System data is abundant, yet data-driven decision making is currently more of an art than a science. Many organizations rely on data analysis for problem detection and diagnosis, but the process continues to be custom and ad hoc. In this paper, we examine the analytics process undertaken by users to mine large data sets, and try to characterize these searches by the operations performed. Furthermore, we take a first stab at a methodical process to automatically suggest operations based on statistical analysis of previous searches performed.
National Science Foundation
Expeditions in Computing