![]() | Only 14 pages are availabe for public view |
Abstract The customers’ information contained in databases have increased dramatically in the last few years, thus organizations built their customer information data warehouse aiming to enhance the process of customer services. Data mining is used to deal with such volume of data to enhance the process of customer services. One of the most important and powerful techniques of data mining is decision trees algorithm. But it faces common problems same as the other data mining techniques which are first, what is the most important attribute to set as start node to begin with. Second, it is more suitable for large and sophisticated business area so it’s complicated and high cost. Third, it’s not easily used even by non¬specialists in the field. This work aims to produce two proposed methodologies to overcome these problems. First one is K-MIAS methodology to select the K-Most important attributes that distinguish different customer types. K-MIAS Methodology consists of three phases. The first phase is data preparation which prepares data for computing calculations. The second phase is K-MIAS algorithm which ranks the quantification levels for each attributes with respect to all attributes to select the k-most important attributes. While the third phase is to visualize data which helps for better data understanding and clarifying the results. Second proposed methodology is SDT which is simple, powerful and low-cost proposed methodology to simulate the decision trees algorithm for different business scopes and natures. SDT methodology consists of three phases. The first phase is data preparation which prepare data for computing calculations, the second phase is SDT algorithm which represents a simulation of decision trees algorithm to find the most important rules that distinguish specific type of customers while the third phase is to visualize results and rules for better understanding and clarifying the results. The two methodologies are efficient, successful and they are easy to use even by non-specialists in the field. In this work, K-MIAS and SDT methodologies are tested by a dataset consist of 1000 instants for trainee’s questionnaire. K-MIAS methodology selects the K-Most important attribute and DTS methodology selects the most important rules and paths that reaches the selected ratio and tested cluster of customers. The two proposed algorithms get target successfully with interesting remarks and finding. |