Mining Weighted Frequent Patterns in Incremental Databases

Abstract

By considering different weights of the items, weighted frequent pattern (WFP) mining becomes an important research issue in data mining and knowledge discovery. However, existing algorithms cannot be applied for incremental and interactive WFP mining because they are based on a static database and require multiple database scans. In this paper, we present a novel tree structure ${\rm IWFPT}_{\textrm{\scriptsize{WA}}}$ (Incremental WFP tree based on weight ascending order) and an algorithm ${\rm IWFP}_{\textrm{\scriptsize{WA}}}$ for incremental and interactive WFP mining using a single database scan. Extensive performance analyses show that our tree structure and algorithm are efficient for incremental and interactive WFP mining.