Abstract:
Nowadays, marketing managers are more concerned with identifying and understanding customer behavior in the online space. Since the customers in online space are not visible, it is much essential to have more information about them to provide better services. Customer segmentation is one way to improve the customer problems in an online space. Identifying characteristics of customers and optimal resource allocation to them according to their value to the company is one of the major concerns in the field of customer relationship management and determining factors in E-business success. The purpose of this study is clustering customers online of a mobile sales website based on their lifetime value and RFM model. At the proposed framework in this study after determining the values of RFM model include recently, frequency and monetary of purchase and weighting them using Shannon entropy, a self-organizing map is applied to the segmentation of customers. The customers are categorized into four main segments and characteristics of customers online in each of the segments are identified. Mobile sales website customers are identified by segmenting customers in terms of the pyramid of customer lifetime value. Finally, suggestions are proposed to improve customer relationship management system.
Machine summary:
The purpose of this study is clustering customers online of a mobile sales website based on their lifetime value and RFM model.
At the proposed framework in this study after determining the values of RFM model include recently, frequency and monetary of purchase and weighting them using Shannon entropy, a self-organizing map is applied to the segmentation of customers.
This study combines a Multi -Criteria -Decision -Method and a data mining framework for clustering mobile internet customers' behavior prediction are provided for the field to identify key customers, choose the right marketing strategies and optimal allocation of resources according to the characteristics of each customer segment in order to improve the performance of customer relationship management system that can be provided.
The special feature of this study compared with previous studies can be combined by customer segmentation based on customer lifetime value has pointed as pyramid, also in this study, to avoid weighting the experts' opinions in parameters model, the method of Shannon's entropy has been used and according to the intelligence and speed of neural networks in data analysis, the self-organizing neural network is used for clustering.
Literature Review Self-organizing networks, many applications in different fields such as science, engineering (Jounela, Vermasvuori, Enden, & Haavisto, 2003), medicine(Moshou, Hostens, Papaioannou, & Ramon, 2005), etc.
Namvar, Gholamian, and KhakAbi, (2010) in their study, RFM analysis of the data and to calculate the lifetime value of customers were using K-Means algorithm.