Visualization of customer value by using self-organizing map (Data from a restaurant in Iran)

Due to the importance of using visualization to demonstrate information including organization financial data and to anticipate the cases such as customer value, this article seeks to predict customer value according to the restaurant data in Iran (Tehran) through visual representation in the second half of the year 1393, equivalent to the year 2014. The statistical population of this research is the restaurant clients who were given specific participation codes and the data regarding 20364 customers was collected randomly among them.For data analysis of restaurant database, initially via using clementine 12.0 clustering software, we clustered data and calculated the customer value, and then we displayed it visually by Matlab neural network toolbox software.According to the results it was found that in the cluster 5 from month 8 to month 12 there was an ascending trend and from month 9 to month 12 in cluster 1, the trend was descending and for the other months it had a steady growth. In this article we have tried to consider all the factors affecting customer value. Finally, this research emphasizes that visual representation of data plays an important role in presenting data in a simple and understandable manner.

Keywords:

visualization , customer value , clustering , neural networks , self-organizing maps