 Okay, we're back here live at information on demand IBM's conference. This is a silicon angle and wiki bonds exclusive coverage With the Cube our flagship program. We go out to advance extract the system from the noise I'm John Furrier the founder silicon angle join my co-host Dave Vellante and our next guest is Michael Koa Lenco who's the assistant professor at NC State North Carolina State University welcome to the Cube So tell us about what you're working on at at the University and how does that relate to the data-driven? Analytics social business theme here at IOD which is hot Okay, well in the business school what we're really concerned with is understanding how people make decisions and Here at IOD we talk a lot about how data comes together how we look at for aggregation of information and do analytics But what we like to focus on is what's the next step analytics is one part of it But you have to make a decision at the end of the day based on the information you have So how do you go about collecting the right information and asking the right questions? So what we do is we work with students and with companies in training them How to do critical thinking and then to augment that process by using these technologies that are available to them So we are talking with Fred Balboni yesterday IBM or old we're old school We're talking about the old days and the language sees coming back, you know data processing Decision support. Do you remember those terms? I mean Nightmare is this real? I mean these are all the same Stuff but different look and feel different dynamics exactly and the thing that I always have to caution people is this isn't you know My form of data decision Support systems and those sorts of things that the technology has really evolved with what we can do and at the same time You know we I've heard several talks over the past couple days about well There's all this new data. No, there's not new data people have been saying things and commenting for years There's now methods for capturing that information and now making use of it. So how do you do that? So the tools have to change to reflect that now many times people are very comfortable with taking an information that is structured in nature Again as your previous speaker sped if you said if you look at Amazon and how they figure out what you like and what you Don't like it's very nice to look at that structured information and figure out different correlations Well, we're mostly focused on now though is to take that unstructured information which comprises Pick a number 80 90% of the information out there and figure out how to extract knowledge from that And we think that there's a target-rich opportunity in that tremendous value But you needed a different type of tool set to do that well now that you have things like Hadoop You have the ability to go in and spread your work amongst many machines in order to take in this information We have tools that allow us to extract that information and to do brute You know very large Indexing of unstructured text now we can apply context and specificity in order to extract the information We need much different than the old data sets and cubes that we used to deal with in decision support systems So I've asked you so there is new data though, right? I mean think about the decision support system There's certainly more of it. We can agree on that, but social data is Newish, isn't it? I mean, I guess your argument would be there's social data before we're just capturing it now Exactly my argument and I would say that yeah, there's more of the you know sensor data now the the ways that we have phones All the things that you can take from that. Yeah, that's new but in a sense, you know, okay I have this and I understand what's going on with it What people have to transition beyond now is not only what's happened in the information that you've gleaned from that But why did it happen and how can I change outcome based on it to move away from just looking at straight correlations? But also to take in information now from these different sources and figure out well, where's my white space? Where can I go now? What's not being said that allows me to drive new business decisions and new business models So talk a little bit about that, you know, John is you're making me laugh You're talking about all these these concepts and terms coming back I remember in the early 2000s must have been 2002 2003 HBR came out with an article on how the best decision makers make decisions based on gut feel And I laughed I'm like that's absurd. They're just not you know They analyze in the data or they're not look at the data the right way And you know then of course Malcolm Gladwell comes out with this book blink and you know confirms that that thinking so Take us back. I mean was that just Nonsense was it actually real is it still the case help us? Squint through those well gut feel I think it's in a sense. It's a misinterpretation of gut feel because again if you look at cognition in the ability to Organize information and create knowledge from it. It's a bunch of different inputs, right? And at the end of the day you may subconsciously think it's a gut feeling, but it's experiential based All right, so there there is an anchor in data now the thing that you have now though is the ability you have a Certain capacity a finite capacity of taking in that information What you want to do along with the gut feeling is augmented with these other tools let them do the heavy lifting So you can focus on knowledge creation and activity actions from it So I would say that yes there was gut feeling. Yes, you did do this But if I look at it it was all experiential based and you had data that allowed you to make that decision now You can de-risk that decision because you can collect more of it and focus on the hard things So that's a great answer and I think I agree with you I mean, it's you're right. That's big course It's based on data now the data and theory should be better So we're gonna make better decisions with with less risk of it I heard in the news this morning that the the Obamacare website, you know The person in charge of it was in front of Congress saying well, we added more servers. Oh, yeah, you just hear that Servers We make better decisions will we make you know, we'll we'll we'll can we de-risk things or is it just that there's Things are happening so much faster. We actually face increased risk Well, I think what you have to look at is is data is one aspect of things and if you're looking at decision making the other and probably the Bigger aspect of that is culture. All right So if I don't have a culture that allows you to go in and ask good questions and to challenge the status quo How are things going to change the person who's headed? You know, we can speculate that the person who's heading that website could have said look you will do it this way And no matter what you find You're going to continue down this path all the data in the world wouldn't change that position, right? It was a cultural thing command and control is contradictory to a data-driven organization And I keep here, you know, again, we hear this over and over again. Well, we have data. We can make better decisions Yes, you can but will you be allowed to does the culture in your organization allow you to take that information and make a better decision? Yes, I was on I was doing a monitoring a panel with the CEO G in a couple weeks ago in Chicago and Florian That'll Meyer was there from Kellogg's business school and his analytics side and we're talking about culture, right? And the big thing was, you know data scientists Everyone's got a data scientist and that was kind of the big theme and his point was Anyone can be a data scientist if the tooling is, you know, simple to use hence we heard that yesterday So that was the message here at yesterday So the but the point he made was he said sit the CEOs out there If you're gonna hire a CFO and the guy walks into the office and says hi, I want to be your CFO What do you know about finance? I don't know but someone might on my team will know finance and his point was Everyone has to know data. You can't just say I'm a CFO and say oh someone on my team will take care of the numbers Because then you can't understand the decision So again, this is this the virtuous circle of innovation around personal It is and again the idea is what at the business school What we're trying to do is get people to become what we call not data scientists But data managers Understand the technology that's available to you and how you can leverage that to make a better decision We don't need you to be that data scientist that goes in and can do the statistical application on a data set What we need you to know how to do is ask a good question and know how to extract the appropriate information To answer that question and you should be able to can you know to converse with the data scientist the IT? Organization whoever understands the different technology platforms to get that information to you So with more data There's potential for more bias and this potential for more politics and decision-making How do you foresee you know from a strategy standpoint addressing that probably is that a problem? Oh, it's definitely a problem And it's actually kind of interesting because what we do first when we we deal with a company who comes into Who wants to start working with big data or with the students in the MBA program is to first teach them about critical thinking and By understanding critical thinking one of the first things that you look for is bias and alternatives And to be able to clearly articulate your problem statement and break it down and understand the subsets of questions The problems that could come from this the bias that's associated with it and the alternatives to the decision Once you're able to do that and understand that the core of the problem now We can start breaking it down and looking for the appropriate pieces of data to either support or refute the position So it has to start there then you bring the technology in now if the organization can't deal with critical thinking They have bigger issues than big data. So you start with a critical thinking process And make sure that those pieces are in place. Yes, absolutely And you provide guidance training Education on each of those piece parts Yes, because what we say is if we have all these things and you're able to do that It enables you to now utilize big data in any way shape or form to answer the question If you look at critical thinking the critical thinking process and understand how you have to clearly state the objectives of the question Well, that sets you up for speed to build up. Excuse me here develop your Unstructured text analytic models now. I have specificity in context now I can start going in querying my data whether it be structured or unstructured to try to bring that information together in the appropriate Context so it's a piecemeal approach you start high and work down into the details So once again, John we see process trumping technology Function always works. Hey, is there lock-in? Yeah, it works like Okay, go with it Mike I have a question for you on on Marsha our previous guest was talking about towards the end, you know her passion around people's centric focus The social web and social businesses people centric the humanization Love that I love that all day long. Well, one of the things she talked about earlier that I want to drill down with you is New waves bring in Dynamics that destroys old ways. So, what do you say to the folks out there that are both looking to change careers around? How to leverage their their their skill sets and to young people in high school who quite frankly are going through the book Oh, and where do I go to college? What should what career should I do and what major should I be in? Data science is not in the side on the radar. Do I go math? Do they go physics? Do they go creative all the above? What's your advice to the young people out there and also folks changing jobs? Do you do you like problem-solving if you like problem-solving? You're going to enjoy your career where you start to look at information and putting things together To solve puzzles and to work on mysteries those sorts of things and you can become pretty much agnostic With regard to what vertical you want to go and you may have a pre Disposition to be in one field or another But overall if you're naturally curious and you want to find out why things work the way they do Then the field of big data is just great for you any advice on disciplines. I mean is it do a little creativity? Well as an economics, I mean This is a hard road map for young people to to navigate. Is it a DB DB DB guy database guy Dba well, it's kind of financial, you know, it's truthfully You know, I I'm an immunologist biochemist that does this and worked in a number of different industries Before I started down this path of looking at how people ask questions and make decisions So I guess my view of it again born with it You're either born a data scientist. I Think it comes down to this this pre-disposition for problem-solving and understanding that so I mean we see musicians Come in and be amazing. We've seen folks with no multiple languages be great data scientists We've seen math geeks. So think about you know, again, do you like to look at again puzzles? Do you like to organize things? Do you like to see how they you know sort so to speak in your head? I it all comes down to yes I think there's a Predelection to doing these sorts of things and there are certain disciplines that reinforce that sort of thinking So if you start thinking about STEM disciplines, they there is a structure that they apply to it That's not to say that you don't have the same thing in the language skills Remember I do unstructured text analytics when I first started in science. I I teased my mother who was a English teacher that I was going to delete the label arts gene. Well, what am I doing now? I'm doing unstructured text analytics based on ontologies and language in order to extract data So so what do you do with IBM? What do I do with IBM? What I do with IBM is I work with both the hardware software and Watson groups to figure out new ways of applying their Technologies to business decision-making that is what sort of toolings do we need to use? How can we make a better interface for the business user? Because one of the big things again in this democratization of data is that everyone's got to have access to it know how to use it So what we do in working with the students and the companies is we collect a lot of feedback about what works What doesn't work? What do they have to change in the software? The same time we have new paradigms on how you deal with data and how you have to handle it from a Technology point of view. How do you store it? How do you manage memory those sorts of things? So not only is it me in the business school with the business question I have partners in the computer science question looking at memory and storage paradigms as well as the engineering department to look at Things like FPGA performance and programming so what we like to do is work with IBM and give them a continuum from the Design stage all the way to the business application stage. Yeah, so we were talking we were to Oracle open world You know a couple months ago now and it came out that the average age of the enterprise app is like 18 or 19 years old So we're talking today about social business. I guarantee most of those apps are none of those apps have anything that looks like social we were talking earlier with Marsha about how Applications have to change the whole interface has to change the user experience has to change I mean it feels like it's a complete do-over in terms of the interface in the experience I think with the interface in the experience. Yes, but you know fundamentally and this is one of the discussions We I have a lot with the IBM group is how do we look at data and how can we start treating data as objects? Because at the application state what I'm doing is I'm trying to figure out the best way to get input and Then once I have that input what form can I hold it in so now that I can rearrange it and use it in any Way shape or form that I want and some of the things that we have even though they're 20 years old are very good at that So how do I how can I take that and not be so constrained in structure that allow me to rearrange data? Reorganize it and utilize it in new forms. So at the end of the day Yes, there's some very good things that have that have been around for a long time and we can leverage them And we have technologies now that can really accelerate look at the DB to blue stuff and how they're doing acceleration there Okay, that's a great form of Organization now, how do I bring unstructured information into it? How do I organize that and how do I leverage that? So again, it's more of an interface question Yeah, like I say treat it as an object as opposed to some inane concept. I can't really understand that's decomposed into a block That's somewhere somehow Okay, cool All right, so final question for you is what's the outlook for in your opinion the future of decision support Do you seeing automation personalization user interface? What are some of the things that you can point to that you say? Hey, you know, we're looking that it's on the fringe It's on the the lunatic fringe to you know, mainstream innovation. That's a right around the corner What do you see those two areas? Well deeper on the corner lunatic fringe to to Like just right around the corner. So it could be automation AI Learning machines, etc. Well, and again, I think when you start all of these things are starting to converge and the question is how much do you want the machine to do versus how much do you want the human to do and Where we want the machine to do a lot of the road stuff the heavy lifting of data and information We want that we that's good for the machine to do It's better for the human to focus on the decision-making part of the program and understand because truthfully even though we talk about Cognitive computing I haven't met a machine that can out do the mind with regard to cognitive computing But what can we do to get it to focus on the mind on decision-making and less worried about bringing the information in? So we drive the technology with regard to getting it to understand the context and specificity Necessary for the cognitive event that the human has to make I've always been fascinated with artificial intelligence and Entologies and cognitive and learning machines. I think it's the what Watson has proven to the world is you can actually put a You know put a face to that with Watson saying hey we can actually build this real-time analytic brains into apps I think that's something that we're watching really closely Michael. Thanks so much on the for coming on the queue really appreciate Thanks for your insight and good luck with your work at NC State and great to have you on This is the cubes exclusive coverage of IBM's information on demand. We'll write back after this short break