Predictive Analytics is “Science for the Masses”

My favorite columnist, Rich Karlgaard, wrote a blog a couple of weeks ago titled, The Trillion Dollar IQ Business. It proposes that an industry around "intelligence-enhancing" products ("intelligence" in terms of human IQ) will be a huge industry and that the "scary smart", those with IQs that make people who barely qualify for Mensa seem average, will inherit the world. Those would be people like Steve Jobs, Bill Gates, Sergey Brin, Mark Zuckerberg, etc. The products of such an industry would include things akin to Viagra or steroids for the brain.
 
Although I mostly agree, I do think that the human brain has a ceiling for IQ. However, I don’t believe there is one for artificial intelligence (AI), even though human intelligence is currently still well in the lead. As a human body could not run a one-minute mile unassisted by devices or becoming something different, I don’t think a human brain could develop an IQ much more an a couple incremental levels above 200 – maybe say 300 (whatever that means).
 
The human brain stores, accesses, and processes information under an architecture that is optimized for competing with other life on Earth. Our brains may be somewhat malleable (we’re capable of learning and modifying our behaviors all our lives), but they all still function under an architecture that has its limits – as does any other architecture. It is a web of trade-offs optimized for the success of our species, at least on Earth.  Perhaps if we could live much longer, we could forge the intellectual throughput of our brains to something equivalent to a much higher IQ – instead of each of us starting from scratch at birth.
 
Anyway, back from science fiction, for we BI practitioners, it’s not a matter of human versus machine, but how we can integrate the best respective aspects of human and machine intelligence into an intelligence greater than each individually. How can we improve the interfaces to streamline the two, allowing the human brain to do what it does best and an "AI" machine to do what it does best? The job of a BI practitioner is to enable human thinkers to better integrate with their machine counterparts. It takes training on both sides. On the human side, this means to build the "scientific mind" of curiousity, passion, open-mindedness, tenacity, fundamental knowledge such as systems-thinking, and last but not least, pragmatism (there is usually a disconnect of some level between theory and practice). On the machine side, this means building machines that can integrate massive amounts of disparate data and present it in a manner conducive towards insight – where "insight" really means something that helps us build the "models in our head" about the world around us.
 
There are many people out there on the cutting edge of this human-machine intellectual partnership. I found one such person recently. I spent my last day of vacation attending an annual event/lecture by a person who is the epitome of a data miner I describe in the paragraph above. Laurie "dragged" me to the University of Idaho’s 2010 Pomology Program Annual Fruit Field Day. It turned out to be an excellent way stimulate my data mining neuron clusters before returning to work in an incredibly interesting way, and one of the highlights of my vacation.
 
Although it was great tasting new varieties of grapes and peaches, this event isn’t just about new varieties of fruit. It’s about the optimization of the production of fruit on many KPIs: Taste, transportability, ripening time (so crops ripening can stagger so as to not overwhelm harvesting efforts), weather, soils, environmental safety (pesticides and fertilizer), nutrition (even anti-oxidants), yield per acre, time to production (the trees bare fruit faster), and water requirements. So beyond supplying new varieties optimized for various circumstances, the products of this facility include best practices given to farmers as well as other entities along the food supply chain.
 
Methods for such optimizations are researched at a facility in Parma, ID, and lead by an incredibly impressive man named Essie Fallahi. Essie is a scientist at heart, but he also possesses a keen understanding of business requirements. He thoroughly understands that there are many aspects to the long supply chain that begins with the farmer and ends with the consumer that must be taken into consideration. Therefore, the product of his efforts have been of immense value to farmers in Idaho and well beyond. I consider R&D like Essie’s to be up there in importance to anything out of Silicon Valley.
 
 

Essie Fallahi (right) and his team.
 
This, my data mining friends, is data mining at its most fun and compelling. And it illustrates the virtue of competition through innovation as opposed to some sort of short-sighted tactic such as suppression, quick wins (with no regard to side-effects), or cheating, which doesn’t usually net any good. The growing proliferation of data mining/predictive analytics helps many more of us become super-productive researchers like Essie. Essie combines his agile human intelligence with a computer’s superior ability to process data from across many sources resulting in a symbiotic fountain-head of innovation.
 
This Facebook photo album has a few pictures from today’s field trip: Facebook Photo Album

About Eugene

Business Intelligence and Predictive Analytics on the Microsoft BI Stack.
This entry was posted in Data Mining and Predictive Analytics. Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s