Mining Analysis Of Mobile Phone Sales Customers Based On Partition And Hierarchical Clustering Method

In today’s world of information technology, mainly in the hope to analyze the customer’s digital, the customer-related data is stored in a system, managers and researchers to facilitate the analysis and processing later. Collection of these data mining can improve sales. Late variety of data about customers, especially the purchase amount and timing of payments, if they can take relevant mining techniques, we can get a lot of valuable conclusions. This is the era of big data phone sales industry.First, data collection, recording customer data in a plurality of mobile phone sales locations. And collected data to data preprocessing, the first supplement missing data and a plurality of variable dimension reduction into two variables: the customer’s current value and potential value of the customer. A small amount of data preprocessing is to modify the erroneous data, complement some of the incomplete data and data dimensionality reduction process. In the pre-processing stage, using the AHP method to quantify the use YAAHP software.Secondly, the use of hierarchical cluster analysis and classification methods for customers to buy sales data generated by mobile phones to analyze, to achieve customer classification. Specific method is to use the method of dividing the K-means method and the method of hierarchical clustering method in the median, all customers will be collected in accordance with the current value and potential value of two variables clustering,eventually divided into four categories: current value high potential high value, low current value of high potential value, the value of low potential value of high current and low current potential value of low value.Finally, the K-means clustering and hierarchical clustering methods to analyze,summarize their respective advantages and disadvantages, and to summarize the present research, after the customers into four categories, salespeople can target different customer groups propose different marketing techniques.