Clustering analysis and supervised methods for antiparticle studies in the PAMELA experiment

In this study a new approach to pattern recognition problems in astroparticle physics is presented. The context in which this work has been developed is the satellite borne experiment PAMELA, whose principal aim is antiparticle studies. In particular the classification problem of the PAMELA imaging calorimeter has been taken into account. This detector is designed for particle identification; due to its high granularity, both in the transversal and in the longitudinal direction, the calorimeter is suitable for reconstructing the spatial development of a shower. For each event the calorimeter is able to provide a 3D image that can be used to discriminate between hadrons and leptons. In this work the available information for each kind of image event class has been pre-processed representing each event by means of discriminating variables. A clustering analysis has been applied to a data set and the classification has been performed using supervised algorithms. Results from simulated data from the PAMELA prototype calorimeter will be shown.