期刊名称：International Journal of Computers and Communications
印刷版ISSN：2074-1294
出版年度：2012
卷号：6
期号：1
页码：1-8
出版社：University Press
摘要：Just like any other aspect of human life on our planet, the advancements in internet technology over the past decades have greatly influenced the way we live, changing the socio-political structure of the nations as it can clearly be seen in the Middle East and northern African countries. In specific terms, they have influenced our behavior in the way that we conduct marketing extending it beyond traditional boundaries both from business actions and consumer’s behavior. With ease of access and the vast amount of information that can be observed on the internet, consumers are lost for choice. These have increased the challenges that both consumers and business people face manifold. The survival conditions for both sides require more information in order to build new models that can respond to these challenges and help the decision making process in line with these changes and making them more efficient. This paper attempts to assess the fuzzy actions of buying behavior from a multinational viewpoint using data that have been collected on consumer risk assessment while attempting to buy a product on the internet. The survey covered a sample of 270male and female participants of different nationality with different levels of income and education. In a hierarchical modeling attempt, Logistic regression is used to identify factors that are significant in the action of purchasing leading to a sensitivity of 0.94 and specificity of 0.81. The resulting significant components from the first stage model are used as inputs in the application of AnswerTreeand more specifically, Classification Regression Tree (CRT) model to formulate what-if decision scenarios to help decision makers improve their targeting and selling options. The sensitivity and specificity of the latter approach were 0.93 and 0.83 respectively and led to the establishment of a set of decision rules each with their probability of success ranging from 0.67 to 1. None of the factors of gender, geographic location or income were significant in the process.
关键词：Tele-marketing; Customer behaviour; Logistic Regression; Decision Tree
