A new multivariate technique for top quark search Original Research Article

We present a new multivariate event classifier, the PDE classifier, which can be used to identify rare signals in the presence of large backgrounds. The classifier is based on a discriminant function which uses kernel density estimates of the signal and background event densities. The PDE method offers the flexibility of a neural network but is conceptually simpler, theoretically better understood, and easier to optimize in practice. The technique is tested on Monte Carlo data using processes tt̄ → e + Et + jets as signal and W + jets → e + Et + jets as background. The performance of the PDE classifier is similar to that of a neural network. Both methods outperform conventional analysis.