Use of neural networks in the analysis of τ → Ks0ντX inclusive decays at LEP

Neutral Networks (NNW) were used to tag the inclusive τ → Ks0ντX decay at the Z0 energy. Two feed-forward NNWs were trained with the backpropagation algorithm, to select, respectively, Z0 → τ+τ− events and Ks0 → π+π− decays. They were used to process the data collected with the DELPHI detector at LEP in 1993. The selection efficiency and purity achieved with the NNW at each step of the analysis were compared with those obtained using linear cuts.