Abstract: Below Poverty Line (BPL) is an economic benchmark and poverty threshold used by the government to indicate economic disadvantage and to identify individuals and households in need of government assistance and aid. In this paper a methodology is shown to classify as BPL or non BPL using Artificial Neural Network. Multi-Layer Perceptron (MLP) network will be used for classification. This classification will take very less time and provides efficient results. Further the effect of network architecture on performance is shown on this application.