Date Added: Apr 2013
Periodicity mining is used for predicting different applications such as prediction, forecasting, etc. It has several application in Time-series databases. Several algorithms are present for detecting the periodicity. But most of the algorithm do not take into account the presence of noise or partial periodicity. Here the authors compare four different types of algorithm. Based on time-wrapping, the first algorithm wraps the time axis to optimally remove the noise at various locations. The second algorithm can be viewed as a variation of the approximate string matching algorithm. The third algorithm is used for partial periodicity detection and in the fourth one periodic detection using suffix tree is done.