Moving from Data-Rich to Decision-Smart: INFORMS Conference The Business of Big Data 2014
We interview the co-chairs of INFORMS Conference The Business of Big Data 2014 (June 22-24, 2014) on Big Data maturity, opportunities assessment, analytics for operations research, conference agenda and more.
- Getting from data discovery to return on investment and real business value
- Bridging the gap between decision-makers, IT managers and analytics professionals
- Expediting the journey from business problem to analytics solution
- Selecting and using the right big data technologies
These challenges along with the new opportunities, success stories and case studies will be the focus of INFORMS Conference The Business of Big Data at San Jose, CA on June 22-24, 2014. The conference is a great opportunity for learning as well as networking with the leaders from industry and academia through a
seamless exchange of information, ideas and perspectives.
The impressive line-up of speakers, which includes Big Data leaders from various industries (including Bill Franks, Chief Analytics Officer, Teradata), will share the best practices through real-world case studies and tutorials on a wide variety of topics. The conference program includes keynote sessions, presentations across 3 tracks (Big Data Case Studies, Big Data 101 and Emerging Trends in Big Data), technology workshops, panel discussions, facilitated networking, gala receptions and exhibits. Thus, this conference is a must-attend for all business managers, decision-makers, IT managers and technologists, and analytics professionals.
Conference fact sheet: http://meetings2.informs.org/bigdata2014/bigdatatrifold.pdf
Conference website: http://meetings2.informs.org/bigdata2014/index.html
The organizing committee co-chairs are Diego Klabjan, Professor & Founding Director, Master of Science in Analytics, Northwestern University, and Margery H. Connor, Senior Operations Researcher, Advanced Analytics Chevron Corp.
Here is my interview with them regarding current Big Data opportunities and challenges, and the conference.
Anmol Rajpurohit: Q1. How do you define Big Data? In terms of technology adoption life-cycle, what do you consider as the current status of Big Data?
Diego Klabjan: For decades relational databases and SQL have been ruling the world. Then came the proliferation of the web and social media. Progressive corporations have realized that SQL cannot handle these new data sources. As a result the technology behind big data has been developed. My definition of big data is that every data set that cannot be analyzed by using relational database technologies is big data. This includes large and quick data such as tweets that simply cannot be handled by relational databases, analyses that are possible with SQL but it would take too much computing time, and problems that would not be economical on relational databases due to cost constraints.
Big data is definitely past early innovators and in the prime of early adopters. I want to point out that there are still several companies that use big data technologies for the sake of using it and having bragging rights. If we include such cases, then we are deeply into the early adoption phase. On the other hand, there are not many big data deployments with a positive ROI, and thus from this perspective we are barely beyond early innovators.
AR: Q2. What is the impact of Big Data trends on the field of Operations Research? What are the biggest opportunities and major challenges?
DK: Operations research has always had impact in data science, in particular the fields of optimization, statistics, and data mining. On the technical side, big data spawned the search for different algorithms able to scale and tailored for the new technologies. Operations research has also always been very strong in analyzing processes and articulating business value. While manufacturing and related industries are often listed as original industries for such studies, big data created process changes in the IT and other organizational units and thus operations research practitioners and scholars have started adapting lessons learned to the data science and big data space.
While the algorithmic side of machine learning algorithms is well studied and algorithms have pretty much became a commodity, identifying problems with positive ROI remains challenging and a great opportunity for operations research studies. In particular data discovery which has become the biggest selling point of big data by its nature has challenges when it comes to ROI. Searching for unknown in the data can at most lead to hypothetical business values.
AR: Q3. Operations Research and Optimization has not received sufficient attention in Predictive Analytics Community. What are the key ideas and algorithms that Data Scientists should know about Operations Research and why?
DK: In many areas the line between operations research, and predictive analytics and machine learning is blurry. For example, regression has been studied by all these communities, and also statistics. Other areas such as classification falls more under machine learning than operations research. On the other hand, optimization is an area where operations research has historically been very strong. For decades the primary applications of sophisticated optimization algorithms have been select military, transportation, and telecommunication problems. In recent years the focus has been equally important to ‘porting’ these state-of-the-art algorithms to machine learning related problems.
AR: Q4. What are the most important questions (or issues) from industry and academia that are on the agenda for INFORMS The Business of Big Data 2014 Conference?
DK: The most pressing issue is definitely how to identify ROI in big data projects. This is an extremely challenging problem since big data is mostly used for data discovery which by nature implies that the business value is hovering in clouds (true clouds and not computing clouds). We will also highlight key emerging technologies to educate decision makers about what to expect in the future and plan accordingly.
We hope that the attendees after attending the meeting will have a much better view about best practices, what problems are best for big data, and what is in the pipeline when it comes to technologies.
AR: Q5. Tell us about the Technology Workshops on the first day of conference. Is there an additional price for attending these workshops?
Margery H. Connor: The technology workshops are 90 minutes long and provide our Exhibitors the opportunity to do a deeper dive into their technology solutions. The workshops also provide the attendees a chance to have a technical discussion with the vendors and hear the questions and perspectives of other attendees. Some workshops are hands-on. Leading analytics companies will be showcasing their software solutions and approaches. There is no additional charge for these workshops.