Customer Clustering Using Semi-Supervised Geographic Information

We present an innovative approach for clustering retail customers using semi-supervised geographic information. The approach aims at clustering (or segm enting) customers not only depending on their age, spending, etc., but also on their dwelling, which can discover userful customer patterns for the retailers marketing strategy. In real retail applications unsupervised clustering faces the problem of normalizing multiple hetergeneous features, which results in limited "ndings. Moreover, human knowledge cannot be incorporated in the process. Consequently, we propose a semi-superivsed approach which supports two kinds of human knowledge on the culstering 1) hard constraint - must-link and cannot-link and 2) soft constraint - distance comparison. The constraints can be appropriately applied in our tick of customer clustering. asad on the constraints, we develop a framework integrating metric learning (by weighing features) and clustering. The experimental results on real customer pro" le, comparing with the unsuperivsed approach, show reasonable clusters. In addition, using the proposed approach, the learned feature weights reveal valuable knowledge on the customers.