Statistical Analysis of Massive Data Streams: Proceedings of a Workshop (2004)
Chapter: TRANSCRIPT OF PRESENTATION
Transcript of Presentation
MS. KELLER-MCNULTY: Okay, I would like to welcome everybody today. I am Sallie Keller-McNulty. I am the current chair of the Committee on Applied and Theoretical Statistics. This workshop is actually sponsored by CATS. That is the acronym for our committee. It is kind of a bit of a déja vu looking out into this room, back to 1995, the nucleus of people who held the first workshop, or at least attended the first workshop that CATS had, on the analysis of massive data sets. It has taken us a while to put a second workshop together. In fact, as CATS tried to think about what makes sense for a workshop today, that really deals with massive amounts of data, is where we decided we would really try to actually jump ahead a bit and try to look at problems of streaming data, massive data streams.
Now, the workshop committee, which consisted of David Scott, Lee Wilkinson, Bill DuMouchel and Jennifer Widom, when they started planning this, they were pretty comfortable with the concept of massive data streams.
I think that, by the time that this actually got together, they debated whether, instead of data streams, it should be data rivers. Several of you have asked me what constitutes a stream, how fast does the data have to flow. I am not qualified to answer that question, but I think our speakers throughout the day should be able to try to address what that means to them.
We need to give a really good thank you to our sponsors for this workshop, which is the Office of Naval Research and the National Security Agency. Now I will turn it over to Jim Schatz from NSA. He will give us an enlightening, boosting talk for the workshop.
