Predict consumer behavior in selecting mobile SIM using artificial neural networks (Case study: mobile phone subscribers in the city of Urmia)

This study sought to develop a model of artificial neural networks that is able to hold variables that influenceconsumer behavior and finally predict the type of SIM that consumer demands. Research's Statistical community is allmobile subscribers in the city of Urmia, in which 384 mobile subscribers have chosen to survey by sampling method. Formodeling and analysis of multi-layer neural network with hyperbolic tangent function, of education, training has been usedby the Forward algorithm that results showed the importance of independent variables in predicting the dependent variable(SIM card mobile phone) to arrange the variables of price, product, rank first and third in importance and the promotionof social class ranked fifth in importance. Gender, education and marital status have the least impact on selecting the typeof SIM card. The overall extension of the model obtained 65.7 percent of the SIM card used in the test samples correctlypredicted.

Keywords:

mobile , artificial neural networks , consumer behavior , predict , MCI , MTN