| Chapter 1: | Introduction |
This is a limited free preview of this book. Please buy full access.
Table 1. Relationship of the two essays
| Methods/Models | ||
| Problems | Data Mining | Marketing |
| Customer Segmentation | New method | Evaluation |
| Free-shipping Promotion | Evaluation | New model |
As shown in Table 1, the first essay addresses customer segmentation problems. In this essay we develop new data mining methods and apply them to customer segmentation problems. Other two popular methods from the marketing literature are used in the evaluation step. In this essay, we want to show how data mining methods can help solve customer segmentation problems in a data rich environment.
The second essay focuses on a specific marketing problem: the relationship between free shipping promotions and Internet shopping behavior. It develops original analytical models that generate hypotheses that are tested on the Internet data. A data mining method called contrast set is used for empirical testing.
Each essay is independently evaluated on different data sets. These data sets all originate from the Web. While data generated online has been extensively studied in the data mining community and used in marketing practice, the marketing research community has not developed a significant body of research based on such data, due partially to the overwhelming amount of data, noise in the data and effort needed to preprocess the data
In the first essay, we study how data mining concepts such as patterns can be used to help represent the underlying behavior governing the generation of the data and how the flexibility in representation can help us develop more effective methods in discovering segments in the data and build more accurate predictive models. We study a new approach to segmenting customer transactions that is based on the idea that natural behavioral patterns may exist in different groups of transactions. At the highest level, the idea is to cluster customer transactions such