Particle identification in the NA48 experiment using neural networks

The possibility of applying Feed-Forward Neural Network for particle identification in the experiment NA48 is investigated. It is shown that exploiting all the information available from the Liquid Krypton Electromagnetic Calorimeter, the efficiency for recognition of electrons and pions can be improved significantly. Using the method developed, it is possible to fulfill the NA48 programme for investigation of rare kaon decays without employing an additional experimental equipment.