Big Data, Data Science, Internet of things (IoT),… these are terms that are interchangeably bouncing around the media all the time these days. The hype is at a fever pitch! Very many people are jumping onto the bandwagon or trying to do so. Governments and businesses world-wide are even betting on this concept to help the world recover from the most destructive economic slump since the Great Depression, in the same manner as the rescue act performed by the Information Technology revolution in the 90’s. The hottest job title in Silicon Valley these days seems to be that of “Data Scientist”. Finland on the other hand has been “relatively” slow to adapt to this new reality. Possibly due to an abundance of caution resulting from the recent problems of Nokia.
So is the hype really justified? Have we not heard this all before? Consider nanotechnology-an area with which I am intimately familiar. Focused research into nano has been going on for close to 30 years now. Billions of Euros have been and are being spent on research in this area. Why has nanotechnology as of yet produced the massive revolution in industry that it once promised? To be sure, nano has and will continue to make progress and produce new and exciting products in the future. But it is not quite there yet. One of the big issues in nanotechnology currently is the difficulty in making the transition from lab-scale to commercial-scale production. The much talked about nanomaterial Graphene is a prime example of this challenge. It is a material which will revolutionize electronics just as the solid-state transistor did, decades ago. But at the moment, producing high-quality graphene in even small quantities is just not possible. It is thus not yet commercially viable. With nanotechnology, one needs to find ways to work with the laws of nature. And the laws of nature can be very annoying and complicated! Additionally, health and safety concerns need to be addressed from the environmental, medical and legislative viewpoints before nanomaterials will be allowed out into the world en masse. Compared to the challenges of taming the Big Data beast, the nanotechnology beast is as yet currently, much wilder. The beast is being tamed, but very slowly. To be sure, once it is tamed it will be truly revolutionary, but that will still take some more time.
As an aside, the synergy of Big Data and nanotechnology is something we can marvel at. Envision a future where nano-sensors are a practical reality. Instead of data from one sensor, imagine being able to get data from a million sensors for the same cost! The accuracy of data based decision making would increase exponentially. Then we would be talking Really Really Big Data! But for the moment, that is still in the conceptual stage. Currently, industry has not invested massively in nanotechnology. The investments are currently mainly coming from the government sector. In the current economic climate, more immediate results are demanded by corporate shareholders.
What of Big Data/Data Science/IoT then? The proposition is that the Big Data revolution will add value to economic activity in a shorter time-scale than nano will. The rewards of Big Data promise to be more immediate and can build upon available infrastructures. It only makes sense that the more information we have, the better decisions we can make. This is why the American NSA invested on the order of 56 billion USD last year on data gathering and analysis-in fact they have been doing “Big Data” for many years now. To give some examples of the use of data science in business, Walmart-the 3rd largest employer in the world after the US defense department and the Peoples Liberation Army of China-is going full speed with Big Data. Other top-caliber companies such as GE, and IBM have all started investing in Big Data in a big way. When so many big players invest in something, it is with the expectation of results, and the sensible thing would be to stand up and take notice. Is Big Data just limited to the big players? No it is not anymore. In fact, the true revolution is that these Big Data technologies are now within the reach of smaller players. The numbers put out there are very compelling. The “Field Guide to Data Science (2013)” published by the well-respected management consulting firm Booz Allen Hamilton gives the following data which should make any manager stand up and take notice.
17-49% increase in productivity when organizations increase data usability by 10%
11-42% return on assets (ROA) when organizations increase data access by 10%
241% increase in return on investments (ROI) when organizations use big data to
1000% increase in ROI when deploying analytics across most of the organization, aligning daily operations with senior management's goals, and incorporating big data
5-6% performance improvement for organizations making data-driven decisions.
The leading challenge of Big Data currently appears to be that of an acute skills shortage. If one is to believe this widely circulated roadmap to becoming a data scientist, then one needs to have an extensive list of intellectually demanding skills to become a full-blow data scientist. The statistician, the machine learning expert, the programmer all have something to contribute immediately. But the nature of the Big Data/Data Science/IoT beast is such that in order to fully exploit its potential, it also requires people with skill sets beyond the traditional ones. If one is talking about the IoT for example, knowing how the sensors that get the data work can be critical in preventing misinterpretation of the data. For, example, for people of my own background there is an exclusive Silicon Valley data scientist training program called the Insight Data Science Fellows postdoctoral program which is rumored to be harder to get admitted to than getting into Harvard University! Many of the biggest players in Silicon Valley are in on it, Facebook, Linkedin, Microsoft etc… are all in there. Once you complete this program, the rate of job placement is 100%. Most people in the Insight program mentioned above are physicists. But there are also people from computational biology, chemistry etc.. Essentially, people who are skilled with math, computers, data handling, presenting results are all needed.
Finland has a clear advantage in the form of a highly educated, disciplined workforce with strong skills in science and technology. Indeed, currently there are large numbers of highly skilled, highly motivated people suffering from unemployment-a huge waste of the resources spent on educating them. And, much has been written recently about the loss of competitiveness suffered by Finland recently. The business sector should take the bull by the horns and join and even take a leading role in this revolution. The message for companies in Finland should be crystal clear. Move fast or be a sitting duck! Adapt to the Big Data revolution or die.
Nissanka Wickremasinghe is a physicist and a Data Science enthusiast. He hails from Sri Lanka and moved to Finland recently with his Finnish wife a few years ago. He was most recently a Marie Curie experienced researcher at VTT, Espoo where he worked on projects involving printed biosensors as well as RFID sensors. He has previously worked as a researcher at EPFL in Lausanne, Switzerland. He obtained his PhD in physics in the area of bio-nano from Rice University, Houston, USA and his BSc in physics and mathematics from the University of Colombo, Sri Lanka.