An efficient Algorithm for Frequent Pattern Mining over Uncertain Data Stream
Mingye Xie, Long Tan · 2019
In recent years, due to the increasing demand for real-time data processing, data mining on stream gradually becomes a research hotspots, mining over uncertain data stream is a more practical compare to over precise data stream, which doesn't take uncertainty of data in real world into consideration. Most of the popular algorithms of uncertain data stream mining are based on tree structure, all branches in the tree need be retrieved to find all frequent patterns while is time-consuming. Considering this defect of tree structure, a new structure Uncertain Item-lists (UIT-lists) and a efficient Uncertain Frequent Stream mining algorithm (UFS-mine) is presented. Experiments show our approach has great performance on runtime and with same memory usage compare to an up-to-state uncertain stream data mining algorithm.