Efficient data mining for data warehousing and e-commerce

This research focuses on association rule mining towards integration with data warehousing, minimizing the required number of database passes, and adding a new form of association mining that is very useful to e-commerce. We present several fundamental solutions to high profile data mining problems. The first solution marries data warehousing and association mining to handle market basket data. We continue with the data warehousing solution to present a new counting structure that has applications throughout all algorithms handling itemset counting. Next, we propose methods for the incremental mining problem when working with dynamic databases. The solutions we provide add new algorithms and employ modified versions of the data structures we present earlier in this dissertation. Finally, we propose a new form of association data mining that has implications to e-commerce.