THE APPLICATION OF THE EXPECTATION MAXIMIZATION ALGORITHM ONTO BIG DATA
Jason Beffel · Carroll Collected (John Carroll University) · 2015
With the growing prevalence of big data, it is interesting to see whether the popular machine-learning tool known as the Expectation Maximization (EM) algorithm (Dempster, Laird, & Rubin, 1977) is applicable and beneficial for data analysis. Can the EM algorithm be used directly on big data without requiring additional changes or adjustments? Yes, the EM algorithm can be applied to a big data set modified or unmodified, and the goal of this paper is to show the technique used to do so.