Once again, I was honored to write a guest post for DataKind. This time is was on the spread of open source software by data-do-gooders. A couple years ago, DataKind hosted a DataDive in Washington D.C. and some of the participants created a mapping software project titled DataTools 2.0. Since then, it has been replicated by a number of groups around the globe. Read the full post on the DataKind blog to find out more.
Continuing this weeks theme on data science colleges, the nice folks at Silk.co created an interactive map of the Data Science University Programs across the globe. Click the map to view the interactive visualization.
Where are the Programs?
Based upon the visualization, it is easy to see most of the programs are in the United States and Western Europe.
What About Degree Types
The data can also easily be broken down by degree type and how the degree is delivered (online/on-campus).
What is Silk.co?
To state it simply, Silk.co is a place to very easily store and visualize data. It looks pretty awesome.
Creating the Awesome Data Science Colleges List has opened the data for some analysis. The above chart shows the states with the most data science programs based upon the total number of colleges in the state (dark red is best, followed by orange, followed by yellow, and finally black is for states with no data science programs).
The Top States
- Washington D.C.
- South Dakota
- New Hampshire
- New York
The States Needing to Create Data Science Programs
- North Dakota
- New Mexico
See the original post at Need to Learn Data Science? These States are the Best! and the analysis at Create US States Choropleth for Data Science Degrees.
I recently compiled a huge list of colleges and universities with data science-related degree programs. The compiled list is available on Github as Awesome Data Science Colleges.
I encourage you to contribute to the list if you know of missing programs.
A great read for people without an extensive math, statistics or computer science background. And still an interesting read for those people.
The book includes tons of non-technical descriptions for data science terms.
HeroX, an organization that runs competitions for big ideas, has recently launched a competition that is relevant to the data science community. It is called the Cognitive Computing Challenge
The challenge sounds fairly simple.
Build a cognitive system that can read a document, then load a database with what it finds.
However, don’t be fooled by the description. Getting a computer to accurately read and “understand” a document is very difficult.
You have until January 11, 2016 to submit a solution, and the winner receives a $200,000 prize.
Anyhow, if you are interested, please join the competition. If you happen to compete and/or win, please leave a comment and I would love to blog about it. Good Luck!
The guide provides some excellent tips on how to get involved.
The algorithm guarantees the same results as traditional K-means, but it produces results with an order of magnitude higher performance.
An abstract of the paper and a PDF download can be accessed at Yinyang K-Means: A Drop-In Replacement of the Classic K-Means with Consistent Speedup.
Organizations everywhere are racing to build analytics/data science teams. Big Data is everywhere and companies don’t want to fall behind. Unfortunately, many organizations are struggling to get started because of questions similar to the following:
- How will Analytics help us?
- What does an analytics team look like in our organization?
- How do we start?
Luckily, the analytics team at 500px, a photography community site, was kind enough to provide a detailed overview, Building Analytics at 500px, of what really happens when building an analytics team. The overview provides:
- And more
If your organization is considering adding an analytics or data science team, this article is definitely worth reading.