Optimization, statistics, big data and business analytics
Last month, we attended the INFORMS 2012 conference in order to learn more about current activities in the field of business analytics, and to present the results of some of the work we've done in this area. The meeting kicked off with a series of interesting technology workshops run by commercial companies as a means of promulgating their software systems; the main insight I got from them was the importance that the community places on high-quality optimization solvers in areas like prescriptive analytics, in which quantitative methods are employed to help make better decisions in business.
The NAG Library contains a variety of optimization routines (for both local and global minimization) - along with, of course, a wide range of solvers for other types of problems in analytics (such as statistical analysis, correlation and regression modelling and time series analysis) and in a variety of other numerical areas. At the conference, we presented the results of some consultancy work performed for a client who was using NAG routines to solve a large-scale constrained optimization problem arising from activities such as price promotion (see abstract 4 on this list for more details).
Optimization for a Client with Large-Scale Constrained Problems: A Case Study, on display at the INFORMS 2012 poster session.
The remainder of the conference consisted of a couple of plenary talks (from Google and eBay), an entertaining panel discussion on the perennial topic of Big Data, and a collection of contributed talks which were arranged in fifteen parallel sessions on topics like The Analytics Process, Decision Analytics, Analytics Around Us, etc. I found the standard of the presentations to be very high; a few personal highlights were:
- Google’s Hal Varian describing the use of their Insights for Search tool, which can be used to identify market trends based on the frequency of searches for specific terms.
- LinkedIn’s Scott Nicholson discussing their analysis of personal data entered by its users, and the way that’s used to build improved user experiences. There's a detailed article about this interesting talk here.
- End-to-End Analytics’ Colin Kessinger drawing a distinction between analysis and decision-making, emphasizing the importance of explaining to the client why the answer is the correct one, and the importance of multiple iterations in the analytics process.
- eBay's Bob Page speaking about eBay's perspective on consumer behaviour, and the increasing importance of mobile technologies. He described how mobile was 'driving engagement' and, as if to provide an illustration of his point, I found myself downloading the eBay app whilst he was talking about it.
- 4i Inc’s Eugene Roytburg giving his impression of the future of analytics and the way it's being increasingly used to drive business decisions. His discussion of the analyst's technical toolset included a reference to NAG, which was pleasing.
- AMPL’s Robert Fourer presenting how the AMPL modelling language works, and describing its advantages and future directions in development.
- DemandTec's Suzanne Valentine talking about their techniques for structuring and analyzing large-scale consumer data. DemandTec, who are NAG users, were recently acquired by IBM as part of their so-called Smarter Commerce initiative.
Our poster presentation was well-received, and the meeting gave us many opportunities for making useful contacts in the field. The conference location - where a weather forecast of 'partly cloudy' apparently meant 'a dazzlingly bright blue sky with a tiny little cloud over San Diego' - wasn't too shabby either.