Meet Jeff Jonas, the Latest IBM Fellow With No College Degree

By David Strom / April 11, 2012 03:07 AM / 4 Comments

This week, IBM announced its next group of IBM Fellows, seven of its employees who share, according to the press release, "a commitment to tackling the world's biggest problems with ingenuity, invention and inspiration." The designation is a big deal for IBM, and over the years only 238 staff members have been so honored.

One of the more interesting choices this time is Jeff Jonas, a 47-year old chief scientist with the company who blogs here. Jonas never graduated from college with any degree but is clearly one of the smarter people you'll ever come across. He is also quite a character.

Unlike many of his fellow Fellows - who have resumes that you might have trouble parsing - Jonas has lived a very interesting life and worked on numerous problems that are easily understood by the rest of us.

Jonas came to IBM through a 2005 acquisition of Systems Research and Development, a company that he founded in 1985 to handle labor reporting, inventory management and other back-office systems consulting. One of his jobs was designing the casino security systems in Las Vegas, where he currently lives. He worked for the surveillance intelligence group of several casinos, and automated various manual processes, adding facial recognition software that was key to slowing down the MIT card counting group. "We built [another] system to immediately identify risk in real time so they could get these people out of the casino quickly." This software is still offered by IBM as its InfoSphere Identity Insight event processing and identity tracking technology.

Jonas is one of these people that look at the world with very careful thinking, always searching for actionable patterns. For example, he helped use his casino risk-management system to track down lost family members after the Katrina flooding of New Orleans. He and his team integrated data across 15 web sites - these web sites were being used by people who said they were seeking family members with those seeking them. I was impressed by how he structured his algorithm so it wasn't going to be used by bill collectors, for example.

He calls this perpetual analytics and sense-making to keep track of data changes and to help advise decision-makers in real time. "As information changes, you want to be able to reconsider earlier decisions. If you want to prevent really bad things from happening, you want to be able to monitor risks and trends while they are happening." You want to monitor the motion of the data, as it were.

His current internal IBM project is called G2. The idea is to "make sense of new observations as they happen, fast enough to do something about it, while the transaction is still happening." His work is looking at how to commingle diverse data and weave them together - especially when things are the same, such as people named Billy and William, who could be the same person. "If you can count things that are the same, you can analyze them better and understand how they are related. It is a bit of a breakthrough technology," he told me in an interview today. "I took what I developed for the casinos and made it more generalized and easier to use." He and IBM plan to offer G2 sometime soon for the paying business public.

He gives another example in his blog:

"If someone has three phone numbers - no big deal. On the other hand, if someone has five different dates of birth, that just doesn't seem quite right does it? That would be confusing. Why is this important? Well, if you are looking to analytics to make important decisions, wouldn't you want to know during the decision making process if there was related confusion ... before [any] action is taken."

So a car's color can change over its lifetime, but its make and model remains the same. A person's Social Security Number should remain the same. "The trick is being able to relate how each of these data points to other things."

You can listen to a podcast interview that Jonas did back in March 2009 here with two other IBMers, where he talks about some of these concepts.

IBM Fellow Jeff Jonas on the evolution of Big Data

Recently named IBM Fellow Jeff Jonas is one of the most interesting big data thought-leaders. He spoke to CNET about the increasing value of data-driven decisions.

by Dave Rosenberg

April 21, 2012 3:08 PM PDT Follow @dr138

Jeff Jonas, Chief Scientist, Entity Analytics, IBM

(Credit: IBM)

Last week I reconnected with Jeff Jonas, chief scientist of the IBM Entity Analytics group and a recently named IBM Fellow, about what's going on in the realm of big data.

When I first met Jonas, back in June of 2010, he was focused on how companies are dealing with the deluge of information associated with Big Data. His focus hasn't changed, but he told me his perspective on how we make sense of data continues to evolve -- especially as we move in and out of demand for real-time versus batch data processing.

New Big Data tools make it much more affordable to gather and organize large sets of data that can be analyzed in its raw form. As advanced analytics applications get applied against that data, it becomes dramatically easier to identify the direct cause-and-effect relationship between business events, regardless of what department is nominally in charge of that event or the process associated with it.

According to Jonas, the three V's -- volume, velocity, and variety -- are the essential characteristics of "Big Data" that will grow exponentially, rather than in a linear fashion. Accordingly, you have to plan for data growth in conjunction with any projects you plan to undertake.

But planning is just one aspect of the Big Data situation. More important is knowing what you want to get out of the data analysis you're performing. Trying to make more sense of data is growing in importance for businesses of all kinds, but the techniques employed are relative to the problem you're trying to solve.

Jonas found that by default most organizations go with a batch approach, using tools like Hadoop and other MapReduce implementations. This approach works well for "thinking apps," where you are looking for information and context to inform a bigger notion or data amalgamation as opposed to a real-time decision. But there are many things that can and should be done in real-time primarily because batch/MapReduce processes are too late with decisions and suggestions.

Based on his experience at IBM, Jonas suggested that as the value of the analysis rises, business users will demand that data be delivered sooner -- even if it's not totally practical. "I need to know now, not later, what this data means to my business."

Ultimately, this all feeds into the work Jonas has been doing around puzzles and problem solving, changing the way to think about the context of the data. There may be more than one aspect to the problem. Moneygram, for example, is using IBM Identity Insight for context accumulation -- weaving data together to address fraud complaints, which have dropped 72 percent since they started using it, proving the value of Big Data analysis in real business terms.

As a side note, Jonas is a big triathlete. He told me that if completes all the rest of the races he's planned this year and with two more next year he's been given the impression that he will be the third human being to have done every international Ironman event. Cheer for him as he swims/bikes/runs by.