Predicting the class of a customer profile is a key task in marketing, which enables businesses to approach the right customer with the right product at the right time through the right channel to satisfy the customer's evolving needs. However, due to costs, privacy and/or data protection, only the business' owned transactional data is typically available for constructing customer profiles. Predicting the class of customer profiles based on such data is challenging, as the data tends to be very large, heavily sparse and highly skewed. We present a new approach that is designed to efficiently and accurately handle the multi-class classification of customer profiles built using sparse and skewed transactional data. Our approach first bins the customer profiles on the basis of the number of items transacted. The discovered bins are then partitioned and prototypes within each of the discovered bins selected to build the multi-class classifier models. The results obtained from using four multi-class classifiers on real-world transactional data from the food sales domain consistently show the critical numbers of items at which the predictive performance of customer profiles can be substantially improved.
展开▼
