Abstract
Market-basket analysis is a well-known business problem, which can be (partially) solved computationally using association rules, mined from transaction data to maximize cross-selling effects. Here, we model the market-basket analysis as a finite mixture density of human consumption behavior according to social and cultural events. This leads to the use of principle component analysis and possibly mixture density analysis of transaction data, which was not apparent before. We compare PCA and association rules mined from a set of benchmark transaction data, to explore similarities and differences between these two data exploration tools.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have