Customer targeting: A neural network approach guided by genetic algorithms

• 成果类型: Article
• 署名作者: Kim, Y; Street, WN; Russell, GJ; Menczer, F
• 署名单位: Utah System of Higher Education; Utah State University; University of Iowa; University of Iowa; Indiana University System; Indiana University Bloomington
• 刊物名称: MANAGEMENT SCIENCE
• ISSN/ISSBN: 0025-1909
• DOI: 10.1287/mnsc.1040.0296
• 发表日期: 2005
• 页码: 264-276
• 关键词: database marketing; Neural Networks; GENETIC ALGORITHMS; customer relationship management

摘要

One of the key problems in database marketing is the identification and profiling of households that are most likely to be interested in a particular product or service. Principal component analysis (PCA) of customer background information followed by logistic regression analysis of response behavior is commonly used by database marketers. In this paper, we propose a new approach that uses artificial neural networks (ANNs) guided by genetic algorithms (GAs) to target households. We show that the resulting selection rule is more accurate and more parsimonious than the PCA/logit rule when the manager has a clear decision criterion. Under vague decision criteria, the new procedure loses its advantage in interpretability, but is still more accurate than PCA/logit in targeting households.