Data mining for direct marketing: problems and solutions

TLDR

Direct marketing targets likely buyers, and data mining is increasingly used in this domain, but challenges such as extreme class imbalance, unsuitable predictive accuracy metrics, and large data volumes complicate model development. The paper aims to discuss methods for coping with these data mining problems in direct marketing, based on the authors’ project experience. The authors review techniques for handling class imbalance, alternative evaluation criteria, and scalable algorithms to manage large datasets in direct‑marketing data mining.

Abstract

Direct marketing is a process of identifying likely buyers of certain products and promoting the products accordingly. It is increasingly used by banks, insurance companies, and the retail industry. Data mining can provide an effective tool for direct marketing. During data mining, several specific problems arise. For example, the class distribution is extremely unbalanced (the response rate is about 1%), the predictive accuracy is no longer suitable for evaluating learning methods, and the number of examples can be too large. In this paper, we discuss methods of coping with these problems based on our experience on direct-marketing projects using data mining.