Big Data Profile: Ford Motor Company's Michael Cavaretta
Ford predictive analytics group lead Michael Cavaretta talks big data, machine learning and what he looks for in new hires.
Michael Cavaretta, leader of Ford Research and Innovation's predictive analytics group.
Michael Cavaretta leads the predictive analytics group at Ford Research and Innovation. Hired by Ford Motor Company in 1998 as a technical specialist for the former data mining group (now called predictive analytics), he was promoted to technical leader in 2001.
Cavaretta's group functions as an internal consultancy at Ford, using technologies such as big data, machine learning, artificial intelligence, data mining, text mining and information retrieval to improve business processes across the enterprise.
InformationWeek spoke to Cavaretta in January via phone.
Name: Michael Cavaretta, technical leader of predictive analytics and data mining for Ford Research and Innovation.
Tenure at job: 12 years.
Size of group: I work at Ford Research and Advanced Engineering, a separate entity from Ford IT. It's about 35 people.
What's been your experience hiring data scientists? My impression is that those people go fast. So it pays to make inroads with faculty or sponsoring contests or even presenting data for interesting problems. That's one method. The other method would be to go after some people who maybe have the bulk of skills in one direction, and then train them in another.
How will 2013 be different in terms of hiring? We're pretty comfortable with the skill sets we have. In general, it's going to get harder before it gets easier.
Career accomplishment of which I'm most proud: The record of different groups at Ford Motor Company that we've worked with -- everything from marketing and quality to manufacturing and human resources. It's been great to work with so many different people and be able to bring analytic capabilities to so many different areas of the company.
Decision I wish I could do over: Personally, I really wish I'd gotten my degree quicker. I spent a few years right after my bachelor's working in industry. And while I think that worked really well, I would have happily shaved the last year of that experience and gotten back into academia quicker.
My most important career influencer: The president of the first consulting company I worked for, right after getting my Ph.D. His biggest advice was that technical projects very rarely fail for technical reasons. They almost always fail for political reasons. That has helped me a lot in the rest of my career.
Current top initiatives: Big data, predictive analytics and machine learning. Ford's a large enterprise with a lot of computer systems that generate a lot of data. So there's a lot of opportunities inside [for big data analytics].
The area we're most excited about has a lot to do with all the data being generated by machines already. Of course, the vehicles are huge sources of data. The Ford plug-in hybrid generates 25 Gbytes of data per hour. That's an area we are very excited about.
A rapidly growing area is machine-to-machine, RFID, embedded sensor networks, the Internet of Things. Huge amounts of data are being generated, and mashing that data up with vehicle data and being able to look at patterns brings value.
Most disruptive force in my industry: It's basically what we've been talking about: Sensor networks, RFID, machine-to-machine communication ... the idea that sensors are becoming so cheap, and can be embedded in so many things, that these networks are generating huge amounts of data. It's an area that's ripe for opportunity.
One thing I'm looking to do better: We're part of a research organization, but we're also responsible for delivering results to the corporation, to the business and to our customers internally. It's that balance between making sure that you have an understanding of what new technologies could help, as well as being able to deliver good results. And [being able to] manage our time well when there are so many people coming to us asking for assistance. That's something that I'd like to put some energy into this year.
What I've learned about getting the most out of a team: When you're looking at a team like ours, focused on advanced technology and bringing it to the business, skills are both in a technical and nontechnical area. It hasn't worked out well when we had people focused on one relatively narrow area on the technical side, but they don't have good feel for the business or they can't tell a story well.
What I look for in an employee: We want somebody who can think well but also tell a story well.
The reasons big data projects go wrong: The biggest difficulties have to do with expectations. People have an expectation that, let's say, a predictive analytics project can be run very much like a traditional IT project, [in which you] gather some requirements, build something and deliver something. Big data and predictive analytics technologies don't fit neatly into that hole.
Michael Cavaretta At A Glance
Education: B.S. in information systems from the University of Michigan; M.S. and Ph.D. in computer science from Wayne State University.
Person I'd most like to have lunch with: My grandfather, who unfortunately passed away when I was very young.
First job: Mowing lawns.
If I weren't involved in IT: Chef, Italian cuisine.
Can data analysis keep students on track and improve college retention rates? Also in the premiere all-digital Analytics' Big Test issue of InformationWeek Education: Higher education is just as prone to tech-based disruption as other industries. (Free with registration.)
6 Tools to Protect Big DataMost IT teams have their conventional databases covered in terms of security and business continuity. But as we enter the era of big data, Hadoop, and NoSQL, protection schemes need to evolve. In fact, big data could drive the next big security strategy shift.
Big Data Brings Big Security ProblemsWhy should big data be more difficult to secure? In a word, variety. But the business won’t wait to use it to predict customer behavior, find correlations across disparate data sources, predict fraud or financial risk, and more.