Correctional Facilities, 2004 (ICPSR 4572) 2. 10,…

D3 Transitions

One of the things that d3.js makes very easy is transitions. Transition are the little animations that connect everything. It's something that can really bring some life to a graph or a plot. In making my visualization, I wanted to make sure to include some cool transitions. You can…

My First D3 Map!

Check this link for the fully interactive model: Census Data Look forward to a complete writeup about this and my McNulty project next week along with tutorials on how I created this! What I Learned Today: 111,111,111 x 111,111,111 = 12,345,678,987,654,321…

Is There Really an Autism Epidemic?

I read an interesting article earlier today about the Autism Epidemic on gizmodo. It brought up some interesting points about how Autism might have been just as prevalent before, but it was just underdiagnosed. It pokes holes in data that shows a very clear rise in Autism in the last…

Mcnulty Visualization Teaser

Here is a preview of the interface I created! There's a working search function: A way to view county info: And a way to generate new maps based on features: What I Learned Today: D3 is incredibly cool!…

D3: The Mighty Data

I finally was able to get my hands on d3.js while trying to make visualizations for my Mcnulty Project. Having significant Javascript and jQuery background definitely helped me get pick it up very quickly. I was actually able to get up an initial map that looks something like this:…

Julia Stats

Julia, is an upcoming langauge for Data Analysis that is blazing fast and even has some advantages over C! Though it's only on version 0.3, it is really showing it's potential! While I am currently mostly using Python, I think giving Julia a shot might not be a bad…

Full Full Stack Javascript

Javscript is a very "fun" language. Thought it wasn't designed to be used for everything, that's slowly what it's becoming. As a programmer, you can use JS on the front end of web development with a multitude of tools (jquery, angular, etc.), JS on the back end with node.js…

Cleaning County Data - No Click Baity Titles Available...

This is a follow up to yesterday's entry about cleaning county election data. The other half (closer to 90%) of my data puzzle was county data for a myriad of other factors found here (the first download). While this dataset provides a wide range of data about counties, it also…

Fishy Data in the 2012 Presidential Elections

Beware of Click-Baity Titles like ^ Another Project at Metis, and another detective that I don't know. Well, meet Detective McNulty of The Wire. He represents our 3rd major project here at Metis which involves machine learning and D3.js. Project McNulty In this project, I am analyzing county level data…

Buster Redirects and Static Pages

Buster (the tool I use to generate static pages from my ghost blog that runs on a node.js server so that I can then push it to Github pages where it's hosted for free) has certain issues when it comes to single page generation because of the way it…

Let's Get Down to Business [Cards]

I received my first set of business cards last week with a Data Scientist title. Things just got real. What I Learned Today: Defensive boxing styles just aren't very fun to watch.…

Metis 1-Month Update

1 Month Done! 2 Months To Go... It has been one month since the start of my experience as a Data Science Fellow at Metis. So to commemorate this joyous occasion, I thought I'd give a short list of the top 5 things I've learned so far to help me…

Random Thoughts: When You're Late, You're Really Late!

I'm not going to go into the politics of being late like done on Forbes, Vitamin Talent, The Wall Street Journal, and many more publications. But rather I wanted to propose a theory I've been thinking about. It's not really a theory, and it's also not original (I'm assuming there's…