A Data Science Lesson from Richard Feynman

Richard Feynman

Richard Feynman

Richard Feynman is one of the greatest scientific minds, and what I love about him, aside from his brilliance, is his perspective on why we perform science.   I’ve been reading the compilation of short works of Feynman titled The Pleasure of Finding Things Out, and I recently came across a section that really hit home with me.

In the world of data science, much is made about the algorithms used to work with data, such as random forests or k-mean clustering.  However, I believe there is a missing component – one that deals the fundamentals underlying data science, and that is the real science of data science.

The Best Way To Learn New Things

Science and business seem like two very different disciplines, but is the best approach to learning any different in these two fields?  These areas of life seem so unique, and the people in them can be quite varying (one with the nerdy pocket protector and the other dressed in the well-tailored suit).  However, both science and business require learning, and the best approach to learning in each is really the same.

The best approach to learning is generally through failure.  For example, Thomas Edison failed an astounding number of times before he invented a working lightbulb, and there are likely thousand of stories about how successes came as a result of many tries and many failures.

In many ways, this is really an application of the scientific method.  I’ve written a number of posts about Stephen Wolfram (such as using Wolfram|Alpha to look at your own social network, his views on big data, computing a theory of everything, and how he created his company).  In the effort to learn even more about how the world works, Wolfram has pushed scientific discovery to the next level, which he’s done with his book A New Kind of Science (NKS for short).

Bad Science

Jen Rhee has done some great homework on bad science and put them into a cool infographic that’s worth looking at.  Here are some of the highlights from her research into bad science:

  • 1 in 3 scientists admit to using questionable research practices
  • 1 in 50 admits falsifying or fabricating data outright
  • Among biomedical researcher trainees at UC-San Diego, 81% said they would modify or fabricate results to win a grant or publish a paper

This is obvious disturbing, and worth highlighting to try and root these things out.  Science is about finding the truth – no matter what it is – and as more businesses start using data science in order to drive business outcomes, we need to make sure that science is about being honest – with the truth and with ourselves.

The scientific method was developed to provide the best way to figure out what the truth is, given the data we’ve got.  It doesn’t make perfect decisions (no method can), but it’s the best method available.

Real scientists (the ones not highlighted in Jen’s research) care about what the data is actually saying and discovering the truth.  When someone cares about something else other than the truth (money, celebrity, fame, etc.), then bad science is what you get.  Of course, when there are people involved, sometimes the truth isn’t the top priority.

Great infographic, Jen!  You can find it here

NKS Now Available on the iPad

I really like Stephen Wolfram’s book A New Kind of Science (or NKS for short) – on how simple computational programs can create amazingly complex things.  I’m also really enamored with the iPad (I’ve written about it before a number of times…).

Now, two of my favorite things are coming together – Wolfram’s NKS is now available on the iPad.  It’s available at Apple’s iTunes store for $9.99 – far less than the hardback version and certainly much lighter!…  (Hmmm…  Two things coming together – kind of like chocolate and peanut butter in a Reese’s Peanut Butter Cup…)

I haven’t purchased an iPad yet, but now I’m really excited to think about possibly maybe starting to look into getting one… (or at least window shopping for one…)

Read more about A New Kind of Science for the iPad on Stephen Wolfram’s blog here

Betelgeuse Explosion?

It’s almost like a cosmic hurricane warning signal…  From Bad Astronomy:  “So, what’s the deal with Betelgeuse? What is it, will it explode, and if so, when?”

The constellation Orion’s biggest star is Betelgeuse (pronounced “beetle-juice”), a red supergiant, and it’s one of the brightest stars in the sky – that’s the “what”.  

Will it explode?  Well, according to some observers, the star is not round, and it’s getting smaller.  This could be the signs of Betelgeuse heading toward a supernova phase, but it also could be explained away by large sunspots that fool our observations.

Bad Astronomy seems to think that this is likely an astronomical doomsday rumor.  It may be interesting only because it’s close to the 2012 doomsday date that is the end of the Mayan calendar.  But read the Bad Astronomy article for yourself to get the best information on the topic.

Could Dark Matter Theory Be Dead?

Recent discovery of a new supernova may lead to questions about dark matter theory. Science aficionados know that there appears to be more gravitational attraction than can be explained by the matter that we can see. So, in order for the extra gravity to be there, it’s been theorized that there must be extra matter that we can’t see – in other words, “dark matter“.

However, this new supernova discovery may shine some light on the darkness.  Apparently, this supernova is spewing calcuim and titanium, and while most reports are focusing on the calcium, it is thought that the much heavier element titanium is the interesting part.  This titanium, which is radioactive and emits positrons as it decays, could be the source of what had previously been attributed to dark matter particles colliding.  If these types of supernovae are creating such an abundance of radioactive heavy elements in the universe, this could help explain the supposed dark matter signatures.

For more info on this subject, here’s a post from Casey Kazan from The Daily Galaxy on the new supernova, and the original post from Anil Ananthaswamy at New Scientist is here.

Book Review: “The Perfect Swarm” by Len Fisher

I love to read, and I especially love to read nonfiction books that help me understand how the world works.  So, I’m going to be providing reviews of some of the books that I’ve been reading lately, and let you in on why I like (or don’t like) them and what I’m learning.

Recently I picked up a copy of The Perfect Swarm by Len Fisher, Ph.D., which focuses on the science of complexity in everyday life  Dr. Fisher has also written books on game theory in real life (Rock, Paper, Scissors) and the optimal way to dunk a doughnut (aptly named How to Dunk a Doughnut).  It turns out that Fisher was awarded the Ig Nobel Prize for the dunking donut work – the award is usually reserved for work that just plain bad, but Fisher has apparently turned his lemon into lemonade…

In The Perfect Swarm, Fisher describes how bees, ants, locusts, and fish use swarm intelligence to guide group movements and to help in the search for food.  Humans use swarm intelligence as well, and Fisher describes multiple interesting ways that we can take advantage of these concepts to make better judgments in everyday life.  Here’s just some of the interesting concepts and principles that come out of his analysis of complexity:

  • To move a group in the direction you want (for example, within a company), lead from the inside, but take care not to let other members of the group know what you are doing.  Just head in the direction that you want to go, and leave it to the laws of the swarm to do the rest.
  • If you are in a crowd in a dangerous situation, use a mixed strategy for escape; follow the crowd 60 percent of the time, and spend the other 40 percent searching out escape routes on your own.
  • If you want to persuade a large group of people, or even start a craze, don’t rely on persuading someone with influence to pass the message on.  It is far better to try for a critical mass of early adopters – people who will take the idea or product up after a single exposure.
  • When confronted with a mass of data that you need to use as the basis of a decision, furst use Benford’s Law to check that the data haven’t been faked.
  • Avoid the perils of groupthink by escaping temporarily from the group environment, doing some independent thinking, and committing yourself to the conclusions of that thinking before returning to the group.
  • Spread your bets evenly.  Instead of choosing one alternative over another, allocate your resources equally to each.

And here’s a really interesting one – especially if you are hiring someone in your company:

  • If you want to give yourself the best chance of choosing the very best option in a situation that doesn’t allow you to go back to the options that you have rejected, look at 37 percent of those available, then choose the next one that is better than any of them.  This will give you a 1 in 3 chance of finding the best option, and a very high chance of finding one in the top few percent.

I am incredibly interested in systems that have simple rules yet create complex behavior.  The world can be modeled as a system using simple rules, and Fisher is quite effective in showing us some interesting and useful things that come from looking at these systems.  It’s a good read, interesting to understand the concepts he’s getting across, and, I would say, worth the time.  You can pick up a copy of The Perfect Swarm here

Video: Stephen Wolfram – Computing a Theory of Everything

Stephen Wolfram recently gave a talk about his efforts to understand the universe around us through computation.  He’s the CEO and founder of Wolfram Research, creator of Mathematica, and author of A New Kind of Science.  Wolfram recently launches his computational knowledge engine, Wolfram|Alpha (I wrote a post about its launch some time back…). 

Here’s the video of his talk, given at a recent TED conference

Watch and learn!

The Universe Just Got Bigger

Last week, Nancy Atkinson posted on Universe Today that astronomers are now understanding why they’ve missed 90% of the observable galaxies, and, with new tools, are now able to see them.

Here’s a paragraph from Atkinson’s article:

“Astronomers have long known that many surveys of distant galaxies miss 90% of their targets, but they didn’t know why. Now, astronomers have determined that a large fraction of galaxies whose light took 10 billion years to reach us have gone undiscovered. This was found with an extremely deep survey using two of the four giant 8.2-meter telescopes that make up ESO’s Very Large Telescope (VLT) and a unique custom-built filter. The survey also helped uncover some of the faintest galaxies ever found at this early stage of the Universe.”

You can read the Universe Today article here

Copernicium – The Newest Element

Element 112 has an official name – Copernicium – named after the 16th-century Polish scientist Nicholas Copernicus, who first theorized that the Earth revolved around the Sun.  Copernicium’s periodic element symbol is Cn.

While the name was announced some time back, it became official only recently.

Wonder how this would look in the periodic table of periodic tables?…

Read the Los Angeles Times article about Copernicium and the other most recently named elements here