Saturday, October 19, 2013

Shouting From the Podium: Data Mining Takes Centre Stage

This past Wednesday, SAS Canada's offices was once again pleasantly overrun by data mining practitioners. I often find myself wondering what the collective, cumulative experience of the attendees must be. Hundreds of years? It must be, at a minimum. On top of it, these are some of the best and brightest from a wide range of industries. Whether business analysts in financial services, actuarial professionals from insurance companies, health analysts from hospitals and the government, these folks certainly bring an impressive array of skills to bear. It's a bit intimidating to be in the same room, I tell you! The only thing to do is deliver a compelling agenda which engages and entertains. I'd like to think we were able to do so at the most recent meeting.

I was privileged to spend time with our keynote speaker in a social context before and after the meeting. Krzysztof Dzieciolowski of Rogers Communications certainly turned out to be something of a star draw. There were former and current colleagues of his scattered throughout the room, and my own experiences with him made it clear why this would be the case. Not only was he a brilliant man - in his spare time, he had previously chaired the Stastistical Society of Canada in Quebec and he currently taught data mining at Concordia University - but he was an entirely pleasant and cheerful person to talk with around a wide range of topics. Over dinner, we discussed our shared Polish heritage, tales of football glory and heroes of the past, the challenges of teaching and implementing data mining solutions... and much more. The time just flew by!

As a speaker, Krzysztof offered a tremendous overview of a complex and interesting topic. Entitled 'Scoring Models, Propensity Transformations and Model Scoring Using SAS', the talk was balanced with a lot of theory combined with practical examples. Given the interest of the audience in asking him questions - I nearly had to pull him out of a well-wishing mob of admirers at the break - I'd say his talk was an unqualified success.

Carrie Dang of Loyalty One was our second presenter, and I can honestly say that I've rarely felt worse for a presenter than I did for poor Carrie. Her topic was riveting, her talk well put together... and her voice was soft and soothing. Developing a scoring model for email campaigns is something that all marketers can relate to and I found myself nodding along with her talk. Ordinarily this is a recipe for great success, but Carrie had the misfortune of dealing with a problem which has become all too common at my meetings: technological failure. Our audio system dropped entirely, and I'm afraid anyone beyond the first 20 rows may not have heard a thing. I apologized profusely to both Carrie and the audience and I appreciate that the quality of the experience must have been significantly negatively affected. For what it's worth I've escalated this technological issue internally and been assured that it will be addressed for future meetings. I've extended an offer to Carrie to take her out for lunch by way of saying 'thank you' and 'I'm sorry'. I hope that it's a start, at least!

Our last presenter of the day was SAS Canada's own Lorne Rothman who has been gracing our stage quite a bit lately. Lorne is such a great, natural presenter that if I could somehow wrangle him into giving a talk at every meeting, I certainly would! He instantly had the crowd relaxed by opining that 'unlike the first two talks, mine isn't interesting'. Great stuff. Lorne's talk was focused around getting the most out of SAS Enterprise Miner as a SAS programmer, and he really showed the flexibility of the code node for all it was worth. I was especially grateful to Lorne as he filled an agenda hole which suddenly and unexpectedly developed. I confess to breathing a sigh of relieve when his topic was so well-received.

As time as gone on in my SAS career I have noticed that data mining has taken centre stage in media publications and public thought. As analytics have become less of a black box and more of an indispensable business requirement, the practitioners who are able to blend the art and science of mathematics, psychology, sociology and business savvy are becoming more and more critical to successful business processes. It's a real privilege to be so deeply involved with this group of individuals.
The presentations will be posted very shortly at the Toronto Data Mining Forum website and you'll be able to access them there shortly. For me, however, there were very little time to pause and reflect. I was about to fly out to Halifax for their user group meeting, one which would drive home the importance and strength of SAS community in a much different way. I'll have more about that meeting in an upcoming post.

Until then...

No comments:

Post a Comment