List of Meta-Resources for Big Data

A “Cliff Notes” for Big Data

Following is a list of meta-resources which have been identified by Dr. Kirk Borne in a blog post he wrote at Data Science Central. The list is made up of links clickable through to the resource they identify. I have also included a link to Dr. Borne’s blog post:

Dr. Kirk Borne writes at Data Science Central:

The flood of articles, webinars, and conferences related to Big Data is generating its own “infoglut”. Consequently, it is really helpful when you find resources that summarize many of the latest developments in one place – a sort of “Cliff Notes” for Big Data.  Here are six meta-resources that I have found useful, plus one additional collection that I authored:

Big Data Meta-ResourcesDr. Borne’s original blog post: Big Data – Seven Meta-Resources for Best Practices, Lessons Learned, Data Stories, Opportunities, and Insights – Data Science Central

The Biggest Big Data Project to Date Should Have Been in Texas

An example of simulated data modelled for the ...

This eulogy of the Superconducting Supercollider (SSC) that should have been built in the US— in TX —struck a chord with me. This is the biggest big data project that has been and it should have been here. How shortsighted we have been. The advancements being made at CERN are not just toward some ephemeral ‘God Particle‘, but also toward data science, operations engineering, “Big Data” stream processing and on and on and on.

We gathered all the nation’s elite
to design and build man’s greatest feat

“Come build it for us” Congress said
“America must be at the forefront, always ahead.”

So come we did, pursuing the dream
to build the machine that collides proton beams

“It’s pork-barrel, useless, garbage” they cried.
“Off with its head” the House puffed with pride.

Our families just stare, confused and upset,
The children all innocent, the spouses with regret

“Why did I come out here?” they wonder to themselves.
“I left the home I lived in since I was just twelve.

”Little was ventured, and little was gained
Except to fill the nation’s physicists with anger and pain.

How Texas Lost the World’s Largest Super Collider | Texas Monthly

Metacademy: Machine Learning and Probabilistic AI Learning Resources

I’ve recently come across a tremendous resource for the discovery of various machine learning and probabilistic artificial intelligence topics and associated educational materials. The site I’m describing is Metacademy.Metacademy Large Cropped Home PageMetacademy is a community-driven, open-source platform to facilitate the collaborative construction of a web of knowledge by domain experts meant to help individuals efficiently learn about any topic of interest (supported by Metacademy and the domain experts). The experts responsible for Metacademy are Roger Grosse and Colorado Reed. In addition to building the site, they organized roughly 350 machine learning and probabilistic artificial intelligence concepts along with related training and learning materials.

While Metacademy is currently focused on machine learning and probabilistic artificial intelligence topics, eventually, it has the goal to cover a much wider breadth of knowledge; e.g. mathematics, engineering, music, medicine, computer science, etc.

The premiss of Metacademy is that a user will search for and click on a concept of interest. Metacademy then produces a “learning plan” which includes the prerequisite concepts which were identified in the web of knowledge previously created by the domain experts. This component of identifying for the student the list of prerequisite concepts is what sets Metacademy apart from other learning sites or course catalogs.

As posted at Metacademy:
… But try learning something of conceptual depth by sifting through Google search results … and you’re in for a lot of agony. Before you learn this concept, you need to learn its prerequisite concepts (sometimes you’re not entirely sure what these are), and the prerequisite concepts may have prerequisites themselves. Pretty soon, you’re deep in dependency hell, switching between twenty different tabs trying to understand the various [pre]prerequisite concepts in order to understand the tutorial article Google returned …

Metacademy’s learning experience revolves around two central components:

  • a “learning plan” in a tabular ‘list view’
    Metacademy Logistic Regression List View
  • a “graph view” of the learning plan which is meant to help explore relationships among concepts
    Metacademy Logistic Regression Graph View

Clicking on the check-mark next to the title of a concept in either the graph or list view marks that [prerequisite] concept as being understood. To not show those concepts which have been marked as being understood, click the “hide” button in the upper right. Note that Metacademy will remember the concepts marked as understood and hidden and will automatically re-apply these selections at future visits.

As Metacademy is a work in progress and limited in scope, please keep an open mind when visiting, but I think that you will find it an interesting, unique and valuable resource, particularly if you are, as I am, actively exploring the world of machine learning.