Application Research Of CRM Data Mining Based On Clustering Algorithms

With the rapid development of information technology, especially the wide use of database and network, data increase abundantly, in which lies vitally important things to the corporation. If we can analyze the considerable data and information effectively, the rules and models and knowledge will be found, which we can use to aid company to make its strategy and marketing plan.Knowledge discovery is defined to abstract previously unknown and potentially useful models from a large amount of data, and data mining is an important component in the procedure of knowledge discovery. It has been becoming hotspots in many fields how to obtain not only the superficial information but also the underneath one from a large scale of data.Customer relationship management (CRM) is a concept of management. A company can analyze and trace into information of each customer, with the result of their demands, companies can make and modify their business process and get profit together with their customers.Clustering is a very important branch of data mining field and has been being used in business, geology exploration and pictures analyzing, etc, because it is not only used as per-treatment method, but also used independently. It doesn't need notation of class and can be used on the dataset directly.The thesis uses K-medoid algorithm on customer value segmentation and improve the quality of it with AGNES Cluster algorithm, decreasing the subjectivity and randomness of the method and increasing the accuracy and creditability, finding different customers with different features, which is very helpful for making market strategy and increasing customer's satisfactory and faithfulness. It is effective for companies to maintain old customers and attract new ones .The result of the experiment also testify thetruth of Pareto rule on retail. It is very meaningful and instructive not only on theory but also in reality.