THE USE OF NEURAL NETWORK CLASSIFIERS FOR HIGGS SEARCHES
ALESSANDRO PETROLINI · International Journal of Modern Physics C · 1992
Neural Network Classifiers are used to separate the signal from the background in a High-Energy Physics problem. The basic principles of Multidimensional Analysis by means of Back-Propagation Neural Network Classifiers and the application to the Higgs search are discussed. A search for the Minimal Standard Model Higgs boson, through the reaction [Formula: see text], using the data collected in 1990 by the DELPHI detector at LEP is made. The technique used allows to reach good detection efficiencies and no evidence of the Minimal Standard Model Higgs boson with mass less than 37 GeV is found. The use of Neural Network Classifiers proves to be a very powerful classification method. A comparison of the method with standard analysis techniques is presented.