International Journal of Emerging Technology and Advanced Engineering Website: www.ijetae.com (ISSN 2250-2459, ISO 9001:2008 Certified Journal, Volume 3, Issue 1, January 2013) 542 BPL Classification Using Multilayer Perceptron: Performance Analysis Preksha Pareek 1 , Prof. Prema K. V. 2 1 Student M.Tech Computer Science (MITS-FET, Lakshmangarh) 2 Professor computer science (MITS-FET, Lakshmangarh) Abstract— Artificial Neural networks have emerged as an important tool for classification. Many neural network models have been proposed for pattern classification, function approximation and regression problems. Among them, the class of multi-layer perceptron networks is most popular. This paper presents an approach for classifying a person as below poverty line or not using Multilayer feed forward neural networks. This network is trained by varying different parameters like learning rate and number of iterations for determining performance level of the network constructed and it describes which training functions and learning parameters are suitable for the application of BPL(Below Poverty Line) classification. Keywords— Artificial Neural Network, Multi Layer feed forward network, Back propagation algorithm, Below Poverty Line . I. INTRODUCTION National estimates of the percentage of the population lying below the poverty line are based on surveys of sub- groups, with the results weighted by the number of people in each group. In India BPL analysis is based on the degree of deprivation in respect of 13 parameter[8]. The need for targeting the poor was felt when economic growth in developing countries benefited relatively developed areas and better-off sections of the population, and more or less bypassed backward areas and poorer sections of the population[8]. Parameters for classifying poverty are: (1) Land holdings :- This is further divided into (i) Nil holding (ii) Less than1 ha (iii) More than 1 and less than 2 ha. Then (2) Types of house: This parameter contains value as (i) Nil house (ii) Kutcha house (iii) Pucca house. (3) Availability of clothing contains values (i) Less than 2 pairs (ii) More than 2 but less than 4 pairs (iii) More than 4 but less than 6. (4) Food security: (i) Less than 1 meal per day in major part of year (ii) Normal 1 meal but sometimes less (iii) Normal 1 meal throughout the year. (5) Sanitation: (i) open (ii) group bathrooms with irregular water supply (iii) group bathrooms with regular water supply. (6) Consumable durables (T.V., electric fan, kitchen appliances): (i) nil (ii) Any one item (iii) Any 2 item . (7) literacy status of highest literate: (i) illiterate (ii) upto primary (iii) completed secondary. (8) Status of household labour: (i) bonded labour (ii) women and child labour (iii) adult males. (9) Means of livelihood: (i) Casual labour (ii) subsistence cultivation (iii) artisan. (10) Status of children: (i) not going to school and working (ii) going to school and working (iii) going to school and not working. (11) Types of indebtedness: (i) daily consumption purpose from normal sources (ii) production purpose from normal sources (iii) for other purpose from normal sources.