This is a limited free preview of this book. Please buy full access.
Profiling and Signature Discovery
In market segmentation research, demographic and socioeconomic variables are often used for profiling purposes. They are used in segmentation studies to profile segments in order to enhance identifiability and accessibility. In this way, segments can be targeted, as media profiles and market areas are often described with demographic variables. In mixture models, profiling of segments is typically performed by using the posterior segment membership probabilities that provide the probability that a particular subject belongs to each of the derived segments. The segments derived from the mixture models have been profiled by most researchers in the second step of the analysis. Several marketing researchers have proposed models that simultaneously profile the derived segments with descriptor variables. In our modeling framework, we incorporate signature discovery techniques. In signature discovery and profiling research within the data mining community, studies have focused primarily on extracting features (variables) and generating rules to represent signatures for an individual customer. Signatures are often used for personalization and fraud detection.
The second essay studies a specific problem. As far as we know, this is the first research to examine free shipping promotions on the Internet in a theoretical framework. Researchers have studied shipping fees before but not specifically issues related to free shipping.
Research Problems and Solutions
In the first essay, we study a new approach to segmenting customer transactions that is based on the idea that there may exist natural behavioral patterns in different groups of transactions. In such cases, appropriate “pattern-based clustering” approaches can constitute an intuitive method for grouping customer transactions. At the highest level, the idea is to cluster customer transactions such that patterns generated from each cluster, while similar to each other within the cluster, are very different from the