KDnuggets Home » News » 2014 » Aug » Opinions, Interviews, Reports » Big Data Innovation Summit 2014 Toronto: Day 1 Highlights ( 16:n16 )

Big Data Innovation Summit 2014 Toronto: Day 1 Highlights


Highlights from the presentations by Big Data leaders from TD Bank, Public Health Ontario and First Nations Education Steering Committee on day 1 of Big Data Innovation Summit 2014 in Toronto, Canada.



As many organizations are now working with unmanageably large data sets, the importance of using and maintaining an analytics platform which can cope with this scale of information is essential. This presents both a challenge and opportunity as organizations must identify patterns and gain actionable results in order to gain a crucial advantage over competitors. Big Data Innovation will help businesses understand & utilize data-driven strategies and discover what disciplines will change because of the advent of data. With a vast amount of data now available, modern businesses are faced with the challenge of storage, management, analysis, visualization, security and disruptive tools & technologies.

Big Data 2014 TorontoThe Big Data Innovation Summit (June 4 & 5, 2014) was organized by the Innovation Enterprise at Toronto, Canada. Illustrated intermittently with case studies, interactive panel sessions and deep-dive discussions, this summit offered solutions and insight from the leaders operating in the Big Data space.

We provide here a summary of selected talks along with the key takeaways.

Here are highlights from Day 1 (Wednesday, June 4, 2014):

Leila LavaeeLeila Lavaee, Data Analytics Lead, Digital Strategy & Customer Experience, TD Bank gave a thought-provoking talk on "'Big Data' Mindset and How it Can Enhance Customer Experience". "Big Data" seems to be a no-brainer and so many more organizations are trying to run big data projects. But last year the focus was more on infrastructure and all costs associated with it. What seems to be missing in the bigger picture is the true value of big data which is if and only if it drives strategic insights. Based on Gartner; 64% of enterprises surveyed indicated they're deploying or planning big data. Yet, they still don't know what to do with it. Leila explained how to develop "big data" mindset in an organization and how to use it for enhancing customer experience.

She suggested that instead of "Big Data" we should rather think about "Right Data". Storing vast amounts of data in an extremely efficient manner does not benefit an enterprise if it isn't using that data to generate insights that drive marketing and business decisions. People have a narrow vision of Big Data, thinking of the term merely in the context of IT infrastructure, storage, cost, training, privacy, security, etc. Big Data is a shift in strategy, and not merely a technology.

She quoted the following from Accenture Technology Report - 2013:

"Analytics have always been a challenge, but partially because businesses have often conducted the process in an exploratory fashion—they've collected available data and then analyzed it, rather than assiduously determining what data will aid their business strategy and then ensuring that the right data is collected to analyze."

Citing Gartner reports, she mentioned that Big Data investments are steadily increasing. The most common business problems, for which Big Data is being leveraged, are: enhancing customer experience, improving process efficiency and designing new products / new business models.

The application of Big Data Mindset to Customer Experience needs seamless integration of Customer Experience surveys, enterprise data warehouse and web analytic tools. It is important to understand online behavior and demographics of customers participating in CEI (Customer Experience Index) surveys & Segmentation. Organizations need to develop personas-specific, concrete representations of target users and use this information from strategy to design to testing. Based on prior research and exploratory interviews, determine the key attributes differentiating each persona and the number of personas. In summary, she emphasized that: it is important to develop Big Data Mindset throughout the organization, which requires significant cultural change - transforming the enterprise to become insight-driven.

Jim TomJim Tom, Chief Information Officer, Public Health Ontario (PHO) gave a good overview of the data analytics challenges in managing healthcare, in his talk "Big Messy Data: The Case of Public Health". Public Health is concerned with monitoring the health of populations (that is, groupings of individuals) and with the measures that can be used to improve their health. The concerns include such short-term issues as identifying an outbreak of infectious disease as well as long-term issues such as development of diabetes. Data complexity ranges from relatively modest datasets from laboratory test results to heterogeneous sources such as the health system, pollution monitoring, climate records and behavioral evidence. Messiness includes timeliness, linkage between phenomena and semantics.

In this eclectic talk, Jim outlined the challenges and PHO’s plans to address the data management and analytic issues. Canada spends around $211 billion on healthcare, which comprises of 11.2% of its GDP. 30% of this money is spent on hospitals, 16% on Drugs and 15% on Physicians. All these expenses are growing year-over-year by around 2.5%. The current challenges include: communicable and infectious disease surveillance, "determinants of health", "co-infections", chronic diseases - time and combination of factors, genomics and how to change people's attitude and behavior towards personal health. Jim's vision for future includes setting up data federation and using Big Data to measure effectiveness of public health messages.

Alan KharaAlan Khara, Director, Information and Communication Technology, First Nations Education Steering Committee (FNESC) talked about the prevalent misconceptions around Big Data in his talk "Five Great Myths About Big Data". Big Data may eventually lead to the discovery of the ‘Theory of Everything’ in Physics one day; however, it will still not become a solution to everything in the world we live in. Following the wrong approach towards Big Data can cost a company, its reputation and its competitive edge in the market. Just knowing Attribution Models and Predictive Analytics is not enough. If you become a believer in one of the myths described below, you will end up with wrong numbers.

Big Data Myths:
  1. Volume is directly proportional to Information: A recent study by Harvard Business School suggests big-data investments fail to perform because most companies cannot handle the information they already have. Thus, instead of going crazy after data collection, companies should rather first focus on making the most of the data they already have.
  2. Structured data is better than Unstructured data: Generally the term structured data (SD) is applied to databases (DB) and unstructured data (UD) applies to everything else (text, audio, video, images). Though we are technologically better equipped to handle structured data, it is the unstructured data which has far greater magnitude of information.
  3. Sort it later policy: With storage getting cheaper day-by-day, companies now can just collect the data, and figure out what to do with it later. In fact, with many data sets, the law of diminishing returns starts to apply. In addition, the more data, the longer it takes to sort through it. Sooner than expected, the volume of data transforms from an opportunity to a challenge.
  4. Prediction Accuracy debate: It's human nature to think that something that is more specific is more accurate. In fact, in a lot of Analytics, prediction is reported without due importance to accuracy, which creates a false confidence, potentially leading to disastrous results. More specific is not always better. It is important to find the balance.
  5. Data is static – Big Data? We are used to the over-simplistic approach of treating data as a static entity. However, in the Big Data world, any particular data point might be reported differently by different institutions, or at different points in time, or by different people at those institutions. Also, the meaning of data changes over time.

 
Highlights from day 2.

Related: