Carbon based materials have been explored for cathode applications in Lithium ion batteries because of their low environmental impact, easy availability of carbon and the rich chemistry of carbon based compounds which allows their properties to be tailored using functionalization and doping. Prototype devices based on carbon nanotubes and graphene which show superior combination of energy and power density have already been demonstrated. The calculation of redox potentials using DFT method to search for promising materials, is a time and resource expensive methodology. Consequently, correlations between redox potentials and electronic properties such as HOMO, LUMO and electron affinity are being explored so as to identify promising materials based on electronic properties which are relatively easily calculated. For DFT based computations of redox potentials, implicit solvent models such as Poisson-Boltzmann solver are used to compute solvation effects, at a reference temperature. The dielectric constant and probe radius of the solvent are inputs for solvation energy calculations. In this work, we intend to train a supervised machine learning algorithm, based on the structures in a given class of material, and investigate if a sophisticated machine learning based model can be used to predict redox potentials based on electronic properties from DFT calculations. It is anticipated that a machine learning model can be used as a rapid screening tool for identifying and designing novel carbon based materials for redox applications.