Clustering of Online Stores from Suppliers, Point of View: Using Clusters Number Optimization in Two-Level SOM

By developing information technology and emerging online markets, planning for these markets and analyzing their details has become a priority for beneficiary organizations. One of the most important online markets in Iran is SIM card credit services that are considered as online stores. Regarding numerous and increasing number of online malls, grouping and classification of online malls from the supplier's point of view is essential to forecast cooperation. In this paper, around three thousand online stores have been studied and analyzed by using the data from one of famous suppliers and they have been clustered according to supplier's indexes. Clustering process has been done using SOM Neural Network in two levels by K-means algorithm since it facilitates analysis of clusters. Although various validation indicators have been developed to determine the best number of clusters. In this paper, an optimizing model of compensatory approach to indexes is presented through combination of multi criteria decision making and aggregation of various indexes.