资 源 简 介
we propose an efficient algorithm named UDS-FIM and a tree structure named UDS-Tree. Firstly, UDS-FIM compresses transaction itemsets to a UDS-Tree as compact as the original FP-Tree; secondly, it mines frequent itemsets from the UDS-Tree without additional scan of transactions. We evaluate our algorithm using sparse and dense datasets including real-world datasets and synthetic datasets; the final experimental results show that UDS-FIM has achieved a good performance under different experimental conditions, especially on real-world datasets.
