market segmentation using neural networks (case study: pharmaceutical market in iran)

like other sciences, market segmentation, as a science, seeks to realize unique needs of human being. in health market, individual diagnostic and curative methods are practiced more than anytime. however, a foundation of this market, pharmaceutical market, is not as advanced as other products and services markets in terms of segmentation. regarding the complexity of pharmaceutical market, it is essential to utilize an appropriate technique to segment this market. the research aims to design neural networks-based mathematical model for iran medicine market that segments it effectively according to multi criteria, and identifies differences among segments. according to data contexts, many possible models were tested for som networks and the model with most competencies to segment market was chosen as effective pharmaceutical market segmentation. such architecture identifies six different segments among medicine market consumers. these segments are distinguishable from the principle segmentation factors point of view. in addition, the performance of the designed model was measured by comparing the results with the results of data segmentation using a classic clustering method (k-means).