When Alex Guazzelli talks about predictive analytics, he likes to use this analogy: “We’ve come a long way since palm reading.”
Instructor Profile: Alex Guazzelli: "Predictive Models with PMML"
As a specialist in this fast-emerging field, Guazzelli has a dual role: Chief Technology Officer at Zementis, a San Diego-based analytics firm; and instructor at UC San Diego Extension.
“We’re living in a highly complex scientific age where enormous amounts of data are being gathered everywhere on everything we do,” said Guazzelli. “Someone needs to make sense out of all that data, and find value and insight from it.”
A native of Brazil who earned his Ph.D. from University of Southern California, Guazzelli later became a psychology faculty member at the University of Washington. There, he applied his knowledge of artificial neural networks to the study of the brain.
“I’ve always been fascinated by how the brain works and how we process information,” he said. “That’s why I love what I do. It’s the science of creating a mathematical representation for every-day life, to learn and predict behavior in much the same way the brain does.”
His current Extension course is “Predictive Models with PMML” (Predictive Model Markup Language), part of the Data Mining Specialized Certificate program.
“When my students ‘get it,’ I feel responsible for that,” he said. “It’s very rewarding."
These days, super-computers and what’s termed “machine learning” (think iPhone's Siri) reproduce many of the brain’s functions — but hardly all of them. Not even close.
“The biggest challenge of computers is to solve mundane tasks we humans perform every day,” he said. “Like making coffee in the morning — a robot doesn’t do that yet."
Meanwhile, serving as an Extension instructor gives Guazzelli and his occasionally over-taxed neurons a special dose of satisfaction.
“The transmission of knowledge is always fascinating,” he said. “Other creatures do it, of course, but we humans excel at it.”