Election Result Forecasting Using Two Layer Perceptron Network

Neural networks are increasingly used to solve highly non linear control problems. The current paper addresses the problem of forecasting the result of general elections in India. The neural network is first made to learn and then the trained network is made to forecast the result of the election. While training the network minimal disturbance principle was followed, which suggests that during training it is advisable to inject new information into the network in such a manner that it disturbs the stored information to the smallest extent possible.