Water treeing is a phenomenon whereby the polymeric cable suffers serious degradation under combined stresses of electrical, thermal and chemical, thus shortening the lifespan of the cable. A high electrical field stress present at the tip of a water tree structure would degrade the surrounding healthy polymeric insulation, leading to electrical treeing and subsequent breakdown of the water tree affected polymeric cable. This paper aims at proposing methods to estimate the electric field stress ratios of water tree structures present in the insulation of water tree affected polymeric cables. The capacitances and lengths of water tree structures present the insulation were measured in water tree affected polymeric samples over a period. The measured data were then used in the proposed methodology to estimate the maximum electric field stress ratios present at the tips of the water tree structures, using electrostatic finite element software. A database of the electric field stress ratios was compiled and subsequently fed into the neural network's back propagation process. With the optimisation of weights in the neural network's back propagation process, an averaged error of less than 5% was found. This newly developed intelligent system provides a simple, time saving and practical way of obtaining electric field ratios.