Predictive Analytics: The Past as Prelude
Editor’s Note: The following is an excerpt from Mike Kuniavsky‘s full article, “The Past as Prelude,” now available in The Connective edition through the WIRED tablet newsstand. See below for full download instructions.
A big promise of the Internet of Everything is that by analyzing millions of new sources of data from embedded, networked devices, we have a better and more efficient experience of the world. The environment automatically predicts our behavior and adjusts to it, anticipating problems and intercepting them before they occur.
The notion is seductive and almost magical: an automatic espresso machine that starts a fresh latte as you’re thinking it’s a good time for coffee; office lights that dim when it’s sunny and electricity is expensive; a taco truck that arrives just as the crowd in the park is getting peckish.
Although exciting in theory, the path from technical capabilities to life improvement is unspecific in the details. Exactly how will our experience of the world, our ability to use all the collected data, become more efficient and more pleasurable?
There’s a term for the technologies behind products that continuously adapt their behavior: predictive analytics. While descriptive analytics tries to explain current and past states of the world, and prescriptive analytics guides decisions by analyzing their potential impact, predictive analytics models continuously changing data streams to anticipate future behavior based on past patterns.
Predictive systems pick actions that their models say are most likely to have desirable outcomes: when to change the temperature so that it’s right when you get home, the fastest route to work based on current traffic, the baby food your child will want next month. When events happen that are outside of its model, when its prediction is wrong, a predictive system adjusts to be more accurate in the future.
For example, the Nest Learning Thermostat starts by recording readings from its onboard sensors and its users’ behavior, to which it probably adds the weather, the time, the day of the week, and secret sauce data (the spot price of energy? how many people are in the house right now? how neighbors have set their Nests?). It (and by “it” I mean the Nest service in the cloud) builds a model of how people prefer heating and air conditioning to work, subtly adjusted for every household. Then the Nest acts on its users’ behalf. It changes the temperature, but it doesn’t just play back past behavior or preset preferences. Instead, it acts on what it believes users want. A Nest remembers when you turned the AC on full blast and, two hours later, felt freezing and turned it off. It knows that when you set it to heat the house to 65, an hour later you’ll turn it to 70. And it assumes that everyone wants to reduce greenhouse emissions and save money. That it’s acting based on the collection of massive amounts of data is what differentiates it from past thermostats. Those devices played back a program, perhaps a complex program, but their behavior was based on what their programmers thought was right. The changing behavior of a Nest is not programmed; it programs itself.
Download The Connective 2.0
1. Download the free WIRED iPad® Edition from the App Store.
2. Tap the WIRED cover in your Newsstand folder to open the app.
3. Tap The Connective cover and download for free.
1. Go to the Newsstand on your device.
2. Search for WIRED.
3. Download and install the app to your device.
4. Tap The Connective cover and download for free.
This article originally appeared on Wired.com. Reprinted with permission.
Join us at the PARC Innovation Lab, September 16 – 17, 2014. Register today!
Read PARC’s new white paper “The Camera Doesn’t Lie: Rapid Observation to Create Better Customer Experiences“
Our work is centered around a series of Focus Areas that we believe are the future of science and technology.
We’re continually developing new technologies, many of which are available for Commercialization.