Statistical Analysis of Massive Data Streams: Proceedings of a Workshop (2004)
Chapter: Concluding Comments
Concluding Comments
Among the research topics presented at this meeting were remote sensing for climate modeling, computer or network intrusion detection monitoring, records from communication networks or Web logs, e-commerce recommendation systems, and real-time imaging problems arising in robotic vision. By design, none of the presentations is so broad that it touches all five major areas of research. However, common threads did emerge, such as statistical modeling, visualization, and the need to store, move, and manipulate massive data streams, which are nontrivial challenges.
The workshop explored both theoretical aspects of dealing with massive data streams as well as practical means of effective analysis of such streams. For many massive data streams, there is no model of the data and little a priori knowledge, and the workshop presenters demonstrated the vital role that statistical modeling, analysis, and visualization play in enabling the essence of otherwise hidden information to be distilled from these data streams.
This is an exciting new realm for statisticians, motivated by a rapid increase in streaming data. The Committee on Applied and Theoretical Statistics (CATS) has already made plans for a workshop on visualization of uncertain information, to be held in the winter of 2005, and the Committee will explore other dimensions of these challenges in the coming years. Please check out the CATS Web site for future events.
