tailieunhanh - GROUPING CUSTOMERS USING FREQUENTLY PURCHASED GOODS SETS

In the international geographic information community one of the contemporary challenges is to develop and deploy distributed technology solutions that provide means for accessing, exploring, and utilizing distributed geographic information and geoprocessing resources over a communication network like the Internet; anywhere, anytime, and with any device. Today, the scientific research and industry development put special emphasis on interoperable geographic information services; henceforth referred to as GI services. The proceeding diffusion of GI services – most notably of geographic model / information management services and geoprocessing services as defined in the GI services taxonomy of ISO 19119 (ISO/TC 211 2002). | GROUPING CUSTOMERS USING FREQUENTLY PURCHASED GOODS SETS Y anbo J. Wang Department of Computer Science The University of Liverpool Chadwick Building Peach Street Liverpool L69 7ZF UK jwang @ ABSTRACT Analysing customers in groups is one of the most fundamental issues in Marketing. It helps companies sufficiently learn from their customers and rationally design their marketing strategies. Given a customer database with n records customers and m attributes one s characteristics stored different approaches can be applied to automatically segment cluster records in divisions. In this paper we propose an Association Rule Mining ARM based approach which introduces transaction market-basket data into current customer segmentation task and automatically recognises customers in groups possibly overlapping using their frequently purchased goods sets noticed as frequent itemsets in ARM . Furthermore a variety of strategies and or techniques . statistics based data mining based machine learning based etc. that are found in current customer segmentation can be applied to deeply extract the similarities in customer characteristics within each pre-generated group which indicates why these particular goods are purchased simultaneously. KEY WORDS Association rule mining clustering frequent itemset marketing and transaction data 1. INTRODUCTION Analysing customers in groups is a fundamental issue in Marketing. By sufficiently learning from customers companies precisely segment their markets customers and logically determine their own target market that ensures a rational design of marketing strategy 3 . Moreover grouping customers motivates the studies of customer relationship management which helps companies improve the profitability of their interactions with customers while at the same time making the interactions appear friendlier through individualisation 4 . Given a customer database with n records customers and m attributes one s characteristics stored .