一种新颖周期模式的挖掘

Wavelet Miner: Mining Rare Periodic Patterns

基于小波数据结构设计了一种用于挖掘新的周期模式(RPP)的算法.该周期模式不同于以支持度为度量标准挖掘的周期或半周期模式,它能有效地发现RPP中P=〈Am→Bn〉这样的模式.实验证明该算法是有效的,且具有很好的延展性.

A algorithm is designed for mining a kind of new periodic pattern （RPP） in our paper, which is based on wavelet data structre. This periodic pattern is different from existing periodic patterns or semi-periodic patterns that are measured using minimum support. It would effectively find the model such as P=〈A^m→B^n〉. The approach is experimentally evaluated to demonstrate that the proposed approach is efficient and promising.