In recent years, the growing competition has motivated companies to find out useful data mining methods to convert massive data into valuable information. How to retrieve meaningful messages from data effectively and efficiently is always the key. Market segmentation is the key to better marketing and customer relationship management. In this research, we propose an effective method to retrieve consumer shopping behavior patterns from the real transaction data. We first transform the transaction data into an easy to represent format that can be used to efficiently represent dynamic customer behaviors; subsequently powerful hierarchical and non-hierarchical clustering algorithms are applied to segregate the overall customer population. These clusters consist of shopping behaviors which are significantly different from each other. These results can be used to better target the customers and develop better promotional strategies.