Abstract:
The data deluge in Astronomy has led to the birth and fast growing of Astroinformatics. We present a work within the framework of such new discipline, related to the applicationof a machine learning method to evaluate photometric redshifts and their Probability Density Functions, used in wide survey projects like EUCLID, and LSST. A second part of the work is dedicated to the application of reliable photo-z's to the measurement of Weak Lensing. The Excess Surface Density is, in fact, proportional to the tangential shear, through a geometrical factor directly related to the measures of accurate photo-z's. We therefore evaluated the impact of photo-z quality by comparing SED template fitting with astroinformatics method.