Observation of Single Top Quark Production at the DZero Experiment

Using $1~\rm{fb^{-1}}$ of data, D0 reported an evidence for the production of single top quarks at 3 standard deviations, in 2007. Three different state-of-the-art multivariate techniques: Decision Trees (DT), Matrix Elements (ME), and Bayesian Neural Networks (BNN), were used to identify the small signal from the overwhelming backgrounds. A weighted average of measurements from all three analyses was then done via the Best Linear Unbiased Estimate (BLUE) to further enhance the sensitivity. Here we present a multivariate approach to combining results, which takes as input the output discriminants of the DT, ME, and BNN analyses. Preliminary results suggest a significant improvement upon the simple weighting technique.