A Hierarchical NeuroBayes-based Algorithm for Full Reconstruction of B Mesons at B Factories

摘要

We describe a new B-meson full reconstruction algorithm designed for theBelle experiment at the B-factory KEKB, an asymmetric e+e- collider thatcollected a data sample of 771.6 x 10^6 BBbar pairs during its running time. Tomaximize the number of reconstructed B decay channels, it utilizes ahierarchical reconstruction procedure and probabilistic calculus instead ofclassical selection cuts. The multivariate analysis package NeuroBayes was usedextensively to hold the balance between highest possible efficiency, robustnessand acceptable consumption of CPU time. In total, 1104 exclusive decay channels were reconstructed, employing 71neural networks altogether. Overall, we correctly reconstruct one B+/- or B0candidate in 0.28% or 0.18% of the BBbar events, respectively. Compared to thecut-based classical reconstruction algorithm used at the Belle experiment, thisis an improvement in efficiency by roughly a factor of 2, depending on theanalysis considered. The new framework also features the ability to choose the desired purity orefficiency of the fully reconstructed sample freely. If the same purity as forthe classical full reconstruction code is desired ~25%, the efficiency is stilllarger by nearly a factor of 2. If, on the other hand, the efficiency is chosenat a similar level as the classical full reconstruction, the purity rises from~25% to nearly 90%.