A New Marketing Channel Management Strategy Based on Frequent Subtree Mining

For most manufacturers, success or failure is determined by how effectively and efficiently their products are sold through their marketing channel members, so the management of marketing channels plays an important role in market competition. Most existing work studies the problem of marketing channel management in a qualitative way. Recently, with the increase of amount of sales data, how to enhance the marketing channel quantitatively is significant. As the marketing channel can be viewed as a tree, in this paper, a new marketing channel management strategy based on frequent subtree mining is proposed. The proposed method is illustrated under the real-world sales data in ERDOS group. Firstly, the tree transaction is formed monthly. For each monthly transaction, only those channel members that pass the basic sales plan will be included. Secondly, we use the TreeMiner algorithm to discover embedded frequent subtrees. Finally, different management strategies are used for different kinds of discovered patterns. We show that our method can correspond to the seven decision areas in traditional marketing channel management. INTRODUCTION For most manufacturers, success or failure is determined by how effectively and efficiently their products are sold through their marketing channel members (e.g., agents, wholesalers, distributors, and retailers). Given this situation, considerable marketing channel research has focused on organizational responsibility for managing channel how interrelationships among a firm and its channel members can be managed better (Achrol and Stern 1988; Anderson et al 1997). Recently, our capabilities of both generating and collecting data have been increasing rapidly. The widespread use of bar codes for most commercial products, and the advances in data collection tools have provided us with huge amounts of data. This explosive growth in data and databases has generated an urgent need for new techniques and tools that can intelligently and automatically transform the processed data into useful information and knowledge. In order to relieve such a data rich but information poor plight, during the late 1980s, a new discipline named data mining emerged (Han and Kamber 2006), which devotes itself to extracting knowledge from huge volumes of data, with the help of the ubiquitous modern computing device, i.e., the computer. Most existing work studies the problem of marketing channel management in a qualitative way (Coughlan et al 2005; Pelton et al 2001). Recently, with the increase of amount of sales data, how to enhance the marketing channel quantitatively is significant. As the marketing channel can be viewed as a tree, in this paper, a new marketing channel management strategy based on frequent subtree mining is proposed. The proposed method is illustrated under the real-world sales data in ERDOS group. Firstly, the tree transaction is formed monthly. For each monthly transaction, only those channel members that pass the basic sales plan will be included. Secondly, we use the TreeMiner algorithm to discover embedded frequent subtrees. Finally, different management strategies are used for different kinds of discovered patterns. To the best of our knowledge, it is the first time to exploit the problem of marketing channel management by using frequent subtree mining. So our work is explorative, many details should be enhanced step by step in practice. New Marketing Channel Management Strategy Based on Frequent Subtree Mining Wang & Gao Communications of the IIMA 5